xref: /macosx-10.10/CPANInternal-159.1/Perl-Tidy-20121207/lib/Perl/Tidy.pm

#
############################################################
#
#    perltidy - a perl script indenter and formatter
#
#    Copyright (c) 2000-2012 by Steve Hancock
#    Distributed under the GPL license agreement; see file COPYING
#
#    This program is free software; you can redistribute it and/or modify
#    it under the terms of the GNU General Public License as published by
#    the Free Software Foundation; either version 2 of the License, or
#    (at your option) any later version.
#
#    This program is distributed in the hope that it will be useful,
#    but WITHOUT ANY WARRANTY; without even the implied warranty of
#    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#    GNU General Public License for more details.
#
#    You should have received a copy of the GNU General Public License along
#    with this program; if not, write to the Free Software Foundation, Inc.,
#    51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
#
#    For brief instructions instructions, try 'perltidy -h'.
#    For more complete documentation, try 'man perltidy'
#    or visit http://perltidy.sourceforge.net
#
#    This script is an example of the default style.  It was formatted with:
#
#      perltidy Tidy.pm
#
#    Code Contributions: See ChangeLog.html for a complete history.
#      Michael Cartmell supplied code for adaptation to VMS and helped with
#        v-strings.
#      Hugh S. Myers supplied sub streamhandle and the supporting code to
#        create a Perl::Tidy module which can operate on strings, arrays, etc.
#      Yves Orton supplied coding to help detect Windows versions.
#      Axel Rose supplied a patch for MacPerl.
#      Sebastien Aperghis-Tramoni supplied a patch for the defined or operator.
#      Dan Tyrell contributed a patch for binary I/O.
#      Ueli Hugenschmidt contributed a patch for -fpsc
#      Sam Kington supplied a patch to identify the initial indentation of
#      entabbed code.
#      jonathan swartz supplied patches for:
#      * .../ pattern, which looks upwards from directory
#      * --notidy, to be used in directories where we want to avoid
#        accidentally tidying
#      * prefilter and postfilter
#      * iterations option
#
#      Many others have supplied key ideas, suggestions, and bug reports;
#        see the CHANGES file.
#
############################################################

package Perl::Tidy;
use 5.004;    # need IO::File from 5.004 or later
BEGIN { $^W = 1; }    # turn on warnings

use strict;
use Exporter;
use Carp;
$|++;

use vars qw{
  $VERSION
  @ISA
  @EXPORT
  $missing_file_spec
  $fh_stderr
};

@ISA    = qw( Exporter );
@EXPORT = qw( &perltidy );

use Cwd;
use IO::File;
use File::Basename;
use File::Copy;

BEGIN {
    ( $VERSION = q($Id: Tidy.pm,v 1.74 2012/12/07 13:56:49 perltidy Exp $) ) =~ s/^.*\s+(\d+)\/(\d+)\/(\d+).*$/$1$2$3/; # all one line for MakeMaker
}

sub streamhandle {

    # given filename and mode (r or w), create an object which:
    #   has a 'getline' method if mode='r', and
    #   has a 'print' method if mode='w'.
    # The objects also need a 'close' method.
    #
    # How the object is made:
    #
    # if $filename is:     Make object using:
    # ----------------     -----------------
    # '-'                  (STDIN if mode = 'r', STDOUT if mode='w')
    # string               IO::File
    # ARRAY  ref           Perl::Tidy::IOScalarArray (formerly IO::ScalarArray)
    # STRING ref           Perl::Tidy::IOScalar      (formerly IO::Scalar)
    # object               object
    #                      (check for 'print' method for 'w' mode)
    #                      (check for 'getline' method for 'r' mode)
    my $ref = ref( my $filename = shift );
    my $mode = shift;
    my $New;
    my $fh;

    # handle a reference
    if ($ref) {
        if ( $ref eq 'ARRAY' ) {
            $New = sub { Perl::Tidy::IOScalarArray->new(@_) };
        }
        elsif ( $ref eq 'SCALAR' ) {
            $New = sub { Perl::Tidy::IOScalar->new(@_) };
        }
        else {

            # Accept an object with a getline method for reading. Note:
            # IO::File is built-in and does not respond to the defined
            # operator.  If this causes trouble, the check can be
            # skipped and we can just let it crash if there is no
            # getline.
            if ( $mode =~ /[rR]/ ) {
                if ( $ref eq 'IO::File' || defined &{ $ref . "::getline" } ) {
                    $New = sub { $filename };
                }
                else {
                    $New = sub { undef };
                    confess <<EOM;
------------------------------------------------------------------------
No 'getline' method is defined for object of class $ref
Please check your call to Perl::Tidy::perltidy.  Trace follows.
------------------------------------------------------------------------
EOM
                }
            }

            # Accept an object with a print method for writing.
            # See note above about IO::File
            if ( $mode =~ /[wW]/ ) {
                if ( $ref eq 'IO::File' || defined &{ $ref . "::print" } ) {
                    $New = sub { $filename };
                }
                else {
                    $New = sub { undef };
                    confess <<EOM;
------------------------------------------------------------------------
No 'print' method is defined for object of class $ref
Please check your call to Perl::Tidy::perltidy. Trace follows.
------------------------------------------------------------------------
EOM
                }
            }
        }
    }

    # handle a string
    else {
        if ( $filename eq '-' ) {
            $New = sub { $mode eq 'w' ? *STDOUT : *STDIN }
        }
        else {
            $New = sub { IO::File->new(@_) };
        }
    }
    $fh = $New->( $filename, $mode )
      or Warn("Couldn't open file:$filename in mode:$mode : $!\n");
    return $fh, ( $ref or $filename );
}

sub find_input_line_ending {

    # Peek at a file and return first line ending character.
    # Quietly return undef in case of any trouble.
    my ($input_file) = @_;
    my $ending;

    # silently ignore input from object or stdin
    if ( ref($input_file) || $input_file eq '-' ) {
        return $ending;
    }
    open( INFILE, $input_file ) || return $ending;

    binmode INFILE;
    my $buf;
    read( INFILE, $buf, 1024 );
    close INFILE;
    if ( $buf && $buf =~ /([\012\015]+)/ ) {
        my $test = $1;

        # dos
        if ( $test =~ /^(\015\012)+$/ ) { $ending = "\015\012" }

        # mac
        elsif ( $test =~ /^\015+$/ ) { $ending = "\015" }

        # unix
        elsif ( $test =~ /^\012+$/ ) { $ending = "\012" }

        # unknown
        else { }
    }

    # no ending seen
    else { }

    return $ending;
}

sub catfile {

    # concatenate a path and file basename
    # returns undef in case of error

    BEGIN { eval "require File::Spec"; $missing_file_spec = $@; }

    # use File::Spec if we can
    unless ($missing_file_spec) {
        return File::Spec->catfile(@_);
    }

    # Perl 5.004 systems may not have File::Spec so we'll make
    # a simple try.  We assume File::Basename is available.
    # return undef if not successful.
    my $name      = pop @_;
    my $path      = join '/', @_;
    my $test_file = $path . $name;
    my ( $test_name, $test_path ) = fileparse($test_file);
    return $test_file if ( $test_name eq $name );
    return undef if ( $^O eq 'VMS' );

    # this should work at least for Windows and Unix:
    $test_file = $path . '/' . $name;
    ( $test_name, $test_path ) = fileparse($test_file);
    return $test_file if ( $test_name eq $name );
    return undef;
}

sub make_temporary_filename {

    # Make a temporary filename.
    # The POSIX tmpnam() function has been unreliable for non-unix systems
    # (at least for the win32 systems that I've tested), so use a pre-defined
    # name for them.  A disadvantage of this is that two perltidy
    # runs in the same working directory may conflict.  However, the chance of
    # that is small and manageable by the user, especially on systems for which
    # the POSIX tmpnam function doesn't work.
    my $name = "perltidy.TMP";
    if ( $^O =~ /win32|dos/i || $^O eq 'VMS' || $^O eq 'MacOs' ) {
        return $name;
    }
    eval "use POSIX qw(tmpnam)";
    if ($@) { return $name }
    use IO::File;

    # just make a couple of tries before giving up and using the default
    for ( 0 .. 3 ) {
        my $tmpname = tmpnam();
        my $fh = IO::File->new( $tmpname, O_RDWR | O_CREAT | O_EXCL );
        if ($fh) {
            $fh->close();
            return ($tmpname);
            last;
        }
    }
    return ($name);
}

# Here is a map of the flow of data from the input source to the output
# line sink:
#
# LineSource-->Tokenizer-->Formatter-->VerticalAligner-->FileWriter-->
#       input                         groups                 output
#       lines   tokens      lines       of          lines    lines
#                                      lines
#
# The names correspond to the package names responsible for the unit processes.
#
# The overall process is controlled by the "main" package.
#
# LineSource is the stream of input lines
#
# Tokenizer analyzes a line and breaks it into tokens, peeking ahead
# if necessary.  A token is any section of the input line which should be
# manipulated as a single entity during formatting.  For example, a single
# ',' character is a token, and so is an entire side comment.  It handles
# the complexities of Perl syntax, such as distinguishing between '<<' as
# a shift operator and as a here-document, or distinguishing between '/'
# as a divide symbol and as a pattern delimiter.
#
# Formatter inserts and deletes whitespace between tokens, and breaks
# sequences of tokens at appropriate points as output lines.  It bases its
# decisions on the default rules as modified by any command-line options.
#
# VerticalAligner collects groups of lines together and tries to line up
# certain tokens, such as '=>', '#', and '=' by adding whitespace.
#
# FileWriter simply writes lines to the output stream.
#
# The Logger package, not shown, records significant events and warning
# messages.  It writes a .LOG file, which may be saved with a
# '-log' or a '-g' flag.

sub perltidy {

    my %defaults = (
        argv                  => undef,
        destination           => undef,
        formatter             => undef,
        logfile               => undef,
        errorfile             => undef,
        perltidyrc            => undef,
        source                => undef,
        stderr                => undef,
        dump_options          => undef,
        dump_options_type     => undef,
        dump_getopt_flags     => undef,
        dump_options_category => undef,
        dump_options_range    => undef,
        dump_abbreviations    => undef,
        prefilter             => undef,
        postfilter            => undef,
    );

    # don't overwrite callers ARGV
    local @ARGV   = @ARGV;
    local *STDERR = *STDERR;

    my %input_hash = @_;

    if ( my @bad_keys = grep { !exists $defaults{$_} } keys %input_hash ) {
        local $" = ')(';
        my @good_keys = sort keys %defaults;
        @bad_keys = sort @bad_keys;
        confess <<EOM;
------------------------------------------------------------------------
Unknown perltidy parameter : (@bad_keys)
perltidy only understands : (@good_keys)
------------------------------------------------------------------------

EOM
    }

    my $get_hash_ref = sub {
        my ($key) = @_;
        my $hash_ref = $input_hash{$key};
        if ( defined($hash_ref) ) {
            unless ( ref($hash_ref) eq 'HASH' ) {
                my $what = ref($hash_ref);
                my $but_is =
                  $what ? "but is ref to $what" : "but is not a reference";
                croak <<EOM;
------------------------------------------------------------------------
error in call to perltidy:
-$key must be reference to HASH $but_is
------------------------------------------------------------------------
EOM
            }
        }
        return $hash_ref;
    };

    %input_hash = ( %defaults, %input_hash );
    my $argv               = $input_hash{'argv'};
    my $destination_stream = $input_hash{'destination'};
    my $errorfile_stream   = $input_hash{'errorfile'};
    my $logfile_stream     = $input_hash{'logfile'};
    my $perltidyrc_stream  = $input_hash{'perltidyrc'};
    my $source_stream      = $input_hash{'source'};
    my $stderr_stream      = $input_hash{'stderr'};
    my $user_formatter     = $input_hash{'formatter'};
    my $prefilter          = $input_hash{'prefilter'};
    my $postfilter         = $input_hash{'postfilter'};

    if ($stderr_stream) {
        ( $fh_stderr, my $stderr_file ) =
          Perl::Tidy::streamhandle( $stderr_stream, 'w' );
        if ( !$fh_stderr ) {
            croak <<EOM;
------------------------------------------------------------------------
Unable to redirect STDERR to $stderr_stream
Please check value of -stderr in call to perltidy
------------------------------------------------------------------------
EOM
        }
    }
    else {
        $fh_stderr = *STDERR;
    }

    sub Warn ($) { $fh_stderr->print( $_[0] ); }

    sub Exit ($) {
        if   ( $_[0] ) { goto ERROR_EXIT }
        else           { goto NORMAL_EXIT }
    }

    sub Die ($) { Warn $_[0]; Exit(1); }

    # extract various dump parameters
    my $dump_options_type     = $input_hash{'dump_options_type'};
    my $dump_options          = $get_hash_ref->('dump_options');
    my $dump_getopt_flags     = $get_hash_ref->('dump_getopt_flags');
    my $dump_options_category = $get_hash_ref->('dump_options_category');
    my $dump_abbreviations    = $get_hash_ref->('dump_abbreviations');
    my $dump_options_range    = $get_hash_ref->('dump_options_range');

    # validate dump_options_type
    if ( defined($dump_options) ) {
        unless ( defined($dump_options_type) ) {
            $dump_options_type = 'perltidyrc';
        }
        unless ( $dump_options_type =~ /^(perltidyrc|full)$/ ) {
            croak <<EOM;
------------------------------------------------------------------------
Please check value of -dump_options_type in call to perltidy;
saw: '$dump_options_type'
expecting: 'perltidyrc' or 'full'
------------------------------------------------------------------------
EOM

        }
    }
    else {
        $dump_options_type = "";
    }

    if ($user_formatter) {

        # if the user defines a formatter, there is no output stream,
        # but we need a null stream to keep coding simple
        $destination_stream = Perl::Tidy::DevNull->new();
    }

    # see if ARGV is overridden
    if ( defined($argv) ) {

        my $rargv = ref $argv;
        if ( $rargv eq 'SCALAR' ) { $argv = $$argv; $rargv = undef }

        # ref to ARRAY
        if ($rargv) {
            if ( $rargv eq 'ARRAY' ) {
                @ARGV = @$argv;
            }
            else {
                croak <<EOM;
------------------------------------------------------------------------
Please check value of -argv in call to perltidy;
it must be a string or ref to ARRAY but is: $rargv
------------------------------------------------------------------------
EOM
            }
        }

        # string
        else {
            my ( $rargv, $msg ) = parse_args($argv);
            if ($msg) {
                Die <<EOM;
Error parsing this string passed to to perltidy with 'argv':
$msg
EOM
            }
            @ARGV = @{$rargv};
        }
    }

    my $rpending_complaint;
    $$rpending_complaint = "";
    my $rpending_logfile_message;
    $$rpending_logfile_message = "";

    my ( $is_Windows, $Windows_type ) = look_for_Windows($rpending_complaint);

    # VMS file names are restricted to a 40.40 format, so we append _tdy
    # instead of .tdy, etc. (but see also sub check_vms_filename)
    my $dot;
    my $dot_pattern;
    if ( $^O eq 'VMS' ) {
        $dot         = '_';
        $dot_pattern = '_';
    }
    else {
        $dot         = '.';
        $dot_pattern = '\.';    # must escape for use in regex
    }

    #---------------------------------------------------------------
    # get command line options
    #---------------------------------------------------------------
    my (
        $rOpts,       $config_file,      $rraw_options,
        $saw_extrude, $saw_pbp,          $roption_string,
        $rexpansion,  $roption_category, $roption_range
      )
      = process_command_line(
        $perltidyrc_stream,  $is_Windows, $Windows_type,
        $rpending_complaint, $dump_options_type,
      );

    #---------------------------------------------------------------
    # Handle requests to dump information
    #---------------------------------------------------------------

    # return or exit immediately after all dumps
    my $quit_now = 0;

    # Getopt parameters and their flags
    if ( defined($dump_getopt_flags) ) {
        $quit_now = 1;
        foreach my $op ( @{$roption_string} ) {
            my $opt  = $op;
            my $flag = "";

            # Examples:
            #  some-option=s
            #  some-option=i
            #  some-option:i
            #  some-option!
            if ( $opt =~ /(.*)(!|=.*|:.*)$/ ) {
                $opt  = $1;
                $flag = $2;
            }
            $dump_getopt_flags->{$opt} = $flag;
        }
    }

    if ( defined($dump_options_category) ) {
        $quit_now = 1;
        %{$dump_options_category} = %{$roption_category};
    }

    if ( defined($dump_options_range) ) {
        $quit_now = 1;
        %{$dump_options_range} = %{$roption_range};
    }

    if ( defined($dump_abbreviations) ) {
        $quit_now = 1;
        %{$dump_abbreviations} = %{$rexpansion};
    }

    if ( defined($dump_options) ) {
        $quit_now = 1;
        %{$dump_options} = %{$rOpts};
    }

    Exit 0 if ($quit_now);

    # make printable string of options for this run as possible diagnostic
    my $readable_options = readable_options( $rOpts, $roption_string );

    # dump from command line
    if ( $rOpts->{'dump-options'} ) {
        print STDOUT $readable_options;
        Exit 0;
    }

    #---------------------------------------------------------------
    # check parameters and their interactions
    #---------------------------------------------------------------
    my $tabsize =
      check_options( $rOpts, $is_Windows, $Windows_type, $rpending_complaint );

    if ($user_formatter) {
        $rOpts->{'format'} = 'user';
    }

    # there must be one entry here for every possible format
    my %default_file_extension = (
        tidy => 'tdy',
        html => 'html',
        user => '',
    );

    # be sure we have a valid output format
    unless ( exists $default_file_extension{ $rOpts->{'format'} } ) {
        my $formats = join ' ',
          sort map { "'" . $_ . "'" } keys %default_file_extension;
        my $fmt = $rOpts->{'format'};
        Die "-format='$fmt' but must be one of: $formats\n";
    }

    my $output_extension = make_extension( $rOpts->{'output-file-extension'},
        $default_file_extension{ $rOpts->{'format'} }, $dot );

    # If the backup extension contains a / character then the backup should
    # be deleted when the -b option is used.   On older versions of
    # perltidy this will generate an error message due to an illegal
    # file name.
    #
    # A backup file will still be generated but will be deleted
    # at the end.  If -bext='/' then this extension will be
    # the default 'bak'.  Otherwise it will be whatever characters
    # remains after all '/' characters are removed.  For example:
    # -bext         extension     slashes
    #  '/'          bak           1
    #  '/delete'    delete        1
    #  'delete/'    delete        1
    #  '/dev/null'  devnull       2    (Currently not allowed)
    my $bext          = $rOpts->{'backup-file-extension'};
    my $delete_backup = ( $rOpts->{'backup-file-extension'} =~ s/\///g );

    # At present only one forward slash is allowed.  In the future multiple
    # slashes may be allowed to allow for other options
    if ( $delete_backup > 1 ) {
        Die "-bext=$bext contains more than one '/'\n";
    }

    my $backup_extension =
      make_extension( $rOpts->{'backup-file-extension'}, 'bak', $dot );

    my $html_toc_extension =
      make_extension( $rOpts->{'html-toc-extension'}, 'toc', $dot );

    my $html_src_extension =
      make_extension( $rOpts->{'html-src-extension'}, 'src', $dot );

    # check for -b option;
    # silently ignore unless beautify mode
    my $in_place_modify = $rOpts->{'backup-and-modify-in-place'}
      && $rOpts->{'format'} eq 'tidy';

    # turn off -b with warnings in case of conflicts with other options
    if ($in_place_modify) {
        if ( $rOpts->{'standard-output'} ) {
            my $msg = "Ignoring -b; you may not use -b and -st together";
            $msg .= " (-pbp contains -st; see manual)" if ($saw_pbp);
            Warn "$msg\n";
            $in_place_modify = 0;
        }
        if ($destination_stream) {
            Warn
"Ignoring -b; you may not specify a destination stream and -b together\n";
            $in_place_modify = 0;
        }
        if ( ref($source_stream) ) {
            Warn
"Ignoring -b; you may not specify a source array and -b together\n";
            $in_place_modify = 0;
        }
        if ( $rOpts->{'outfile'} ) {
            Warn "Ignoring -b; you may not use -b and -o together\n";
            $in_place_modify = 0;
        }
        if ( defined( $rOpts->{'output-path'} ) ) {
            Warn "Ignoring -b; you may not use -b and -opath together\n";
            $in_place_modify = 0;
        }
    }

    Perl::Tidy::Formatter::check_options($rOpts);
    if ( $rOpts->{'format'} eq 'html' ) {
        Perl::Tidy::HtmlWriter->check_options($rOpts);
    }

    # make the pattern of file extensions that we shouldn't touch
    my $forbidden_file_extensions = "(($dot_pattern)(LOG|DEBUG|ERR|TEE)";
    if ($output_extension) {
        my $ext = quotemeta($output_extension);
        $forbidden_file_extensions .= "|$ext";
    }
    if ( $in_place_modify && $backup_extension ) {
        my $ext = quotemeta($backup_extension);
        $forbidden_file_extensions .= "|$ext";
    }
    $forbidden_file_extensions .= ')$';

    # Create a diagnostics object if requested;
    # This is only useful for code development
    my $diagnostics_object = undef;
    if ( $rOpts->{'DIAGNOSTICS'} ) {
        $diagnostics_object = Perl::Tidy::Diagnostics->new();
    }

    # no filenames should be given if input is from an array
    if ($source_stream) {
        if ( @ARGV > 0 ) {
            Die
"You may not specify any filenames when a source array is given\n";
        }

        # we'll stuff the source array into ARGV
        unshift( @ARGV, $source_stream );

        # No special treatment for source stream which is a filename.
        # This will enable checks for binary files and other bad stuff.
        $source_stream = undef unless ref($source_stream);
    }

    # use stdin by default if no source array and no args
    else {
        unshift( @ARGV, '-' ) unless @ARGV;
    }

    #---------------------------------------------------------------
    # Ready to go...
    # main loop to process all files in argument list
    #---------------------------------------------------------------
    my $number_of_files = @ARGV;
    my $formatter       = undef;
    my $tokenizer       = undef;
    while ( my $input_file = shift @ARGV ) {
        my $fileroot;
        my $input_file_permissions;

        #---------------------------------------------------------------
        # prepare this input stream
        #---------------------------------------------------------------
        if ($source_stream) {
            $fileroot = "perltidy";
        }
        elsif ( $input_file eq '-' ) {    # '-' indicates input from STDIN
            $fileroot = "perltidy";       # root name to use for .ERR, .LOG, etc
            $in_place_modify = 0;
        }
        else {
            $fileroot = $input_file;
            unless ( -e $input_file ) {

                # file doesn't exist - check for a file glob
                if ( $input_file =~ /([\?\*\[\{])/ ) {

                    # Windows shell may not remove quotes, so do it
                    my $input_file = $input_file;
                    if ( $input_file =~ /^\'(.+)\'$/ ) { $input_file = $1 }
                    if ( $input_file =~ /^\"(.+)\"$/ ) { $input_file = $1 }
                    my $pattern = fileglob_to_re($input_file);
                    ##eval "/$pattern/";
                    if ( !$@ && opendir( DIR, './' ) ) {
                        my @files =
                          grep { /$pattern/ && !-d $_ } readdir(DIR);
                        closedir(DIR);
                        if (@files) {
                            unshift @ARGV, @files;
                            next;
                        }
                    }
                }
                Warn "skipping file: '$input_file': no matches found\n";
                next;
            }

            unless ( -f $input_file ) {
                Warn "skipping file: $input_file: not a regular file\n";
                next;
            }

            # As a safety precaution, skip zero length files.
            # If for example a source file got clobberred somehow,
            # the old .tdy or .bak files might still exist so we
            # shouldn't overwrite them with zero length files.
            unless ( -s $input_file ) {
                Warn "skipping file: $input_file: Zero size\n";
                next;
            }

            unless ( ( -T $input_file ) || $rOpts->{'force-read-binary'} ) {
                Warn
                  "skipping file: $input_file: Non-text (override with -f)\n";
                next;
            }

            # we should have a valid filename now
            $fileroot               = $input_file;
            $input_file_permissions = ( stat $input_file )[2] & 07777;

            if ( $^O eq 'VMS' ) {
                ( $fileroot, $dot ) = check_vms_filename($fileroot);
            }

            # add option to change path here
            if ( defined( $rOpts->{'output-path'} ) ) {

                my ( $base, $old_path ) = fileparse($fileroot);
                my $new_path = $rOpts->{'output-path'};
                unless ( -d $new_path ) {
                    unless ( mkdir $new_path, 0777 ) {
                        Die "unable to create directory $new_path: $!\n";
                    }
                }
                my $path = $new_path;
                $fileroot = catfile( $path, $base );
                unless ($fileroot) {
                    Die <<EOM;
------------------------------------------------------------------------
Problem combining $new_path and $base to make a filename; check -opath
------------------------------------------------------------------------
EOM
                }
            }
        }

        # Skip files with same extension as the output files because
        # this can lead to a messy situation with files like
        # script.tdy.tdy.tdy ... or worse problems ...  when you
        # rerun perltidy over and over with wildcard input.
        if (
            !$source_stream
            && (   $input_file =~ /$forbidden_file_extensions/o
                || $input_file eq 'DIAGNOSTICS' )
          )
        {
            Warn "skipping file: $input_file: wrong extension\n";
            next;
        }

        # the 'source_object' supplies a method to read the input file
        my $source_object =
          Perl::Tidy::LineSource->new( $input_file, $rOpts,
            $rpending_logfile_message );
        next unless ($source_object);

        # Prefilters and postfilters: The prefilter is a code reference
        # that will be applied to the source before tidying, and the
        # postfilter is a code reference to the result before outputting.
        if ($prefilter) {
            my $buf = '';
            while ( my $line = $source_object->get_line() ) {
                $buf .= $line;
            }
            $buf = $prefilter->($buf);

            $source_object = Perl::Tidy::LineSource->new( \$buf, $rOpts,
                $rpending_logfile_message );
        }

        # register this file name with the Diagnostics package
        $diagnostics_object->set_input_file($input_file)
          if $diagnostics_object;

        #---------------------------------------------------------------
        # prepare the output stream
        #---------------------------------------------------------------
        my $output_file = undef;
        my $actual_output_extension;

        if ( $rOpts->{'outfile'} ) {

            if ( $number_of_files <= 1 ) {

                if ( $rOpts->{'standard-output'} ) {
                    my $msg = "You may not use -o and -st together";
                    $msg .= " (-pbp contains -st; see manual)" if ($saw_pbp);
                    Die "$msg\n";
                }
                elsif ($destination_stream) {
                    Die
"You may not specify a destination array and -o together\n";
                }
                elsif ( defined( $rOpts->{'output-path'} ) ) {
                    Die "You may not specify -o and -opath together\n";
                }
                elsif ( defined( $rOpts->{'output-file-extension'} ) ) {
                    Die "You may not specify -o and -oext together\n";
                }
                $output_file = $rOpts->{outfile};

                # make sure user gives a file name after -o
                if ( $output_file =~ /^-/ ) {
                    Die "You must specify a valid filename after -o\n";
                }

                # do not overwrite input file with -o
                if ( defined($input_file_permissions)
                    && ( $output_file eq $input_file ) )
                {
                    Die "Use 'perltidy -b $input_file' to modify in-place\n";
                }
            }
            else {
                Die "You may not use -o with more than one input file\n";
            }
        }
        elsif ( $rOpts->{'standard-output'} ) {
            if ($destination_stream) {
                my $msg =
                  "You may not specify a destination array and -st together\n";
                $msg .= " (-pbp contains -st; see manual)" if ($saw_pbp);
                Die "$msg\n";
            }
            $output_file = '-';

            if ( $number_of_files <= 1 ) {
            }
            else {
                Die "You may not use -st with more than one input file\n";
            }
        }
        elsif ($destination_stream) {
            $output_file = $destination_stream;
        }
        elsif ($source_stream) {    # source but no destination goes to stdout
            $output_file = '-';
        }
        elsif ( $input_file eq '-' ) {
            $output_file = '-';
        }
        else {
            if ($in_place_modify) {
                $output_file = IO::File->new_tmpfile()
                  or Die "cannot open temp file for -b option: $!\n";
            }
            else {
                $actual_output_extension = $output_extension;
                $output_file             = $fileroot . $output_extension;
            }
        }

        # the 'sink_object' knows how to write the output file
        my $tee_file = $fileroot . $dot . "TEE";

        my $line_separator = $rOpts->{'output-line-ending'};
        if ( $rOpts->{'preserve-line-endings'} ) {
            $line_separator = find_input_line_ending($input_file);
        }

        # Eventually all I/O may be done with binmode, but for now it is
        # only done when a user requests a particular line separator
        # through the -ple or -ole flags
        my $binmode = 0;
        if   ( defined($line_separator) ) { $binmode        = 1 }
        else                              { $line_separator = "\n" }

        my ( $sink_object, $postfilter_buffer );
        if ($postfilter) {
            $sink_object =
              Perl::Tidy::LineSink->new( \$postfilter_buffer, $tee_file,
                $line_separator, $rOpts, $rpending_logfile_message, $binmode );
        }
        else {
            $sink_object =
              Perl::Tidy::LineSink->new( $output_file, $tee_file,
                $line_separator, $rOpts, $rpending_logfile_message, $binmode );
        }

        #---------------------------------------------------------------
        # initialize the error logger for this file
        #---------------------------------------------------------------
        my $warning_file = $fileroot . $dot . "ERR";
        if ($errorfile_stream) { $warning_file = $errorfile_stream }
        my $log_file = $fileroot . $dot . "LOG";
        if ($logfile_stream) { $log_file = $logfile_stream }

        my $logger_object =
          Perl::Tidy::Logger->new( $rOpts, $log_file, $warning_file,
            $fh_stderr, $saw_extrude );
        write_logfile_header(
            $rOpts,        $logger_object, $config_file,
            $rraw_options, $Windows_type,  $readable_options,
        );
        if ($$rpending_logfile_message) {
            $logger_object->write_logfile_entry($$rpending_logfile_message);
        }
        if ($$rpending_complaint) {
            $logger_object->complain($$rpending_complaint);
        }

        #---------------------------------------------------------------
        # initialize the debug object, if any
        #---------------------------------------------------------------
        my $debugger_object = undef;
        if ( $rOpts->{DEBUG} ) {
            $debugger_object =
              Perl::Tidy::Debugger->new( $fileroot . $dot . "DEBUG" );
        }

        #---------------------------------------------------------------
        # loop over iterations for one source stream
        #---------------------------------------------------------------

        # We will do a convergence test if 3 or more iterations are allowed.
        # It would be pointless for fewer because we have to make at least
        # two passes before we can see if we are converged, and the test
        # would just slow things down.
        my $max_iterations = $rOpts->{'iterations'};
        my $convergence_log_message;
        my %saw_md5;
        my $do_convergence_test = $max_iterations > 2;
        if ($do_convergence_test) {
            eval "use Digest::MD5 qw(md5_hex)";
            $do_convergence_test = !$@;
        }

        # save objects to allow redirecting output during iterations
        my $sink_object_final     = $sink_object;
        my $debugger_object_final = $debugger_object;
        my $logger_object_final   = $logger_object;

        for ( my $iter = 1 ; $iter <= $max_iterations ; $iter++ ) {

            # send output stream to temp buffers until last iteration
            my $sink_buffer;
            if ( $iter < $max_iterations ) {
                $sink_object =
                  Perl::Tidy::LineSink->new( \$sink_buffer, $tee_file,
                    $line_separator, $rOpts, $rpending_logfile_message,
                    $binmode );
            }
            else {
                $sink_object = $sink_object_final;
            }

            # Save logger, debugger output only on pass 1 because:
            # (1) line number references must be to the starting
            # source, not an intermediate result, and
            # (2) we need to know if there are errors so we can stop the
            # iterations early if necessary.
            if ( $iter > 1 ) {
                $debugger_object = undef;
                $logger_object   = undef;
            }

            #------------------------------------------------------------
            # create a formatter for this file : html writer or
            # pretty printer
            #------------------------------------------------------------

            # we have to delete any old formatter because, for safety,
            # the formatter will check to see that there is only one.
            $formatter = undef;

            if ($user_formatter) {
                $formatter = $user_formatter;
            }
            elsif ( $rOpts->{'format'} eq 'html' ) {
                $formatter =
                  Perl::Tidy::HtmlWriter->new( $fileroot, $output_file,
                    $actual_output_extension, $html_toc_extension,
                    $html_src_extension );
            }
            elsif ( $rOpts->{'format'} eq 'tidy' ) {
                $formatter = Perl::Tidy::Formatter->new(
                    logger_object      => $logger_object,
                    diagnostics_object => $diagnostics_object,
                    sink_object        => $sink_object,
                );
            }
            else {
                Die "I don't know how to do -format=$rOpts->{'format'}\n";
            }

            unless ($formatter) {
                Die "Unable to continue with $rOpts->{'format'} formatting\n";
            }

            #---------------------------------------------------------------
            # create the tokenizer for this file
            #---------------------------------------------------------------
            $tokenizer = undef;                     # must destroy old tokenizer
            $tokenizer = Perl::Tidy::Tokenizer->new(
                source_object      => $source_object,
                logger_object      => $logger_object,
                debugger_object    => $debugger_object,
                diagnostics_object => $diagnostics_object,
                tabsize            => $tabsize,

                starting_level      => $rOpts->{'starting-indentation-level'},
                indent_columns      => $rOpts->{'indent-columns'},
                look_for_hash_bang  => $rOpts->{'look-for-hash-bang'},
                look_for_autoloader => $rOpts->{'look-for-autoloader'},
                look_for_selfloader => $rOpts->{'look-for-selfloader'},
                trim_qw             => $rOpts->{'trim-qw'},

                continuation_indentation =>
                  $rOpts->{'continuation-indentation'},
                outdent_labels => $rOpts->{'outdent-labels'},
            );

            #---------------------------------------------------------------
            # now we can do it
            #---------------------------------------------------------------
            process_this_file( $tokenizer, $formatter );

            #---------------------------------------------------------------
            # close the input source and report errors
            #---------------------------------------------------------------
            $source_object->close_input_file();

            # line source for next iteration (if any) comes from the current
            # temporary output buffer
            if ( $iter < $max_iterations ) {

                $sink_object->close_output_file();
                $source_object =
                  Perl::Tidy::LineSource->new( \$sink_buffer, $rOpts,
                    $rpending_logfile_message );

                # stop iterations if errors or converged
                my $stop_now = $logger_object->{_warning_count};
                if ($stop_now) {
                    $convergence_log_message = <<EOM;
Stopping iterations because of errors.
EOM
                }
                elsif ($do_convergence_test) {
                    my $digest = md5_hex($sink_buffer);
                    if ( !$saw_md5{$digest} ) {
                        $saw_md5{$digest} = $iter;
                    }
                    else {

                        # Deja vu, stop iterating
                        $stop_now = 1;
                        my $iterm = $iter - 1;
                        if ( $saw_md5{$digest} != $iterm ) {

                            # Blinking (oscillating) between two stable
                            # end states.  This has happened in the past
                            # but at present there are no known instances.
                            $convergence_log_message = <<EOM;
Blinking. Output for iteration $iter same as for $saw_md5{$digest}.
EOM
                            $diagnostics_object->write_diagnostics(
                                $convergence_log_message)
                              if $diagnostics_object;
                        }
                        else {
                            $convergence_log_message = <<EOM;
Converged.  Output for iteration $iter same as for iter $iterm.
EOM
                            $diagnostics_object->write_diagnostics(
                                $convergence_log_message)
                              if $diagnostics_object && $iterm > 2;
                        }
                    }
                } ## end if ($do_convergence_test)

                if ($stop_now) {

                    # we are stopping the iterations early;
                    # copy the output stream to its final destination
                    $sink_object = $sink_object_final;
                    while ( my $line = $source_object->get_line() ) {
                        $sink_object->write_line($line);
                    }
                    $source_object->close_input_file();
                    last;
                }
            } ## end if ( $iter < $max_iterations)
        }    # end loop over iterations for one source file

        # restore objects which have been temporarily undefined
        # for second and higher iterations
        $debugger_object = $debugger_object_final;
        $logger_object   = $logger_object_final;

        $logger_object->write_logfile_entry($convergence_log_message)
          if $convergence_log_message;

        #---------------------------------------------------------------
        # Perform any postfilter operation
        #---------------------------------------------------------------
        if ($postfilter) {
            $sink_object->close_output_file();
            $sink_object =
              Perl::Tidy::LineSink->new( $output_file, $tee_file,
                $line_separator, $rOpts, $rpending_logfile_message, $binmode );
            my $buf = $postfilter->($postfilter_buffer);
            $source_object =
              Perl::Tidy::LineSource->new( \$buf, $rOpts,
                $rpending_logfile_message );
            while ( my $line = $source_object->get_line() ) {
                $sink_object->write_line($line);
            }
            $source_object->close_input_file();
        }

        # Save names of the input and output files for syntax check
        my $ifname = $input_file;
        my $ofname = $output_file;

        #---------------------------------------------------------------
        # handle the -b option (backup and modify in-place)
        #---------------------------------------------------------------
        if ($in_place_modify) {
            unless ( -f $input_file ) {

                # oh, oh, no real file to backup ..
                # shouldn't happen because of numerous preliminary checks
                Die
"problem with -b backing up input file '$input_file': not a file\n";
            }
            my $backup_name = $input_file . $backup_extension;
            if ( -f $backup_name ) {
                unlink($backup_name)
                  or Die
"unable to remove previous '$backup_name' for -b option; check permissions: $!\n";
            }

            # backup the input file
            # we use copy for symlinks, move for regular files
            if ( -l $input_file ) {
                File::Copy::copy( $input_file, $backup_name )
                  or Die "File::Copy failed trying to backup source: $!";
            }
            else {
                rename( $input_file, $backup_name )
                  or Die
"problem renaming $input_file to $backup_name for -b option: $!\n";
            }
            $ifname = $backup_name;

            # copy the output to the original input file
            # NOTE: it would be nice to just close $output_file and use
            # File::Copy::copy here, but in this case $output_file is the
            # handle of an open nameless temporary file so we would lose
            # everything if we closed it.
            seek( $output_file, 0, 0 )
              or Die "unable to rewind a temporary file for -b option: $!\n";
            my $fout = IO::File->new("> $input_file")
              or Die
"problem re-opening $input_file for write for -b option; check file and directory permissions: $!\n";
            binmode $fout;
            my $line;
            while ( $line = $output_file->getline() ) {
                $fout->print($line);
            }
            $fout->close();
            $output_file = $input_file;
            $ofname      = $input_file;
        }

        #---------------------------------------------------------------
        # clean up and report errors
        #---------------------------------------------------------------
        $sink_object->close_output_file()    if $sink_object;
        $debugger_object->close_debug_file() if $debugger_object;

        # set output file permissions
        if ( $output_file && -f $output_file && !-l $output_file ) {
            if ($input_file_permissions) {

                # give output script same permissions as input script, but
                # make it user-writable or else we can't run perltidy again.
                # Thus we retain whatever executable flags were set.
                if ( $rOpts->{'format'} eq 'tidy' ) {
                    chmod( $input_file_permissions | 0600, $output_file );
                }

                # else use default permissions for html and any other format
            }
        }

        #---------------------------------------------------------------
        # Do syntax check if requested and possible
        #---------------------------------------------------------------
        my $infile_syntax_ok = 0;    # -1 no  0=don't know   1 yes
        if (   $logger_object
            && $rOpts->{'check-syntax'}
            && $ifname
            && $ofname )
        {
            $infile_syntax_ok =
              check_syntax( $ifname, $ofname, $logger_object, $rOpts );
        }

        #---------------------------------------------------------------
        # remove the original file for in-place modify as follows:
        #   $delete_backup=0 never
        #   $delete_backup=1 only if no errors
        #   $delete_backup>1 always  : NOT ALLOWED, too risky, see above
        #---------------------------------------------------------------
        if (   $in_place_modify
            && $delete_backup
            && -f $ifname
            && ( $delete_backup > 1 || !$logger_object->{_warning_count} ) )
        {

            # As an added safety precaution, do not delete the source file
            # if its size has dropped from positive to zero, since this
            # could indicate a disaster of some kind, including a hardware
            # failure.  Actually, this could happen if you had a file of
            # all comments (or pod) and deleted everything with -dac (-dap)
            # for some reason.
            if ( !-s $output_file && -s $ifname && $delete_backup == 1 ) {
                Warn(
"output file '$output_file' missing or zero length; original '$ifname' not deleted\n"
                );
            }
            else {
                unlink($ifname)
                  or Die
"unable to remove previous '$ifname' for -b option; check permissions: $!\n";
            }
        }

        $logger_object->finish( $infile_syntax_ok, $formatter )
          if $logger_object;
    }    # end of main loop to process all files

  NORMAL_EXIT:
    return 0;

  ERROR_EXIT:
    return 1;
}    # end of main program perltidy

sub get_stream_as_named_file {

    # Return the name of a file containing a stream of data, creating
    # a temporary file if necessary.
    # Given:
    #  $stream - the name of a file or stream
    # Returns:
    #  $fname = name of file if possible, or undef
    #  $if_tmpfile = true if temp file, undef if not temp file
    #
    # This routine is needed for passing actual files to Perl for
    # a syntax check.
    my ($stream) = @_;
    my $is_tmpfile;
    my $fname;
    if ($stream) {
        if ( ref($stream) ) {
            my ( $fh_stream, $fh_name ) =
              Perl::Tidy::streamhandle( $stream, 'r' );
            if ($fh_stream) {
                my ( $fout, $tmpnam );

                # TODO: fix the tmpnam routine to return an open filehandle
                $tmpnam = Perl::Tidy::make_temporary_filename();
                $fout = IO::File->new( $tmpnam, 'w' );

                if ($fout) {
                    $fname      = $tmpnam;
                    $is_tmpfile = 1;
                    binmode $fout;
                    while ( my $line = $fh_stream->getline() ) {
                        $fout->print($line);
                    }
                    $fout->close();
                }
                $fh_stream->close();
            }
        }
        elsif ( $stream ne '-' && -f $stream ) {
            $fname = $stream;
        }
    }
    return ( $fname, $is_tmpfile );
}

sub fileglob_to_re {

    # modified (corrected) from version in find2perl
    my $x = shift;
    $x =~ s#([./^\$()])#\\$1#g;    # escape special characters
    $x =~ s#\*#.*#g;               # '*' -> '.*'
    $x =~ s#\?#.#g;                # '?' -> '.'
    "^$x\\z";                      # match whole word
}

sub make_extension {

    # Make a file extension, including any leading '.' if necessary
    # The '.' may actually be an '_' under VMS
    my ( $extension, $default, $dot ) = @_;

    # Use the default if none specified
    $extension = $default unless ($extension);

    # Only extensions with these leading characters get a '.'
    # This rule gives the user some freedom
    if ( $extension =~ /^[a-zA-Z0-9]/ ) {
        $extension = $dot . $extension;
    }
    return $extension;
}

sub write_logfile_header {
    my (
        $rOpts,        $logger_object, $config_file,
        $rraw_options, $Windows_type,  $readable_options
    ) = @_;
    $logger_object->write_logfile_entry(
"perltidy version $VERSION log file on a $^O system, OLD_PERL_VERSION=$]\n"
    );
    if ($Windows_type) {
        $logger_object->write_logfile_entry("Windows type is $Windows_type\n");
    }
    my $options_string = join( ' ', @$rraw_options );

    if ($config_file) {
        $logger_object->write_logfile_entry(
            "Found Configuration File >>> $config_file \n");
    }
    $logger_object->write_logfile_entry(
        "Configuration and command line parameters for this run:\n");
    $logger_object->write_logfile_entry("$options_string\n");

    if ( $rOpts->{'DEBUG'} || $rOpts->{'show-options'} ) {
        $rOpts->{'logfile'} = 1;    # force logfile to be saved
        $logger_object->write_logfile_entry(
            "Final parameter set for this run\n");
        $logger_object->write_logfile_entry(
            "------------------------------------\n");

        $logger_object->write_logfile_entry($readable_options);

        $logger_object->write_logfile_entry(
            "------------------------------------\n");
    }
    $logger_object->write_logfile_entry(
        "To find error messages search for 'WARNING' with your editor\n");
}

sub generate_options {

    ######################################################################
    # Generate and return references to:
    #  @option_string - the list of options to be passed to Getopt::Long
    #  @defaults - the list of default options
    #  %expansion - a hash showing how all abbreviations are expanded
    #  %category - a hash giving the general category of each option
    #  %option_range - a hash giving the valid ranges of certain options

    # Note: a few options are not documented in the man page and usage
    # message. This is because these are experimental or debug options and
    # may or may not be retained in future versions.
    #
    # Here are the undocumented flags as far as I know.  Any of them
    # may disappear at any time.  They are mainly for fine-tuning
    # and debugging.
    #
    # fll --> fuzzy-line-length           # a trivial parameter which gets
    #                                       turned off for the extrude option
    #                                       which is mainly for debugging
    # scl --> short-concatenation-item-length   # helps break at '.'
    # recombine                           # for debugging line breaks
    # valign                              # for debugging vertical alignment
    # I   --> DIAGNOSTICS                 # for debugging
    ######################################################################

    # here is a summary of the Getopt codes:
    # <none> does not take an argument
    # =s takes a mandatory string
    # :s takes an optional string  (DO NOT USE - filenames will get eaten up)
    # =i takes a mandatory integer
    # :i takes an optional integer (NOT RECOMMENDED - can cause trouble)
    # ! does not take an argument and may be negated
    #  i.e., -foo and -nofoo are allowed
    # a double dash signals the end of the options list
    #
    #---------------------------------------------------------------
    # Define the option string passed to GetOptions.
    #---------------------------------------------------------------

    my @option_string   = ();
    my %expansion       = ();
    my %option_category = ();
    my %option_range    = ();
    my $rexpansion      = \%expansion;

    # names of categories in manual
    # leading integers will allow sorting
    my @category_name = (
        '0. I/O control',
        '1. Basic formatting options',
        '2. Code indentation control',
        '3. Whitespace control',
        '4. Comment controls',
        '5. Linebreak controls',
        '6. Controlling list formatting',
        '7. Retaining or ignoring existing line breaks',
        '8. Blank line control',
        '9. Other controls',
        '10. HTML options',
        '11. pod2html options',
        '12. Controlling HTML properties',
        '13. Debugging',
    );

    #  These options are parsed directly by perltidy:
    #    help h
    #    version v
    #  However, they are included in the option set so that they will
    #  be seen in the options dump.

    # These long option names have no abbreviations or are treated specially
    @option_string = qw(
      html!
      noprofile
      no-profile
      npro
      recombine!
      valign!
      notidy
    );

    my $category = 13;    # Debugging
    foreach (@option_string) {
        my $opt = $_;     # must avoid changing the actual flag
        $opt =~ s/!$//;
        $option_category{$opt} = $category_name[$category];
    }

    $category = 11;                                       # HTML
    $option_category{html} = $category_name[$category];

    # routine to install and check options
    my $add_option = sub {
        my ( $long_name, $short_name, $flag ) = @_;
        push @option_string, $long_name . $flag;
        $option_category{$long_name} = $category_name[$category];
        if ($short_name) {
            if ( $expansion{$short_name} ) {
                my $existing_name = $expansion{$short_name}[0];
                Die
"redefining abbreviation $short_name for $long_name; already used for $existing_name\n";
            }
            $expansion{$short_name} = [$long_name];
            if ( $flag eq '!' ) {
                my $nshort_name = 'n' . $short_name;
                my $nolong_name = 'no' . $long_name;
                if ( $expansion{$nshort_name} ) {
                    my $existing_name = $expansion{$nshort_name}[0];
                    Die
"attempting to redefine abbreviation $nshort_name for $nolong_name; already used for $existing_name\n";
                }
                $expansion{$nshort_name} = [$nolong_name];
            }
        }
    };

    # Install long option names which have a simple abbreviation.
    # Options with code '!' get standard negation ('no' for long names,
    # 'n' for abbreviations).  Categories follow the manual.

    ###########################
    $category = 0;    # I/O_Control
    ###########################
    $add_option->( 'backup-and-modify-in-place', 'b',     '!' );
    $add_option->( 'backup-file-extension',      'bext',  '=s' );
    $add_option->( 'force-read-binary',          'f',     '!' );
    $add_option->( 'format',                     'fmt',   '=s' );
    $add_option->( 'iterations',                 'it',    '=i' );
    $add_option->( 'logfile',                    'log',   '!' );
    $add_option->( 'logfile-gap',                'g',     ':i' );
    $add_option->( 'outfile',                    'o',     '=s' );
    $add_option->( 'output-file-extension',      'oext',  '=s' );
    $add_option->( 'output-path',                'opath', '=s' );
    $add_option->( 'profile',                    'pro',   '=s' );
    $add_option->( 'quiet',                      'q',     '!' );
    $add_option->( 'standard-error-output',      'se',    '!' );
    $add_option->( 'standard-output',            'st',    '!' );
    $add_option->( 'warning-output',             'w',     '!' );

    # options which are both toggle switches and values moved here
    # to hide from tidyview (which does not show category 0 flags):
    # -ole moved here from category 1
    # -sil moved here from category 2
    $add_option->( 'output-line-ending',         'ole', '=s' );
    $add_option->( 'starting-indentation-level', 'sil', '=i' );

    ########################################
    $category = 1;    # Basic formatting options
    ########################################
    $add_option->( 'check-syntax',                 'syn',  '!' );
    $add_option->( 'entab-leading-whitespace',     'et',   '=i' );
    $add_option->( 'indent-columns',               'i',    '=i' );
    $add_option->( 'maximum-line-length',          'l',    '=i' );
    $add_option->( 'variable-maximum-line-length', 'vmll', '!' );
    $add_option->( 'whitespace-cycle',             'wc',   '=i' );
    $add_option->( 'perl-syntax-check-flags',      'pscf', '=s' );
    $add_option->( 'preserve-line-endings',        'ple',  '!' );
    $add_option->( 'tabs',                         't',    '!' );
    $add_option->( 'default-tabsize',              'dt',   '=i' );

    ########################################
    $category = 2;    # Code indentation control
    ########################################
    $add_option->( 'continuation-indentation',           'ci',   '=i' );
    $add_option->( 'line-up-parentheses',                'lp',   '!' );
    $add_option->( 'outdent-keyword-list',               'okwl', '=s' );
    $add_option->( 'outdent-keywords',                   'okw',  '!' );
    $add_option->( 'outdent-labels',                     'ola',  '!' );
    $add_option->( 'outdent-long-quotes',                'olq',  '!' );
    $add_option->( 'indent-closing-brace',               'icb',  '!' );
    $add_option->( 'closing-token-indentation',          'cti',  '=i' );
    $add_option->( 'closing-paren-indentation',          'cpi',  '=i' );
    $add_option->( 'closing-brace-indentation',          'cbi',  '=i' );
    $add_option->( 'closing-square-bracket-indentation', 'csbi', '=i' );
    $add_option->( 'brace-left-and-indent',              'bli',  '!' );
    $add_option->( 'brace-left-and-indent-list',         'blil', '=s' );

    ########################################
    $category = 3;    # Whitespace control
    ########################################
    $add_option->( 'add-semicolons',                            'asc',   '!' );
    $add_option->( 'add-whitespace',                            'aws',   '!' );
    $add_option->( 'block-brace-tightness',                     'bbt',   '=i' );
    $add_option->( 'brace-tightness',                           'bt',    '=i' );
    $add_option->( 'delete-old-whitespace',                     'dws',   '!' );
    $add_option->( 'delete-semicolons',                         'dsm',   '!' );
    $add_option->( 'nospace-after-keyword',                     'nsak',  '=s' );
    $add_option->( 'nowant-left-space',                         'nwls',  '=s' );
    $add_option->( 'nowant-right-space',                        'nwrs',  '=s' );
    $add_option->( 'paren-tightness',                           'pt',    '=i' );
    $add_option->( 'space-after-keyword',                       'sak',   '=s' );
    $add_option->( 'space-for-semicolon',                       'sfs',   '!' );
    $add_option->( 'space-function-paren',                      'sfp',   '!' );
    $add_option->( 'space-keyword-paren',                       'skp',   '!' );
    $add_option->( 'space-terminal-semicolon',                  'sts',   '!' );
    $add_option->( 'square-bracket-tightness',                  'sbt',   '=i' );
    $add_option->( 'square-bracket-vertical-tightness',         'sbvt',  '=i' );
    $add_option->( 'square-bracket-vertical-tightness-closing', 'sbvtc', '=i' );
    $add_option->( 'tight-secret-operators',                    'tso',   '!' );
    $add_option->( 'trim-qw',                                   'tqw',   '!' );
    $add_option->( 'want-left-space',                           'wls',   '=s' );
    $add_option->( 'want-right-space',                          'wrs',   '=s' );

    ########################################
    $category = 4;    # Comment controls
    ########################################
    $add_option->( 'closing-side-comment-else-flag',    'csce', '=i' );
    $add_option->( 'closing-side-comment-interval',     'csci', '=i' );
    $add_option->( 'closing-side-comment-list',         'cscl', '=s' );
    $add_option->( 'closing-side-comment-maximum-text', 'csct', '=i' );
    $add_option->( 'closing-side-comment-prefix',       'cscp', '=s' );
    $add_option->( 'closing-side-comment-warnings',     'cscw', '!' );
    $add_option->( 'closing-side-comments',             'csc',  '!' );
    $add_option->( 'closing-side-comments-balanced',    'cscb', '!' );
    $add_option->( 'format-skipping',                   'fs',   '!' );
    $add_option->( 'format-skipping-begin',             'fsb',  '=s' );
    $add_option->( 'format-skipping-end',               'fse',  '=s' );
    $add_option->( 'hanging-side-comments',             'hsc',  '!' );
    $add_option->( 'indent-block-comments',             'ibc',  '!' );
    $add_option->( 'indent-spaced-block-comments',      'isbc', '!' );
    $add_option->( 'fixed-position-side-comment',       'fpsc', '=i' );
    $add_option->( 'minimum-space-to-comment',          'msc',  '=i' );
    $add_option->( 'outdent-long-comments',             'olc',  '!' );
    $add_option->( 'outdent-static-block-comments',     'osbc', '!' );
    $add_option->( 'static-block-comment-prefix',       'sbcp', '=s' );
    $add_option->( 'static-block-comments',             'sbc',  '!' );
    $add_option->( 'static-side-comment-prefix',        'sscp', '=s' );
    $add_option->( 'static-side-comments',              'ssc',  '!' );
    $add_option->( 'ignore-side-comment-lengths',       'iscl', '!' );

    ########################################
    $category = 5;    # Linebreak controls
    ########################################
    $add_option->( 'add-newlines',                            'anl',   '!' );
    $add_option->( 'block-brace-vertical-tightness',          'bbvt',  '=i' );
    $add_option->( 'block-brace-vertical-tightness-list',     'bbvtl', '=s' );
    $add_option->( 'brace-vertical-tightness',                'bvt',   '=i' );
    $add_option->( 'brace-vertical-tightness-closing',        'bvtc',  '=i' );
    $add_option->( 'cuddled-else',                            'ce',    '!' );
    $add_option->( 'delete-old-newlines',                     'dnl',   '!' );
    $add_option->( 'opening-brace-always-on-right',           'bar',   '!' );
    $add_option->( 'opening-brace-on-new-line',               'bl',    '!' );
    $add_option->( 'opening-hash-brace-right',                'ohbr',  '!' );
    $add_option->( 'opening-paren-right',                     'opr',   '!' );
    $add_option->( 'opening-square-bracket-right',            'osbr',  '!' );
    $add_option->( 'opening-anonymous-sub-brace-on-new-line', 'asbl',  '!' );
    $add_option->( 'opening-sub-brace-on-new-line',           'sbl',   '!' );
    $add_option->( 'paren-vertical-tightness',                'pvt',   '=i' );
    $add_option->( 'paren-vertical-tightness-closing',        'pvtc',  '=i' );
    $add_option->( 'stack-closing-block-brace',               'scbb',  '!' );
    $add_option->( 'stack-closing-hash-brace',                'schb',  '!' );
    $add_option->( 'stack-closing-paren',                     'scp',   '!' );
    $add_option->( 'stack-closing-square-bracket',            'scsb',  '!' );
    $add_option->( 'stack-opening-block-brace',               'sobb',  '!' );
    $add_option->( 'stack-opening-hash-brace',                'sohb',  '!' );
    $add_option->( 'stack-opening-paren',                     'sop',   '!' );
    $add_option->( 'stack-opening-square-bracket',            'sosb',  '!' );
    $add_option->( 'vertical-tightness',                      'vt',    '=i' );
    $add_option->( 'vertical-tightness-closing',              'vtc',   '=i' );
    $add_option->( 'want-break-after',                        'wba',   '=s' );
    $add_option->( 'want-break-before',                       'wbb',   '=s' );
    $add_option->( 'break-after-all-operators',               'baao',  '!' );
    $add_option->( 'break-before-all-operators',              'bbao',  '!' );
    $add_option->( 'keep-interior-semicolons',                'kis',   '!' );

    ########################################
    $category = 6;    # Controlling list formatting
    ########################################
    $add_option->( 'break-at-old-comma-breakpoints', 'boc', '!' );
    $add_option->( 'comma-arrow-breakpoints',        'cab', '=i' );
    $add_option->( 'maximum-fields-per-table',       'mft', '=i' );

    ########################################
    $category = 7;    # Retaining or ignoring existing line breaks
    ########################################
    $add_option->( 'break-at-old-keyword-breakpoints',   'bok', '!' );
    $add_option->( 'break-at-old-logical-breakpoints',   'bol', '!' );
    $add_option->( 'break-at-old-ternary-breakpoints',   'bot', '!' );
    $add_option->( 'break-at-old-attribute-breakpoints', 'boa', '!' );
    $add_option->( 'ignore-old-breakpoints',             'iob', '!' );

    ########################################
    $category = 8;    # Blank line control
    ########################################
    $add_option->( 'blanks-before-blocks',            'bbb',  '!' );
    $add_option->( 'blanks-before-comments',          'bbc',  '!' );
    $add_option->( 'blank-lines-before-subs',         'blbs', '=i' );
    $add_option->( 'blank-lines-before-packages',     'blbp', '=i' );
    $add_option->( 'long-block-line-count',           'lbl',  '=i' );
    $add_option->( 'maximum-consecutive-blank-lines', 'mbl',  '=i' );
    $add_option->( 'keep-old-blank-lines',            'kbl',  '=i' );

    ########################################
    $category = 9;    # Other controls
    ########################################
    $add_option->( 'delete-block-comments',        'dbc',  '!' );
    $add_option->( 'delete-closing-side-comments', 'dcsc', '!' );
    $add_option->( 'delete-pod',                   'dp',   '!' );
    $add_option->( 'delete-side-comments',         'dsc',  '!' );
    $add_option->( 'tee-block-comments',           'tbc',  '!' );
    $add_option->( 'tee-pod',                      'tp',   '!' );
    $add_option->( 'tee-side-comments',            'tsc',  '!' );
    $add_option->( 'look-for-autoloader',          'lal',  '!' );
    $add_option->( 'look-for-hash-bang',           'x',    '!' );
    $add_option->( 'look-for-selfloader',          'lsl',  '!' );
    $add_option->( 'pass-version-line',            'pvl',  '!' );

    ########################################
    $category = 13;    # Debugging
    ########################################
    $add_option->( 'DEBUG',                           'D',    '!' );
    $add_option->( 'DIAGNOSTICS',                     'I',    '!' );
    $add_option->( 'dump-defaults',                   'ddf',  '!' );
    $add_option->( 'dump-long-names',                 'dln',  '!' );
    $add_option->( 'dump-options',                    'dop',  '!' );
    $add_option->( 'dump-profile',                    'dpro', '!' );
    $add_option->( 'dump-short-names',                'dsn',  '!' );
    $add_option->( 'dump-token-types',                'dtt',  '!' );
    $add_option->( 'dump-want-left-space',            'dwls', '!' );
    $add_option->( 'dump-want-right-space',           'dwrs', '!' );
    $add_option->( 'fuzzy-line-length',               'fll',  '!' );
    $add_option->( 'help',                            'h',    '' );
    $add_option->( 'short-concatenation-item-length', 'scl',  '=i' );
    $add_option->( 'show-options',                    'opt',  '!' );
    $add_option->( 'version',                         'v',    '' );
    $add_option->( 'memoize',                         'mem',  '!' );

    #---------------------------------------------------------------------

    # The Perl::Tidy::HtmlWriter will add its own options to the string
    Perl::Tidy::HtmlWriter->make_getopt_long_names( \@option_string );

    ########################################
    # Set categories 10, 11, 12
    ########################################
    # Based on their known order
    $category = 12;    # HTML properties
    foreach my $opt (@option_string) {
        my $long_name = $opt;
        $long_name =~ s/(!|=.*|:.*)$//;
        unless ( defined( $option_category{$long_name} ) ) {
            if ( $long_name =~ /^html-linked/ ) {
                $category = 10;    # HTML options
            }
            elsif ( $long_name =~ /^pod2html/ ) {
                $category = 11;    # Pod2html
            }
            $option_category{$long_name} = $category_name[$category];
        }
    }

    #---------------------------------------------------------------
    # Assign valid ranges to certain options
    #---------------------------------------------------------------
    # In the future, these may be used to make preliminary checks
    # hash keys are long names
    # If key or value is undefined:
    #   strings may have any value
    #   integer ranges are >=0
    # If value is defined:
    #   value is [qw(any valid words)] for strings
    #   value is [min, max] for integers
    #   if min is undefined, there is no lower limit
    #   if max is undefined, there is no upper limit
    # Parameters not listed here have defaults
    %option_range = (
        'format'             => [ 'tidy', 'html', 'user' ],
        'output-line-ending' => [ 'dos',  'win',  'mac', 'unix' ],

        'block-brace-tightness'    => [ 0, 2 ],
        'brace-tightness'          => [ 0, 2 ],
        'paren-tightness'          => [ 0, 2 ],
        'square-bracket-tightness' => [ 0, 2 ],

        'block-brace-vertical-tightness'            => [ 0, 2 ],
        'brace-vertical-tightness'                  => [ 0, 2 ],
        'brace-vertical-tightness-closing'          => [ 0, 2 ],
        'paren-vertical-tightness'                  => [ 0, 2 ],
        'paren-vertical-tightness-closing'          => [ 0, 2 ],
        'square-bracket-vertical-tightness'         => [ 0, 2 ],
        'square-bracket-vertical-tightness-closing' => [ 0, 2 ],
        'vertical-tightness'                        => [ 0, 2 ],
        'vertical-tightness-closing'                => [ 0, 2 ],

        'closing-brace-indentation'          => [ 0, 3 ],
        'closing-paren-indentation'          => [ 0, 3 ],
        'closing-square-bracket-indentation' => [ 0, 3 ],
        'closing-token-indentation'          => [ 0, 3 ],

        'closing-side-comment-else-flag' => [ 0, 2 ],
        'comma-arrow-breakpoints'        => [ 0, 5 ],
    );

    # Note: we could actually allow negative ci if someone really wants it:
    # $option_range{'continuation-indentation'} = [ undef, undef ];

    #---------------------------------------------------------------
    # Assign default values to the above options here, except
    # for 'outfile' and 'help'.
    # These settings should approximate the perlstyle(1) suggestions.
    #---------------------------------------------------------------
    my @defaults = qw(
      add-newlines
      add-semicolons
      add-whitespace
      blanks-before-blocks
      blanks-before-comments
      blank-lines-before-subs=1
      blank-lines-before-packages=1
      block-brace-tightness=0
      block-brace-vertical-tightness=0
      brace-tightness=1
      brace-vertical-tightness-closing=0
      brace-vertical-tightness=0
      break-at-old-logical-breakpoints
      break-at-old-ternary-breakpoints
      break-at-old-attribute-breakpoints
      break-at-old-keyword-breakpoints
      comma-arrow-breakpoints=5
      nocheck-syntax
      closing-side-comment-interval=6
      closing-side-comment-maximum-text=20
      closing-side-comment-else-flag=0
      closing-side-comments-balanced
      closing-paren-indentation=0
      closing-brace-indentation=0
      closing-square-bracket-indentation=0
      continuation-indentation=2
      delete-old-newlines
      delete-semicolons
      fuzzy-line-length
      hanging-side-comments
      indent-block-comments
      indent-columns=4
      iterations=1
      keep-old-blank-lines=1
      long-block-line-count=8
      look-for-autoloader
      look-for-selfloader
      maximum-consecutive-blank-lines=1
      maximum-fields-per-table=0
      maximum-line-length=80
      memoize
      minimum-space-to-comment=4
      nobrace-left-and-indent
      nocuddled-else
      nodelete-old-whitespace
      nohtml
      nologfile
      noquiet
      noshow-options
      nostatic-side-comments
      notabs
      nowarning-output
      outdent-labels
      outdent-long-quotes
      outdent-long-comments
      paren-tightness=1
      paren-vertical-tightness-closing=0
      paren-vertical-tightness=0
      pass-version-line
      recombine
      valign
      short-concatenation-item-length=8
      space-for-semicolon
      square-bracket-tightness=1
      square-bracket-vertical-tightness-closing=0
      square-bracket-vertical-tightness=0
      static-block-comments
      trim-qw
      format=tidy
      backup-file-extension=bak
      format-skipping
      default-tabsize=8

      pod2html
      html-table-of-contents
      html-entities
    );

    push @defaults, "perl-syntax-check-flags=-c -T";

    #---------------------------------------------------------------
    # Define abbreviations which will be expanded into the above primitives.
    # These may be defined recursively.
    #---------------------------------------------------------------
    %expansion = (
        %expansion,
        'freeze-newlines'   => [qw(noadd-newlines nodelete-old-newlines)],
        'fnl'               => [qw(freeze-newlines)],
        'freeze-whitespace' => [qw(noadd-whitespace nodelete-old-whitespace)],
        'fws'               => [qw(freeze-whitespace)],
        'freeze-blank-lines' =>
          [qw(maximum-consecutive-blank-lines=0 keep-old-blank-lines=2)],
        'fbl'                => [qw(freeze-blank-lines)],
        'indent-only'        => [qw(freeze-newlines freeze-whitespace)],
        'outdent-long-lines' => [qw(outdent-long-quotes outdent-long-comments)],
        'nooutdent-long-lines' =>
          [qw(nooutdent-long-quotes nooutdent-long-comments)],
        'noll' => [qw(nooutdent-long-lines)],
        'io'   => [qw(indent-only)],
        'delete-all-comments' =>
          [qw(delete-block-comments delete-side-comments delete-pod)],
        'nodelete-all-comments' =>
          [qw(nodelete-block-comments nodelete-side-comments nodelete-pod)],
        'dac'  => [qw(delete-all-comments)],
        'ndac' => [qw(nodelete-all-comments)],
        'gnu'  => [qw(gnu-style)],
        'pbp'  => [qw(perl-best-practices)],
        'tee-all-comments' =>
          [qw(tee-block-comments tee-side-comments tee-pod)],
        'notee-all-comments' =>
          [qw(notee-block-comments notee-side-comments notee-pod)],
        'tac'   => [qw(tee-all-comments)],
        'ntac'  => [qw(notee-all-comments)],
        'html'  => [qw(format=html)],
        'nhtml' => [qw(format=tidy)],
        'tidy'  => [qw(format=tidy)],

        'swallow-optional-blank-lines'   => [qw(kbl=0)],
        'noswallow-optional-blank-lines' => [qw(kbl=1)],
        'sob'                            => [qw(kbl=0)],
        'nsob'                           => [qw(kbl=1)],

        'break-after-comma-arrows'   => [qw(cab=0)],
        'nobreak-after-comma-arrows' => [qw(cab=1)],
        'baa'                        => [qw(cab=0)],
        'nbaa'                       => [qw(cab=1)],

        'blanks-before-subs'   => [qw(blbs=1 blbp=1)],
        'bbs'                  => [qw(blbs=1 blbp=1)],
        'noblanks-before-subs' => [qw(blbs=0 blbp=0)],
        'nbbs'                 => [qw(blbs=0 blbp=0)],

        'break-at-old-trinary-breakpoints' => [qw(bot)],

        'cti=0' => [qw(cpi=0 cbi=0 csbi=0)],
        'cti=1' => [qw(cpi=1 cbi=1 csbi=1)],
        'cti=2' => [qw(cpi=2 cbi=2 csbi=2)],
        'icp'   => [qw(cpi=2 cbi=2 csbi=2)],
        'nicp'  => [qw(cpi=0 cbi=0 csbi=0)],

        'closing-token-indentation=0' => [qw(cpi=0 cbi=0 csbi=0)],
        'closing-token-indentation=1' => [qw(cpi=1 cbi=1 csbi=1)],
        'closing-token-indentation=2' => [qw(cpi=2 cbi=2 csbi=2)],
        'indent-closing-paren'        => [qw(cpi=2 cbi=2 csbi=2)],
        'noindent-closing-paren'      => [qw(cpi=0 cbi=0 csbi=0)],

        'vt=0' => [qw(pvt=0 bvt=0 sbvt=0)],
        'vt=1' => [qw(pvt=1 bvt=1 sbvt=1)],
        'vt=2' => [qw(pvt=2 bvt=2 sbvt=2)],

        'vertical-tightness=0' => [qw(pvt=0 bvt=0 sbvt=0)],
        'vertical-tightness=1' => [qw(pvt=1 bvt=1 sbvt=1)],
        'vertical-tightness=2' => [qw(pvt=2 bvt=2 sbvt=2)],

        'vtc=0' => [qw(pvtc=0 bvtc=0 sbvtc=0)],
        'vtc=1' => [qw(pvtc=1 bvtc=1 sbvtc=1)],
        'vtc=2' => [qw(pvtc=2 bvtc=2 sbvtc=2)],

        'vertical-tightness-closing=0' => [qw(pvtc=0 bvtc=0 sbvtc=0)],
        'vertical-tightness-closing=1' => [qw(pvtc=1 bvtc=1 sbvtc=1)],
        'vertical-tightness-closing=2' => [qw(pvtc=2 bvtc=2 sbvtc=2)],

        'otr'                   => [qw(opr ohbr osbr)],
        'opening-token-right'   => [qw(opr ohbr osbr)],
        'notr'                  => [qw(nopr nohbr nosbr)],
        'noopening-token-right' => [qw(nopr nohbr nosbr)],

        'sot'                    => [qw(sop sohb sosb)],
        'nsot'                   => [qw(nsop nsohb nsosb)],
        'stack-opening-tokens'   => [qw(sop sohb sosb)],
        'nostack-opening-tokens' => [qw(nsop nsohb nsosb)],

        'sct'                    => [qw(scp schb scsb)],
        'stack-closing-tokens'   => => [qw(scp schb scsb)],
        'nsct'                   => [qw(nscp nschb nscsb)],
        'nostack-opening-tokens' => [qw(nscp nschb nscsb)],

        'sac'                    => [qw(sot sct)],
        'nsac'                   => [qw(nsot nsct)],
        'stack-all-containers'   => [qw(sot sct)],
        'nostack-all-containers' => [qw(nsot nsct)],

        'act=0'                      => [qw(pt=0 sbt=0 bt=0 bbt=0)],
        'act=1'                      => [qw(pt=1 sbt=1 bt=1 bbt=1)],
        'act=2'                      => [qw(pt=2 sbt=2 bt=2 bbt=2)],
        'all-containers-tightness=0' => [qw(pt=0 sbt=0 bt=0 bbt=0)],
        'all-containers-tightness=1' => [qw(pt=1 sbt=1 bt=1 bbt=1)],
        'all-containers-tightness=2' => [qw(pt=2 sbt=2 bt=2 bbt=2)],

        'stack-opening-block-brace'   => [qw(bbvt=2 bbvtl=*)],
        'sobb'                        => [qw(bbvt=2 bbvtl=*)],
        'nostack-opening-block-brace' => [qw(bbvt=0)],
        'nsobb'                       => [qw(bbvt=0)],

        'converge'   => [qw(it=4)],
        'noconverge' => [qw(it=1)],
        'conv'       => [qw(it=4)],
        'nconv'      => [qw(it=1)],

        # 'mangle' originally deleted pod and comments, but to keep it
        # reversible, it no longer does.  But if you really want to
        # delete them, just use:
        #   -mangle -dac

        # An interesting use for 'mangle' is to do this:
        #    perltidy -mangle myfile.pl -st | perltidy -o myfile.pl.new
        # which will form as many one-line blocks as possible

        'mangle' => [
            qw(
              check-syntax
              keep-old-blank-lines=0
              delete-old-newlines
              delete-old-whitespace
              delete-semicolons
              indent-columns=0
              maximum-consecutive-blank-lines=0
              maximum-line-length=100000
              noadd-newlines
              noadd-semicolons
              noadd-whitespace
              noblanks-before-blocks
              blank-lines-before-subs=0
              blank-lines-before-packages=0
              notabs
              )
        ],

        # 'extrude' originally deleted pod and comments, but to keep it
        # reversible, it no longer does.  But if you really want to
        # delete them, just use
        #   extrude -dac
        #
        # An interesting use for 'extrude' is to do this:
        #    perltidy -extrude myfile.pl -st | perltidy -o myfile.pl.new
        # which will break up all one-line blocks.

        'extrude' => [
            qw(
              check-syntax
              ci=0
              delete-old-newlines
              delete-old-whitespace
              delete-semicolons
              indent-columns=0
              maximum-consecutive-blank-lines=0
              maximum-line-length=1
              noadd-semicolons
              noadd-whitespace
              noblanks-before-blocks
              blank-lines-before-subs=0
              blank-lines-before-packages=0
              nofuzzy-line-length
              notabs
              norecombine
              )
        ],

        # this style tries to follow the GNU Coding Standards (which do
        # not really apply to perl but which are followed by some perl
        # programmers).
        'gnu-style' => [
            qw(
              lp bl noll pt=2 bt=2 sbt=2 cpi=1 csbi=1 cbi=1
              )
        ],

        # Style suggested in Damian Conway's Perl Best Practices
        'perl-best-practices' => [
            qw(l=78 i=4 ci=4 st se vt=2 cti=0 pt=1 bt=1 sbt=1 bbt=1 nsfs nolq),
q(wbb=% + - * / x != == >= <= =~ !~ < > | & = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=)
        ],

        # Additional styles can be added here
    );

    Perl::Tidy::HtmlWriter->make_abbreviated_names( \%expansion );

    # Uncomment next line to dump all expansions for debugging:
    # dump_short_names(\%expansion);
    return (
        \@option_string,   \@defaults, \%expansion,
        \%option_category, \%option_range
    );

}    # end of generate_options

# Memoize process_command_line. Given same @ARGV passed in, return same
# values and same @ARGV back.
# This patch was supplied by Jonathan Swartz Nov 2012 and significantly speeds
# up masontidy (https://metacpan.org/module/masontidy)

my %process_command_line_cache;

sub process_command_line {

    my (
        $perltidyrc_stream,  $is_Windows, $Windows_type,
        $rpending_complaint, $dump_options_type
    ) = @_;

    my $use_cache = !defined($perltidyrc_stream) && !$dump_options_type;
    if ($use_cache) {
        my $cache_key = join( chr(28), @ARGV );
        if ( my $result = $process_command_line_cache{$cache_key} ) {
            my ( $argv, @retvals ) = @$result;
            @ARGV = @$argv;
            return @retvals;
        }
        else {
            my @retvals = _process_command_line(@_);
            $process_command_line_cache{$cache_key} = [ \@ARGV, @retvals ]
              if $retvals[0]->{'memoize'};
            return @retvals;
        }
    }
    else {
        return _process_command_line(@_);
    }
}

# (note the underscore here)
sub _process_command_line {

    my (
        $perltidyrc_stream,  $is_Windows, $Windows_type,
        $rpending_complaint, $dump_options_type
    ) = @_;

    use Getopt::Long;

    my (
        $roption_string,   $rdefaults, $rexpansion,
        $roption_category, $roption_range
    ) = generate_options();

    #---------------------------------------------------------------
    # set the defaults by passing the above list through GetOptions
    #---------------------------------------------------------------
    my %Opts = ();
    {
        local @ARGV;
        my $i;

        # do not load the defaults if we are just dumping perltidyrc
        unless ( $dump_options_type eq 'perltidyrc' ) {
            for $i (@$rdefaults) { push @ARGV, "--" . $i }
        }

        # Patch to save users Getopt::Long configuration
        # and set to Getopt::Long defaults.  Use eval to avoid
        # breaking old versions of Perl without these routines.
        my $glc;
        eval { $glc = Getopt::Long::Configure() };
        unless ($@) {
            eval { Getopt::Long::ConfigDefaults() };
        }
        else { $glc = undef }

        if ( !GetOptions( \%Opts, @$roption_string ) ) {
            Die "Programming Bug: error in setting default options";
        }

        # Patch to put the previous Getopt::Long configuration back
        eval { Getopt::Long::Configure($glc) } if defined $glc;
    }

    my $word;
    my @raw_options        = ();
    my $config_file        = "";
    my $saw_ignore_profile = 0;
    my $saw_extrude        = 0;
    my $saw_pbp            = 0;
    my $saw_dump_profile   = 0;
    my $i;

    #---------------------------------------------------------------
    # Take a first look at the command-line parameters.  Do as many
    # immediate dumps as possible, which can avoid confusion if the
    # perltidyrc file has an error.
    #---------------------------------------------------------------
    foreach $i (@ARGV) {

        $i =~ s/^--/-/;
        if ( $i =~ /^-(npro|noprofile|no-profile)$/ ) {
            $saw_ignore_profile = 1;
        }

        # note: this must come before -pro and -profile, below:
        elsif ( $i =~ /^-(dump-profile|dpro)$/ ) {
            $saw_dump_profile = 1;
        }
        elsif ( $i =~ /^-(pro|profile)=(.+)/ ) {
            if ($config_file) {
                Warn
"Only one -pro=filename allowed, using '$2' instead of '$config_file'\n";
            }
            $config_file = $2;

            # resolve <dir>/.../<file>, meaning look upwards from directory
            if ( defined($config_file) ) {
                if ( my ( $start_dir, $search_file ) =
                    ( $config_file =~ m{^(.*)\.\.\./(.*)$} ) )
                {
                    $start_dir = '.' if !$start_dir;
                    $start_dir = Cwd::realpath($start_dir);
                    if ( my $found_file =
                        find_file_upwards( $start_dir, $search_file ) )
                    {
                        $config_file = $found_file;
                    }
                }
            }
            unless ( -e $config_file ) {
                Warn "cannot find file given with -pro=$config_file: $!\n";
                $config_file = "";
            }
        }
        elsif ( $i =~ /^-(pro|profile)=?$/ ) {
            Die "usage: -pro=filename or --profile=filename, no spaces\n";
        }
        elsif ( $i =~ /^-extrude$/ ) {
            $saw_extrude = 1;
        }
        elsif ( $i =~ /^-(pbp|perl-best-practices)$/ ) {
            $saw_pbp = 1;
        }
        elsif ( $i =~ /^-(help|h|HELP|H|\?)$/ ) {
            usage();
            Exit 0;
        }
        elsif ( $i =~ /^-(version|v)$/ ) {
            show_version();
            Exit 0;
        }
        elsif ( $i =~ /^-(dump-defaults|ddf)$/ ) {
            dump_defaults(@$rdefaults);
            Exit 0;
        }
        elsif ( $i =~ /^-(dump-long-names|dln)$/ ) {
            dump_long_names(@$roption_string);
            Exit 0;
        }
        elsif ( $i =~ /^-(dump-short-names|dsn)$/ ) {
            dump_short_names($rexpansion);
            Exit 0;
        }
        elsif ( $i =~ /^-(dump-token-types|dtt)$/ ) {
            Perl::Tidy::Tokenizer->dump_token_types(*STDOUT);
            Exit 0;
        }
    }

    if ( $saw_dump_profile && $saw_ignore_profile ) {
        Warn "No profile to dump because of -npro\n";
        Exit 1;
    }

    #---------------------------------------------------------------
    # read any .perltidyrc configuration file
    #---------------------------------------------------------------
    unless ($saw_ignore_profile) {

        # resolve possible conflict between $perltidyrc_stream passed
        # as call parameter to perltidy and -pro=filename on command
        # line.
        if ($perltidyrc_stream) {
            if ($config_file) {
                Warn <<EOM;
 Conflict: a perltidyrc configuration file was specified both as this
 perltidy call parameter: $perltidyrc_stream
 and with this -profile=$config_file.
 Using -profile=$config_file.
EOM
            }
            else {
                $config_file = $perltidyrc_stream;
            }
        }

        # look for a config file if we don't have one yet
        my $rconfig_file_chatter;
        $$rconfig_file_chatter = "";
        $config_file =
          find_config_file( $is_Windows, $Windows_type, $rconfig_file_chatter,
            $rpending_complaint )
          unless $config_file;

        # open any config file
        my $fh_config;
        if ($config_file) {
            ( $fh_config, $config_file ) =
              Perl::Tidy::streamhandle( $config_file, 'r' );
            unless ($fh_config) {
                $$rconfig_file_chatter .=
                  "# $config_file exists but cannot be opened\n";
            }
        }

        if ($saw_dump_profile) {
            dump_config_file( $fh_config, $config_file, $rconfig_file_chatter );
            Exit 0;
        }

        if ($fh_config) {

            my ( $rconfig_list, $death_message, $_saw_pbp ) =
              read_config_file( $fh_config, $config_file, $rexpansion );
            Die $death_message if ($death_message);
            $saw_pbp ||= $_saw_pbp;

            # process any .perltidyrc parameters right now so we can
            # localize errors
            if (@$rconfig_list) {
                local @ARGV = @$rconfig_list;

                expand_command_abbreviations( $rexpansion, \@raw_options,
                    $config_file );

                if ( !GetOptions( \%Opts, @$roption_string ) ) {
                    Die
"Error in this config file: $config_file  \nUse -npro to ignore this file, -h for help'\n";
                }

                # Anything left in this local @ARGV is an error and must be
                # invalid bare words from the configuration file.  We cannot
                # check this earlier because bare words may have been valid
                # values for parameters.  We had to wait for GetOptions to have
                # a look at @ARGV.
                if (@ARGV) {
                    my $count = @ARGV;
                    my $str   = "\'" . pop(@ARGV) . "\'";
                    while ( my $param = pop(@ARGV) ) {
                        if ( length($str) < 70 ) {
                            $str .= ", '$param'";
                        }
                        else {
                            $str .= ", ...";
                            last;
                        }
                    }
                    Die <<EOM;
There are $count unrecognized values in the configuration file '$config_file':
$str
Use leading dashes for parameters.  Use -npro to ignore this file.
EOM
                }

                # Undo any options which cause premature exit.  They are not
                # appropriate for a config file, and it could be hard to
                # diagnose the cause of the premature exit.
                foreach (
                    qw{
                    dump-defaults
                    dump-long-names
                    dump-options
                    dump-profile
                    dump-short-names
                    dump-token-types
                    dump-want-left-space
                    dump-want-right-space
                    help
                    stylesheet
                    version
                    }
                  )
                {

                    if ( defined( $Opts{$_} ) ) {
                        delete $Opts{$_};
                        Warn "ignoring --$_ in config file: $config_file\n";
                    }
                }
            }
        }
    }

    #---------------------------------------------------------------
    # now process the command line parameters
    #---------------------------------------------------------------
    expand_command_abbreviations( $rexpansion, \@raw_options, $config_file );

    local $SIG{'__WARN__'} = sub { Warn $_[0] };
    if ( !GetOptions( \%Opts, @$roption_string ) ) {
        Die "Error on command line; for help try 'perltidy -h'\n";
    }

    return (
        \%Opts,       $config_file,      \@raw_options,
        $saw_extrude, $saw_pbp,          $roption_string,
        $rexpansion,  $roption_category, $roption_range
    );
}    # end of process_command_line

sub check_options {

    my ( $rOpts, $is_Windows, $Windows_type, $rpending_complaint ) = @_;

    #---------------------------------------------------------------
    # check and handle any interactions among the basic options..
    #---------------------------------------------------------------

    # Since -vt, -vtc, and -cti are abbreviations, but under
    # msdos, an unquoted input parameter like vtc=1 will be
    # seen as 2 parameters, vtc and 1, so the abbreviations
    # won't be seen.  Therefore, we will catch them here if
    # they get through.

    if ( defined $rOpts->{'vertical-tightness'} ) {
        my $vt = $rOpts->{'vertical-tightness'};
        $rOpts->{'paren-vertical-tightness'}          = $vt;
        $rOpts->{'square-bracket-vertical-tightness'} = $vt;
        $rOpts->{'brace-vertical-tightness'}          = $vt;
    }

    if ( defined $rOpts->{'vertical-tightness-closing'} ) {
        my $vtc = $rOpts->{'vertical-tightness-closing'};
        $rOpts->{'paren-vertical-tightness-closing'}          = $vtc;
        $rOpts->{'square-bracket-vertical-tightness-closing'} = $vtc;
        $rOpts->{'brace-vertical-tightness-closing'}          = $vtc;
    }

    if ( defined $rOpts->{'closing-token-indentation'} ) {
        my $cti = $rOpts->{'closing-token-indentation'};
        $rOpts->{'closing-square-bracket-indentation'} = $cti;
        $rOpts->{'closing-brace-indentation'}          = $cti;
        $rOpts->{'closing-paren-indentation'}          = $cti;
    }

    # In quiet mode, there is no log file and hence no way to report
    # results of syntax check, so don't do it.
    if ( $rOpts->{'quiet'} ) {
        $rOpts->{'check-syntax'} = 0;
    }

    # can't check syntax if no output
    if ( $rOpts->{'format'} ne 'tidy' ) {
        $rOpts->{'check-syntax'} = 0;
    }

    # Never let Windows 9x/Me systems run syntax check -- this will prevent a
    # wide variety of nasty problems on these systems, because they cannot
    # reliably run backticks.  Don't even think about changing this!
    if (   $rOpts->{'check-syntax'}
        && $is_Windows
        && ( !$Windows_type || $Windows_type =~ /^(9|Me)/ ) )
    {
        $rOpts->{'check-syntax'} = 0;
    }

    # It's really a bad idea to check syntax as root unless you wrote
    # the script yourself.  FIXME: not sure if this works with VMS
    unless ($is_Windows) {

        if ( $< == 0 && $rOpts->{'check-syntax'} ) {
            $rOpts->{'check-syntax'} = 0;
            $$rpending_complaint .=
"Syntax check deactivated for safety; you shouldn't run this as root\n";
        }
    }

    # check iteration count and quietly fix if necessary:
    # - iterations option only applies to code beautification mode
    # - the convergence check should stop most runs on iteration 2, and
    #   virtually all on iteration 3.  But we'll allow up to 6.
    if ( $rOpts->{'format'} ne 'tidy' ) {
        $rOpts->{'iterations'} = 1;
    }
    elsif ( defined( $rOpts->{'iterations'} ) ) {
        if    ( $rOpts->{'iterations'} <= 0 ) { $rOpts->{'iterations'} = 1 }
        elsif ( $rOpts->{'iterations'} > 6 )  { $rOpts->{'iterations'} = 6 }
    }
    else {
        $rOpts->{'iterations'} = 1;
    }

    # check for reasonable number of blank lines and fix to avoid problems
    if ( $rOpts->{'blank-lines-before-subs'} ) {
        if ( $rOpts->{'blank-lines-before-subs'} < 0 ) {
            $rOpts->{'blank-lines-before-subs'} = 0;
            Warn "negative value of -blbs, setting 0\n";
        }
        if ( $rOpts->{'blank-lines-before-subs'} > 100 ) {
            Warn "unreasonably large value of -blbs, reducing\n";
            $rOpts->{'blank-lines-before-subs'} = 100;
        }
    }
    if ( $rOpts->{'blank-lines-before-packages'} ) {
        if ( $rOpts->{'blank-lines-before-packages'} < 0 ) {
            Warn "negative value of -blbp, setting 0\n";
            $rOpts->{'blank-lines-before-packages'} = 0;
        }
        if ( $rOpts->{'blank-lines-before-packages'} > 100 ) {
            Warn "unreasonably large value of -blbp, reducing\n";
            $rOpts->{'blank-lines-before-packages'} = 100;
        }
    }

    # see if user set a non-negative logfile-gap
    if ( defined( $rOpts->{'logfile-gap'} ) && $rOpts->{'logfile-gap'} >= 0 ) {

        # a zero gap will be taken as a 1
        if ( $rOpts->{'logfile-gap'} == 0 ) {
            $rOpts->{'logfile-gap'} = 1;
        }

        # setting a non-negative logfile gap causes logfile to be saved
        $rOpts->{'logfile'} = 1;
    }

    # not setting logfile gap, or setting it negative, causes default of 50
    else {
        $rOpts->{'logfile-gap'} = 50;
    }

    # set short-cut flag when only indentation is to be done.
    # Note that the user may or may not have already set the
    # indent-only flag.
    if (   !$rOpts->{'add-whitespace'}
        && !$rOpts->{'delete-old-whitespace'}
        && !$rOpts->{'add-newlines'}
        && !$rOpts->{'delete-old-newlines'} )
    {
        $rOpts->{'indent-only'} = 1;
    }

    # -isbc implies -ibc
    if ( $rOpts->{'indent-spaced-block-comments'} ) {
        $rOpts->{'indent-block-comments'} = 1;
    }

    # -bli flag implies -bl
    if ( $rOpts->{'brace-left-and-indent'} ) {
        $rOpts->{'opening-brace-on-new-line'} = 1;
    }

    if (   $rOpts->{'opening-brace-always-on-right'}
        && $rOpts->{'opening-brace-on-new-line'} )
    {
        Warn <<EOM;
 Conflict: you specified both 'opening-brace-always-on-right' (-bar) and
  'opening-brace-on-new-line' (-bl).  Ignoring -bl.
EOM
        $rOpts->{'opening-brace-on-new-line'} = 0;
    }

    # it simplifies things if -bl is 0 rather than undefined
    if ( !defined( $rOpts->{'opening-brace-on-new-line'} ) ) {
        $rOpts->{'opening-brace-on-new-line'} = 0;
    }

    # -sbl defaults to -bl if not defined
    if ( !defined( $rOpts->{'opening-sub-brace-on-new-line'} ) ) {
        $rOpts->{'opening-sub-brace-on-new-line'} =
          $rOpts->{'opening-brace-on-new-line'};
    }

    if ( $rOpts->{'entab-leading-whitespace'} ) {
        if ( $rOpts->{'entab-leading-whitespace'} < 0 ) {
            Warn "-et=n must use a positive integer; ignoring -et\n";
            $rOpts->{'entab-leading-whitespace'} = undef;
        }

        # entab leading whitespace has priority over the older 'tabs' option
        if ( $rOpts->{'tabs'} ) { $rOpts->{'tabs'} = 0; }
    }

    # set a default tabsize to be used in guessing the starting indentation
    # level if and only if this run does not use tabs and the old code does
    # use tabs
    if ( $rOpts->{'default-tabsize'} ) {
        if ( $rOpts->{'default-tabsize'} < 0 ) {
            Warn "negative value of -dt, setting 0\n";
            $rOpts->{'default-tabsize'} = 0;
        }
        if ( $rOpts->{'default-tabsize'} > 20 ) {
            Warn "unreasonably large value of -dt, reducing\n";
            $rOpts->{'default-tabsize'} = 20;
        }
    }
    else {
        $rOpts->{'default-tabsize'} = 8;
    }

    # Define $tabsize, the number of spaces per tab for use in
    # guessing the indentation of source lines with leading tabs.
    # Assume same as for this run if tabs are used , otherwise assume
    # a default value, typically 8
    my $tabsize =
        $rOpts->{'entab-leading-whitespace'}
      ? $rOpts->{'entab-leading-whitespace'}
      : $rOpts->{'tabs'} ? $rOpts->{'indent-columns'}
      :                    $rOpts->{'default-tabsize'};
    return $tabsize;
}

sub find_file_upwards {
    my ( $search_dir, $search_file ) = @_;

    $search_dir =~ s{/+$}{};
    $search_file =~ s{^/+}{};

    while (1) {
        my $try_path = "$search_dir/$search_file";
        if ( -f $try_path ) {
            return $try_path;
        }
        elsif ( $search_dir eq '/' ) {
            return undef;
        }
        else {
            $search_dir = dirname($search_dir);
        }
    }
}

sub expand_command_abbreviations {

    # go through @ARGV and expand any abbreviations

    my ( $rexpansion, $rraw_options, $config_file ) = @_;
    my ($word);

    # set a pass limit to prevent an infinite loop;
    # 10 should be plenty, but it may be increased to allow deeply
    # nested expansions.
    my $max_passes = 10;
    my @new_argv   = ();

    # keep looping until all expansions have been converted into actual
    # dash parameters..
    for ( my $pass_count = 0 ; $pass_count <= $max_passes ; $pass_count++ ) {
        my @new_argv     = ();
        my $abbrev_count = 0;

        # loop over each item in @ARGV..
        foreach $word (@ARGV) {

            # convert any leading 'no-' to just 'no'
            if ( $word =~ /^(-[-]?no)-(.*)/ ) { $word = $1 . $2 }

            # if it is a dash flag (instead of a file name)..
            if ( $word =~ /^-[-]?([\w\-]+)(.*)/ ) {

                my $abr   = $1;
                my $flags = $2;

                # save the raw input for debug output in case of circular refs
                if ( $pass_count == 0 ) {
                    push( @$rraw_options, $word );
                }

                # recombine abbreviation and flag, if necessary,
                # to allow abbreviations with arguments such as '-vt=1'
                if ( $rexpansion->{ $abr . $flags } ) {
                    $abr   = $abr . $flags;
                    $flags = "";
                }

                # if we see this dash item in the expansion hash..
                if ( $rexpansion->{$abr} ) {
                    $abbrev_count++;

                    # stuff all of the words that it expands to into the
                    # new arg list for the next pass
                    foreach my $abbrev ( @{ $rexpansion->{$abr} } ) {
                        next unless $abbrev;    # for safety; shouldn't happen
                        push( @new_argv, '--' . $abbrev . $flags );
                    }
                }

                # not in expansion hash, must be actual long name
                else {
                    push( @new_argv, $word );
                }
            }

            # not a dash item, so just save it for the next pass
            else {
                push( @new_argv, $word );
            }
        }    # end of this pass

        # update parameter list @ARGV to the new one
        @ARGV = @new_argv;
        last unless ( $abbrev_count > 0 );

        # make sure we are not in an infinite loop
        if ( $pass_count == $max_passes ) {
            local $" = ')(';
            Warn <<EOM;
I'm tired. We seem to be in an infinite loop trying to expand aliases.
Here are the raw options;
(rraw_options)
EOM
            my $num = @new_argv;
            if ( $num < 50 ) {
                Warn <<EOM;
After $max_passes passes here is ARGV
(@new_argv)
EOM
            }
            else {
                Warn <<EOM;
After $max_passes passes ARGV has $num entries
EOM
            }

            if ($config_file) {
                Die <<"DIE";
Please check your configuration file $config_file for circular-references.
To deactivate it, use -npro.
DIE
            }
            else {
                Die <<'DIE';
Program bug - circular-references in the %expansion hash, probably due to
a recent program change.
DIE
            }
        }    # end of check for circular references
    }    # end of loop over all passes
}

# Debug routine -- this will dump the expansion hash
sub dump_short_names {
    my $rexpansion = shift;
    print STDOUT <<EOM;
List of short names.  This list shows how all abbreviations are
translated into other abbreviations and, eventually, into long names.
New abbreviations may be defined in a .perltidyrc file.
For a list of all long names, use perltidy --dump-long-names (-dln).
--------------------------------------------------------------------------
EOM
    foreach my $abbrev ( sort keys %$rexpansion ) {
        my @list = @{ $$rexpansion{$abbrev} };
        print STDOUT "$abbrev --> @list\n";
    }
}

sub check_vms_filename {

    # given a valid filename (the perltidy input file)
    # create a modified filename and separator character
    # suitable for VMS.
    #
    # Contributed by Michael Cartmell
    #
    my ( $base, $path ) = fileparse( $_[0] );

    # remove explicit ; version
    $base =~ s/;-?\d*$//

      # remove explicit . version ie two dots in filename NB ^ escapes a dot
      or $base =~ s/(          # begin capture $1
                  (?:^|[^^])\. # match a dot not preceded by a caret
                  (?:          # followed by nothing
                    |          # or
                    .*[^^]     # anything ending in a non caret
                  )
                )              # end capture $1
                \.-?\d*$       # match . version number
              /$1/x;

    # normalise filename, if there are no unescaped dots then append one
    $base .= '.' unless $base =~ /(?:^|[^^])\./;

    # if we don't already have an extension then we just append the extension
    my $separator = ( $base =~ /\.$/ ) ? "" : "_";
    return ( $path . $base, $separator );
}

sub Win_OS_Type {

    # TODO: are these more standard names?
    # Win32s Win95 Win98 WinMe WinNT3.51 WinNT4 Win2000 WinXP/.Net Win2003

    # Returns a string that determines what MS OS we are on.
    # Returns win32s,95,98,Me,NT3.51,NT4,2000,XP/.Net,Win2003
    # Returns blank string if not an MS system.
    # Original code contributed by: Yves Orton
    # We need to know this to decide where to look for config files

    my $rpending_complaint = shift;
    my $os                 = "";
    return $os unless $^O =~ /win32|dos/i;    # is it a MS box?

    # Systems built from Perl source may not have Win32.pm
    # But probably have Win32::GetOSVersion() anyway so the
    # following line is not 'required':
    # return $os unless eval('require Win32');

    # Use the standard API call to determine the version
    my ( $undef, $major, $minor, $build, $id );
    eval { ( $undef, $major, $minor, $build, $id ) = Win32::GetOSVersion() };

    #
    #    NAME                   ID   MAJOR  MINOR
    #    Windows NT 4           2      4       0
    #    Windows 2000           2      5       0
    #    Windows XP             2      5       1
    #    Windows Server 2003    2      5       2

    return "win32s" unless $id;    # If id==0 then its a win32s box.
    $os = {                        # Magic numbers from MSDN
                                   # documentation of GetOSVersion
        1 => {
            0  => "95",
            10 => "98",
            90 => "Me"
        },
        2 => {
            0  => "2000",          # or NT 4, see below
            1  => "XP/.Net",
            2  => "Win2003",
            51 => "NT3.51"
        }
    }->{$id}->{$minor};

    # If $os is undefined, the above code is out of date.  Suggested updates
    # are welcome.
    unless ( defined $os ) {
        $os = "";
        $$rpending_complaint .= <<EOS;
Error trying to discover Win_OS_Type: $id:$major:$minor Has no name of record!
We won't be able to look for a system-wide config file.
EOS
    }

    # Unfortunately the logic used for the various versions isn't so clever..
    # so we have to handle an outside case.
    return ( $os eq "2000" && $major != 5 ) ? "NT4" : $os;
}

sub is_unix {
    return
         ( $^O !~ /win32|dos/i )
      && ( $^O ne 'VMS' )
      && ( $^O ne 'OS2' )
      && ( $^O ne 'MacOS' );
}

sub look_for_Windows {

    # determine Windows sub-type and location of
    # system-wide configuration files
    my $rpending_complaint = shift;
    my $is_Windows         = ( $^O =~ /win32|dos/i );
    my $Windows_type       = Win_OS_Type($rpending_complaint) if $is_Windows;
    return ( $is_Windows, $Windows_type );
}

sub find_config_file {

    # look for a .perltidyrc configuration file
    # For Windows also look for a file named perltidy.ini
    my ( $is_Windows, $Windows_type, $rconfig_file_chatter,
        $rpending_complaint ) = @_;

    $$rconfig_file_chatter .= "# Config file search...system reported as:";
    if ($is_Windows) {
        $$rconfig_file_chatter .= "Windows $Windows_type\n";
    }
    else {
        $$rconfig_file_chatter .= " $^O\n";
    }

    # sub to check file existence and record all tests
    my $exists_config_file = sub {
        my $config_file = shift;
        return 0 unless $config_file;
        $$rconfig_file_chatter .= "# Testing: $config_file\n";
        return -f $config_file;
    };

    my $config_file;

    # look in current directory first
    $config_file = ".perltidyrc";
    return $config_file if $exists_config_file->($config_file);
    if ($is_Windows) {
        $config_file = "perltidy.ini";
        return $config_file if $exists_config_file->($config_file);
    }

    # Default environment vars.
    my @envs = qw(PERLTIDY HOME);

    # Check the NT/2k/XP locations, first a local machine def, then a
    # network def
    push @envs, qw(USERPROFILE HOMESHARE) if $^O =~ /win32/i;

    # Now go through the environment ...
    foreach my $var (@envs) {
        $$rconfig_file_chatter .= "# Examining: \$ENV{$var}";
        if ( defined( $ENV{$var} ) ) {
            $$rconfig_file_chatter .= " = $ENV{$var}\n";

            # test ENV{ PERLTIDY } as file:
            if ( $var eq 'PERLTIDY' ) {
                $config_file = "$ENV{$var}";
                return $config_file if $exists_config_file->($config_file);
            }

            # test ENV as directory:
            $config_file = catfile( $ENV{$var}, ".perltidyrc" );
            return $config_file if $exists_config_file->($config_file);

            if ($is_Windows) {
                $config_file = catfile( $ENV{$var}, "perltidy.ini" );
                return $config_file if $exists_config_file->($config_file);
            }
        }
        else {
            $$rconfig_file_chatter .= "\n";
        }
    }

    # then look for a system-wide definition
    # where to look varies with OS
    if ($is_Windows) {

        if ($Windows_type) {
            my ( $os, $system, $allusers ) =
              Win_Config_Locs( $rpending_complaint, $Windows_type );

            # Check All Users directory, if there is one.
            # i.e. C:\Documents and Settings\User\perltidy.ini
            if ($allusers) {

                $config_file = catfile( $allusers, ".perltidyrc" );
                return $config_file if $exists_config_file->($config_file);

                $config_file = catfile( $allusers, "perltidy.ini" );
                return $config_file if $exists_config_file->($config_file);
            }

            # Check system directory.
            # retain old code in case someone has been able to create
            # a file with a leading period.
            $config_file = catfile( $system, ".perltidyrc" );
            return $config_file if $exists_config_file->($config_file);

            $config_file = catfile( $system, "perltidy.ini" );
            return $config_file if $exists_config_file->($config_file);
        }
    }

    # Place to add customization code for other systems
    elsif ( $^O eq 'OS2' ) {
    }
    elsif ( $^O eq 'MacOS' ) {
    }
    elsif ( $^O eq 'VMS' ) {
    }

    # Assume some kind of Unix
    else {

        $config_file = "/usr/local/etc/perltidyrc";
        return $config_file if $exists_config_file->($config_file);

        $config_file = "/etc/perltidyrc";
        return $config_file if $exists_config_file->($config_file);
    }

    # Couldn't find a config file
    return;
}

sub Win_Config_Locs {

    # In scalar context returns the OS name (95 98 ME NT3.51 NT4 2000 XP),
    # or undef if its not a win32 OS.  In list context returns OS, System
    # Directory, and All Users Directory.  All Users will be empty on a
    # 9x/Me box.  Contributed by: Yves Orton.

    my $rpending_complaint = shift;
    my $os = (@_) ? shift : Win_OS_Type();
    return unless $os;

    my $system   = "";
    my $allusers = "";

    if ( $os =~ /9[58]|Me/ ) {
        $system = "C:/Windows";
    }
    elsif ( $os =~ /NT|XP|200?/ ) {
        $system = ( $os =~ /XP/ ) ? "C:/Windows/" : "C:/WinNT/";
        $allusers =
          ( $os =~ /NT/ )
          ? "C:/WinNT/profiles/All Users/"
          : "C:/Documents and Settings/All Users/";
    }
    else {

        # This currently would only happen on a win32s computer.  I dont have
        # one to test, so I am unsure how to proceed.  Suggestions welcome!
        $$rpending_complaint .=
"I dont know a sensible place to look for config files on an $os system.\n";
        return;
    }
    return wantarray ? ( $os, $system, $allusers ) : $os;
}

sub dump_config_file {
    my $fh                   = shift;
    my $config_file          = shift;
    my $rconfig_file_chatter = shift;
    print STDOUT "$$rconfig_file_chatter";
    if ($fh) {
        print STDOUT "# Dump of file: '$config_file'\n";
        while ( my $line = $fh->getline() ) { print STDOUT $line }
        eval { $fh->close() };
    }
    else {
        print STDOUT "# ...no config file found\n";
    }
}

sub read_config_file {

    my ( $fh, $config_file, $rexpansion ) = @_;
    my @config_list = ();
    my $saw_pbp;

    # file is bad if non-empty $death_message is returned
    my $death_message = "";

    my $name = undef;
    my $line_no;
    while ( my $line = $fh->getline() ) {
        $line_no++;
        chomp $line;
        ( $line, $death_message ) =
          strip_comment( $line, $config_file, $line_no );
        last if ($death_message);
        next unless $line;
        $line =~ s/^\s*(.*?)\s*$/$1/;    # trim both ends
        next unless $line;

        # look for something of the general form
        #    newname { body }
        # or just
        #    body

        my $body = $line;
        my ($newname);
        if ( $line =~ /^((\w+)\s*\{)(.*)\}$/ ) {
            ( $newname, $body ) = ( $2, $3, );
        }
        if ($body) {

            if ( !$saw_pbp && $body =~ /-(pbp|perl-best-practices)/ ) {
                $saw_pbp = 1;
            }

            # handle a new alias definition
            if ($newname) {
                if ($name) {
                    $death_message =
"No '}' seen after $name and before $newname in config file $config_file line $.\n";
                    last;
                }
                $name = $newname;

                if ( ${$rexpansion}{$name} ) {
                    local $" = ')(';
                    my @names = sort keys %$rexpansion;
                    $death_message =
                        "Here is a list of all installed aliases\n(@names)\n"
                      . "Attempting to redefine alias ($name) in config file $config_file line $.\n";
                    last;
                }
                ${$rexpansion}{$name} = [];
            }

            # now do the body
            if ($body) {

                my ( $rbody_parts, $msg ) = parse_args($body);
                if ($msg) {
                    $death_message = <<EOM;
Error reading file '$config_file' at line number $line_no.
$msg
Please fix this line or use -npro to avoid reading this file
EOM
                    last;
                }

                if ($name) {

                    # remove leading dashes if this is an alias
                    foreach (@$rbody_parts) { s/^\-+//; }
                    push @{ ${$rexpansion}{$name} }, @$rbody_parts;
                }
                else {
                    push( @config_list, @$rbody_parts );
                }
            }
        }
    }
    eval { $fh->close() };
    return ( \@config_list, $death_message, $saw_pbp );
}

sub strip_comment {

    # Strip any comment from a command line
    my ( $instr, $config_file, $line_no ) = @_;
    my $msg = "";

    # check for full-line comment
    if ( $instr =~ /^\s*#/ ) {
        return ( "", $msg );
    }

    # nothing to do if no comments
    if ( $instr !~ /#/ ) {
        return ( $instr, $msg );
    }

    # handle case of no quotes
    elsif ( $instr !~ /['"]/ ) {

        # We now require a space before the # of a side comment
        # this allows something like:
        #    -sbcp=#
        # Otherwise, it would have to be quoted:
        #    -sbcp='#'
        $instr =~ s/\s+\#.*$//;
        return ( $instr, $msg );
    }

    # handle comments and quotes
    my $outstr     = "";
    my $quote_char = "";
    while (1) {

        # looking for ending quote character
        if ($quote_char) {
            if ( $instr =~ /\G($quote_char)/gc ) {
                $quote_char = "";
                $outstr .= $1;
            }
            elsif ( $instr =~ /\G(.)/gc ) {
                $outstr .= $1;
            }

            # error..we reached the end without seeing the ending quote char
            else {
                $msg = <<EOM;
Error reading file $config_file at line number $line_no.
Did not see ending quote character <$quote_char> in this text:
$instr
Please fix this line or use -npro to avoid reading this file
EOM
                last;
            }
        }

        # accumulating characters and looking for start of a quoted string
        else {
            if ( $instr =~ /\G([\"\'])/gc ) {
                $outstr .= $1;
                $quote_char = $1;
            }

            # Note: not yet enforcing the space-before-hash rule for side
            # comments if the parameter is quoted.
            elsif ( $instr =~ /\G#/gc ) {
                last;
            }
            elsif ( $instr =~ /\G(.)/gc ) {
                $outstr .= $1;
            }
            else {
                last;
            }
        }
    }
    return ( $outstr, $msg );
}

sub parse_args {

    # Parse a command string containing multiple string with possible
    # quotes, into individual commands.  It might look like this, for example:
    #
    #    -wba=" + - "  -some-thing -wbb='. && ||'
    #
    # There is no need, at present, to handle escaped quote characters.
    # (They are not perltidy tokens, so needn't be in strings).

    my ($body)     = @_;
    my @body_parts = ();
    my $quote_char = "";
    my $part       = "";
    my $msg        = "";
    while (1) {

        # looking for ending quote character
        if ($quote_char) {
            if ( $body =~ /\G($quote_char)/gc ) {
                $quote_char = "";
            }
            elsif ( $body =~ /\G(.)/gc ) {
                $part .= $1;
            }

            # error..we reached the end without seeing the ending quote char
            else {
                if ( length($part) ) { push @body_parts, $part; }
                $msg = <<EOM;
Did not see ending quote character <$quote_char> in this text:
$body
EOM
                last;
            }
        }

        # accumulating characters and looking for start of a quoted string
        else {
            if ( $body =~ /\G([\"\'])/gc ) {
                $quote_char = $1;
            }
            elsif ( $body =~ /\G(\s+)/gc ) {
                if ( length($part) ) { push @body_parts, $part; }
                $part = "";
            }
            elsif ( $body =~ /\G(.)/gc ) {
                $part .= $1;
            }
            else {
                if ( length($part) ) { push @body_parts, $part; }
                last;
            }
        }
    }
    return ( \@body_parts, $msg );
}

sub dump_long_names {

    my @names = sort @_;
    print STDOUT <<EOM;
# Command line long names (passed to GetOptions)
#---------------------------------------------------------------
# here is a summary of the Getopt codes:
# <none> does not take an argument
# =s takes a mandatory string
# :s takes an optional string
# =i takes a mandatory integer
# :i takes an optional integer
# ! does not take an argument and may be negated
#  i.e., -foo and -nofoo are allowed
# a double dash signals the end of the options list
#
#---------------------------------------------------------------
EOM

    foreach (@names) { print STDOUT "$_\n" }
}

sub dump_defaults {
    my @defaults = sort @_;
    print STDOUT "Default command line options:\n";
    foreach (@_) { print STDOUT "$_\n" }
}

sub readable_options {

    # return options for this run as a string which could be
    # put in a perltidyrc file
    my ( $rOpts, $roption_string ) = @_;
    my %Getopt_flags;
    my $rGetopt_flags    = \%Getopt_flags;
    my $readable_options = "# Final parameter set for this run.\n";
    $readable_options .=
      "# See utility 'perltidyrc_dump.pl' for nicer formatting.\n";
    foreach my $opt ( @{$roption_string} ) {
        my $flag = "";
        if ( $opt =~ /(.*)(!|=.*)$/ ) {
            $opt  = $1;
            $flag = $2;
        }
        if ( defined( $rOpts->{$opt} ) ) {
            $rGetopt_flags->{$opt} = $flag;
        }
    }
    foreach my $key ( sort keys %{$rOpts} ) {
        my $flag   = $rGetopt_flags->{$key};
        my $value  = $rOpts->{$key};
        my $prefix = '--';
        my $suffix = "";
        if ($flag) {
            if ( $flag =~ /^=/ ) {
                if ( $value !~ /^\d+$/ ) { $value = '"' . $value . '"' }
                $suffix = "=" . $value;
            }
            elsif ( $flag =~ /^!/ ) {
                $prefix .= "no" unless ($value);
            }
            else {

                # shouldn't happen
                $readable_options .=
                  "# ERROR in dump_options: unrecognized flag $flag for $key\n";
            }
        }
        $readable_options .= $prefix . $key . $suffix . "\n";
    }
    return $readable_options;
}

sub show_version {
    print STDOUT <<"EOM";
This is perltidy, v$VERSION

Copyright 2000-2012, Steve Hancock

Perltidy is free software and may be copied under the terms of the GNU
General Public License, which is included in the distribution files.

Complete documentation for perltidy can be found using 'man perltidy'
or on the internet at http://perltidy.sourceforge.net.
EOM
}

sub usage {

    print STDOUT <<EOF;
This is perltidy version $VERSION, a perl script indenter.  Usage:

    perltidy [ options ] file1 file2 file3 ...
            (output goes to file1.tdy, file2.tdy, file3.tdy, ...)
    perltidy [ options ] file1 -o outfile
    perltidy [ options ] file1 -st >outfile
    perltidy [ options ] <infile >outfile

Options have short and long forms. Short forms are shown; see
man pages for long forms.  Note: '=s' indicates a required string,
and '=n' indicates a required integer.

I/O control
 -h      show this help
 -o=file name of the output file (only if single input file)
 -oext=s change output extension from 'tdy' to s
 -opath=path  change path to be 'path' for output files
 -b      backup original to .bak and modify file in-place
 -bext=s change default backup extension from 'bak' to s
 -q      deactivate error messages (for running under editor)
 -w      include non-critical warning messages in the .ERR error output
 -syn    run perl -c to check syntax (default under unix systems)
 -log    save .LOG file, which has useful diagnostics
 -f      force perltidy to read a binary file
 -g      like -log but writes more detailed .LOG file, for debugging scripts
 -opt    write the set of options actually used to a .LOG file
 -npro   ignore .perltidyrc configuration command file
 -pro=file   read configuration commands from file instead of .perltidyrc
 -st     send output to standard output, STDOUT
 -se     send all error output to standard error output, STDERR
 -v      display version number to standard output and quit

Basic Options:
 -i=n    use n columns per indentation level (default n=4)
 -t      tabs: use one tab character per indentation level, not recommeded
 -nt     no tabs: use n spaces per indentation level (default)
 -et=n   entab leading whitespace n spaces per tab; not recommended
 -io     "indent only": just do indentation, no other formatting.
 -sil=n  set starting indentation level to n;  use if auto detection fails
 -ole=s  specify output line ending (s=dos or win, mac, unix)
 -ple    keep output line endings same as input (input must be filename)

Whitespace Control
 -fws    freeze whitespace; this disables all whitespace changes
           and disables the following switches:
 -bt=n   sets brace tightness,  n= (0 = loose, 1=default, 2 = tight)
 -bbt    same as -bt but for code block braces; same as -bt if not given
 -bbvt   block braces vertically tight; use with -bl or -bli
 -bbvtl=s  make -bbvt to apply to selected list of block types
 -pt=n   paren tightness (n=0, 1 or 2)
 -sbt=n  square bracket tightness (n=0, 1, or 2)
 -bvt=n  brace vertical tightness,
         n=(0=open, 1=close unless multiple steps on a line, 2=always close)
 -pvt=n  paren vertical tightness (see -bvt for n)
 -sbvt=n square bracket vertical tightness (see -bvt for n)
 -bvtc=n closing brace vertical tightness:
         n=(0=open, 1=sometimes close, 2=always close)
 -pvtc=n closing paren vertical tightness, see -bvtc for n.
 -sbvtc=n closing square bracket vertical tightness, see -bvtc for n.
 -ci=n   sets continuation indentation=n,  default is n=2 spaces
 -lp     line up parentheses, brackets, and non-BLOCK braces
 -sfs    add space before semicolon in for( ; ; )
 -aws    allow perltidy to add whitespace (default)
 -dws    delete all old non-essential whitespace
 -icb    indent closing brace of a code block
 -cti=n  closing indentation of paren, square bracket, or non-block brace:
         n=0 none, =1 align with opening, =2 one full indentation level
 -icp    equivalent to -cti=2
 -wls=s  want space left of tokens in string; i.e. -nwls='+ - * /'
 -wrs=s  want space right of tokens in string;
 -sts    put space before terminal semicolon of a statement
 -sak=s  put space between keywords given in s and '(';
 -nsak=s no space between keywords in s and '('; i.e. -nsak='my our local'

Line Break Control
 -fnl    freeze newlines; this disables all line break changes
            and disables the following switches:
 -anl    add newlines;  ok to introduce new line breaks
 -bbs    add blank line before subs and packages
 -bbc    add blank line before block comments
 -bbb    add blank line between major blocks
 -kbl=n  keep old blank lines? 0=no, 1=some, 2=all
 -mbl=n  maximum consecutive blank lines to output (default=1)
 -ce     cuddled else; use this style: '} else {'
 -dnl    delete old newlines (default)
 -l=n    maximum line length;  default n=80
 -bl     opening brace on new line
 -sbl    opening sub brace on new line.  value of -bl is used if not given.
 -bli    opening brace on new line and indented
 -bar    opening brace always on right, even for long clauses
 -vt=n   vertical tightness (requires -lp); n controls break after opening
         token: 0=never  1=no break if next line balanced   2=no break
 -vtc=n  vertical tightness of closing container; n controls if closing
         token starts new line: 0=always  1=not unless list  1=never
 -wba=s  want break after tokens in string; i.e. wba=': .'
 -wbb=s  want break before tokens in string

Following Old Breakpoints
 -kis    keep interior semicolons.  Allows multiple statements per line.
 -boc    break at old comma breaks: turns off all automatic list formatting
 -bol    break at old logical breakpoints: or, and, ||, && (default)
 -bok    break at old list keyword breakpoints such as map, sort (default)
 -bot    break at old conditional (ternary ?:) operator breakpoints (default)
 -boa    break at old attribute breakpoints
 -cab=n  break at commas after a comma-arrow (=>):
         n=0 break at all commas after =>
         n=1 stable: break unless this breaks an existing one-line container
         n=2 break only if a one-line container cannot be formed
         n=3 do not treat commas after => specially at all

Comment controls
 -ibc    indent block comments (default)
 -isbc   indent spaced block comments; may indent unless no leading space
 -msc=n  minimum desired spaces to side comment, default 4
 -fpsc=n fix position for side comments; default 0;
 -csc    add or update closing side comments after closing BLOCK brace
 -dcsc   delete closing side comments created by a -csc command
 -cscp=s change closing side comment prefix to be other than '## end'
 -cscl=s change closing side comment to apply to selected list of blocks
 -csci=n minimum number of lines needed to apply a -csc tag, default n=6
 -csct=n maximum number of columns of appended text, default n=20
 -cscw   causes warning if old side comment is overwritten with -csc

 -sbc    use 'static block comments' identified by leading '##' (default)
 -sbcp=s change static block comment identifier to be other than '##'
 -osbc   outdent static block comments

 -ssc    use 'static side comments' identified by leading '##' (default)
 -sscp=s change static side comment identifier to be other than '##'

Delete selected text
 -dac    delete all comments AND pod
 -dbc    delete block comments
 -dsc    delete side comments
 -dp     delete pod

Send selected text to a '.TEE' file
 -tac    tee all comments AND pod
 -tbc    tee block comments
 -tsc    tee side comments
 -tp     tee pod

Outdenting
 -olq    outdent long quoted strings (default)
 -olc    outdent a long block comment line
 -ola    outdent statement labels
 -okw    outdent control keywords (redo, next, last, goto, return)
 -okwl=s specify alternative keywords for -okw command

Other controls
 -mft=n  maximum fields per table; default n=40
 -x      do not format lines before hash-bang line (i.e., for VMS)
 -asc    allows perltidy to add a ';' when missing (default)
 -dsm    allows perltidy to delete an unnecessary ';'  (default)

Combinations of other parameters
 -gnu     attempt to follow GNU Coding Standards as applied to perl
 -mangle  remove as many newlines as possible (but keep comments and pods)
 -extrude  insert as many newlines as possible

Dump and die, debugging
 -dop    dump options used in this run to standard output and quit
 -ddf    dump default options to standard output and quit
 -dsn    dump all option short names to standard output and quit
 -dln    dump option long names to standard output and quit
 -dpro   dump whatever configuration file is in effect to standard output
 -dtt    dump all token types to standard output and quit

HTML
 -html write an html file (see 'man perl2web' for many options)
       Note: when -html is used, no indentation or formatting are done.
       Hint: try perltidy -html -css=mystyle.css filename.pl
       and edit mystyle.css to change the appearance of filename.html.
       -nnn gives line numbers
       -pre only writes out <pre>..</pre> code section
       -toc places a table of contents to subs at the top (default)
       -pod passes pod text through pod2html (default)
       -frm write html as a frame (3 files)
       -text=s extra extension for table of contents if -frm, default='toc'
       -sext=s extra extension for file content if -frm, default='src'

A prefix of "n" negates short form toggle switches, and a prefix of "no"
negates the long forms.  For example, -nasc means don't add missing
semicolons.

If you are unable to see this entire text, try "perltidy -h | more"
For more detailed information, and additional options, try "man perltidy",
or go to the perltidy home page at http://perltidy.sourceforge.net
EOF

}

sub process_this_file {

    my ( $truth, $beauty ) = @_;

    # loop to process each line of this file
    while ( my $line_of_tokens = $truth->get_line() ) {
        $beauty->write_line($line_of_tokens);
    }

    # finish up
    eval { $beauty->finish_formatting() };
    $truth->report_tokenization_errors();
}

sub check_syntax {

    # Use 'perl -c' to make sure that we did not create bad syntax
    # This is a very good independent check for programming errors
    #
    # Given names of the input and output files, ($istream, $ostream),
    # we do the following:
    # - check syntax of the input file
    # - if bad, all done (could be an incomplete code snippet)
    # - if infile syntax ok, then check syntax of the output file;
    #   - if outfile syntax bad, issue warning; this implies a code bug!
    # - set and return flag "infile_syntax_ok" : =-1 bad 0 unknown 1 good

    my ( $istream, $ostream, $logger_object, $rOpts ) = @_;
    my $infile_syntax_ok = 0;
    my $line_of_dashes   = '-' x 42 . "\n";

    my $flags = $rOpts->{'perl-syntax-check-flags'};

    # be sure we invoke perl with -c
    # note: perl will accept repeated flags like '-c -c'.  It is safest
    # to append another -c than try to find an interior bundled c, as
    # in -Tc, because such a 'c' might be in a quoted string, for example.
    if ( $flags !~ /(^-c|\s+-c)/ ) { $flags .= " -c" }

    # be sure we invoke perl with -x if requested
    # same comments about repeated parameters applies
    if ( $rOpts->{'look-for-hash-bang'} ) {
        if ( $flags !~ /(^-x|\s+-x)/ ) { $flags .= " -x" }
    }

    # this shouldn't happen unless a termporary file couldn't be made
    if ( $istream eq '-' ) {
        $logger_object->write_logfile_entry(
            "Cannot run perl -c on STDIN and STDOUT\n");
        return $infile_syntax_ok;
    }

    $logger_object->write_logfile_entry(
        "checking input file syntax with perl $flags\n");

    # Not all operating systems/shells support redirection of the standard
    # error output.
    my $error_redirection = ( $^O eq 'VMS' ) ? "" : '2>&1';

    my ( $istream_filename, $perl_output ) =
      do_syntax_check( $istream, $flags, $error_redirection );
    $logger_object->write_logfile_entry(
        "Input stream passed to Perl as file $istream_filename\n");
    $logger_object->write_logfile_entry($line_of_dashes);
    $logger_object->write_logfile_entry("$perl_output\n");

    if ( $perl_output =~ /syntax\s*OK/ ) {
        $infile_syntax_ok = 1;
        $logger_object->write_logfile_entry($line_of_dashes);
        $logger_object->write_logfile_entry(
            "checking output file syntax with perl $flags ...\n");
        my ( $ostream_filename, $perl_output ) =
          do_syntax_check( $ostream, $flags, $error_redirection );
        $logger_object->write_logfile_entry(
            "Output stream passed to Perl as file $ostream_filename\n");
        $logger_object->write_logfile_entry($line_of_dashes);
        $logger_object->write_logfile_entry("$perl_output\n");

        unless ( $perl_output =~ /syntax\s*OK/ ) {
            $logger_object->write_logfile_entry($line_of_dashes);
            $logger_object->warning(
"The output file has a syntax error when tested with perl $flags $ostream !\n"
            );
            $logger_object->warning(
                "This implies an error in perltidy; the file $ostream is bad\n"
            );
            $logger_object->report_definite_bug();

            # the perl version number will be helpful for diagnosing the problem
            $logger_object->write_logfile_entry(
                qx/perl -v $error_redirection/ . "\n" );
        }
    }
    else {

        # Only warn of perl -c syntax errors.  Other messages,
        # such as missing modules, are too common.  They can be
        # seen by running with perltidy -w
        $logger_object->complain("A syntax check using perl $flags\n");
        $logger_object->complain(
            "for the output in file $istream_filename gives:\n");
        $logger_object->complain($line_of_dashes);
        $logger_object->complain("$perl_output\n");
        $logger_object->complain($line_of_dashes);
        $infile_syntax_ok = -1;
        $logger_object->write_logfile_entry($line_of_dashes);
        $logger_object->write_logfile_entry(
"The output file will not be checked because of input file problems\n"
        );
    }
    return $infile_syntax_ok;
}

sub do_syntax_check {
    my ( $stream, $flags, $error_redirection ) = @_;

    # We need a named input file for executing perl
    my ( $stream_filename, $is_tmpfile ) = get_stream_as_named_file($stream);

    # TODO: Need to add name of file to log somewhere
    # otherwise Perl output is hard to read
    if ( !$stream_filename ) { return $stream_filename, "" }

    # We have to quote the filename in case it has unusual characters
    # or spaces.  Example: this filename #CM11.pm# gives trouble.
    my $quoted_stream_filename = '"' . $stream_filename . '"';

    # Under VMS something like -T will become -t (and an error) so we
    # will put quotes around the flags.  Double quotes seem to work on
    # Unix/Windows/VMS, but this may not work on all systems.  (Single
    # quotes do not work under Windows).  It could become necessary to
    # put double quotes around each flag, such as:  -"c"  -"T"
    # We may eventually need some system-dependent coding here.
    $flags = '"' . $flags . '"';

    # now wish for luck...
    my $msg = qx/perl $flags $quoted_stream_filename $error_redirection/;

    unlink $stream_filename if ($is_tmpfile);
    return $stream_filename, $msg;
}

#####################################################################
#
# This is a stripped down version of IO::Scalar
# Given a reference to a scalar, it supplies either:
# a getline method which reads lines (mode='r'), or
# a print method which reads lines (mode='w')
#
#####################################################################
package Perl::Tidy::IOScalar;
use Carp;

sub new {
    my ( $package, $rscalar, $mode ) = @_;
    my $ref = ref $rscalar;
    if ( $ref ne 'SCALAR' ) {
        confess <<EOM;
------------------------------------------------------------------------
expecting ref to SCALAR but got ref to ($ref); trace follows:
------------------------------------------------------------------------
EOM

    }
    if ( $mode eq 'w' ) {
        $$rscalar = "";
        return bless [ $rscalar, $mode ], $package;
    }
    elsif ( $mode eq 'r' ) {

        # Convert a scalar to an array.
        # This avoids looking for "\n" on each call to getline
        #
        # NOTES: The -1 count is needed to avoid loss of trailing blank lines
        # (which might be important in a DATA section).
        my @array;
        if ( $rscalar && ${$rscalar} ) {
            @array = map { $_ .= "\n" } split /\n/, ${$rscalar}, -1;

            # remove possible extra blank line introduced with split
            if ( @array && $array[-1] eq "\n" ) { pop @array }
        }
        my $i_next = 0;
        return bless [ \@array, $mode, $i_next ], $package;
    }
    else {
        confess <<EOM;
------------------------------------------------------------------------
expecting mode = 'r' or 'w' but got mode ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
}

sub getline {
    my $self = shift;
    my $mode = $self->[1];
    if ( $mode ne 'r' ) {
        confess <<EOM;
------------------------------------------------------------------------
getline call requires mode = 'r' but mode = ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
    my $i = $self->[2]++;
    return $self->[0]->[$i];
}

sub print {
    my $self = shift;
    my $mode = $self->[1];
    if ( $mode ne 'w' ) {
        confess <<EOM;
------------------------------------------------------------------------
print call requires mode = 'w' but mode = ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
    ${ $self->[0] } .= $_[0];
}
sub close { return }

#####################################################################
#
# This is a stripped down version of IO::ScalarArray
# Given a reference to an array, it supplies either:
# a getline method which reads lines (mode='r'), or
# a print method which reads lines (mode='w')
#
# NOTE: this routine assumes that that there aren't any embedded
# newlines within any of the array elements.  There are no checks
# for that.
#
#####################################################################
package Perl::Tidy::IOScalarArray;
use Carp;

sub new {
    my ( $package, $rarray, $mode ) = @_;
    my $ref = ref $rarray;
    if ( $ref ne 'ARRAY' ) {
        confess <<EOM;
------------------------------------------------------------------------
expecting ref to ARRAY but got ref to ($ref); trace follows:
------------------------------------------------------------------------
EOM

    }
    if ( $mode eq 'w' ) {
        @$rarray = ();
        return bless [ $rarray, $mode ], $package;
    }
    elsif ( $mode eq 'r' ) {
        my $i_next = 0;
        return bless [ $rarray, $mode, $i_next ], $package;
    }
    else {
        confess <<EOM;
------------------------------------------------------------------------
expecting mode = 'r' or 'w' but got mode ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
}

sub getline {
    my $self = shift;
    my $mode = $self->[1];
    if ( $mode ne 'r' ) {
        confess <<EOM;
------------------------------------------------------------------------
getline requires mode = 'r' but mode = ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
    my $i = $self->[2]++;
    return $self->[0]->[$i];
}

sub print {
    my $self = shift;
    my $mode = $self->[1];
    if ( $mode ne 'w' ) {
        confess <<EOM;
------------------------------------------------------------------------
print requires mode = 'w' but mode = ($mode); trace follows:
------------------------------------------------------------------------
EOM
    }
    push @{ $self->[0] }, $_[0];
}
sub close { return }

#####################################################################
#
# the Perl::Tidy::LineSource class supplies an object with a 'get_line()' method
# which returns the next line to be parsed
#
#####################################################################

package Perl::Tidy::LineSource;

sub new {

    my ( $class, $input_file, $rOpts, $rpending_logfile_message ) = @_;

    my $input_line_ending;
    if ( $rOpts->{'preserve-line-endings'} ) {
        $input_line_ending = Perl::Tidy::find_input_line_ending($input_file);
    }

    ( my $fh, $input_file ) = Perl::Tidy::streamhandle( $input_file, 'r' );
    return undef unless $fh;

    # in order to check output syntax when standard output is used,
    # or when it is an object, we have to make a copy of the file
    if ( ( $input_file eq '-' || ref $input_file ) && $rOpts->{'check-syntax'} )
    {

        # Turning off syntax check when input output is used.
        # The reason is that temporary files cause problems on
        # on many systems.
        $rOpts->{'check-syntax'} = 0;

        $$rpending_logfile_message .= <<EOM;
Note: --syntax check will be skipped because standard input is used
EOM

    }

    return bless {
        _fh                => $fh,
        _filename          => $input_file,
        _input_line_ending => $input_line_ending,
        _rinput_buffer     => [],
        _started           => 0,
    }, $class;
}

sub close_input_file {
    my $self = shift;

    # Only close physical files, not STDIN and other objects
    my $filename = $self->{_filename};
    if ( $filename ne '-' && !ref $filename ) {
        eval { $self->{_fh}->close() };
    }
}

sub get_line {
    my $self          = shift;
    my $line          = undef;
    my $fh            = $self->{_fh};
    my $rinput_buffer = $self->{_rinput_buffer};

    if ( scalar(@$rinput_buffer) ) {
        $line = shift @$rinput_buffer;
    }
    else {
        $line = $fh->getline();

        # patch to read raw mac files under unix, dos
        # see if the first line has embedded \r's
        if ( $line && !$self->{_started} ) {
            if ( $line =~ /[\015][^\015\012]/ ) {

                # found one -- break the line up and store in a buffer
                @$rinput_buffer = map { $_ . "\n" } split /\015/, $line;
                my $count = @$rinput_buffer;
                $line = shift @$rinput_buffer;
            }
            $self->{_started}++;
        }
    }
    return $line;
}

#####################################################################
#
# the Perl::Tidy::LineSink class supplies a write_line method for
# actual file writing
#
#####################################################################

package Perl::Tidy::LineSink;

sub new {

    my ( $class, $output_file, $tee_file, $line_separator, $rOpts,
        $rpending_logfile_message, $binmode )
      = @_;
    my $fh     = undef;
    my $fh_tee = undef;

    my $output_file_open = 0;

    if ( $rOpts->{'format'} eq 'tidy' ) {
        ( $fh, $output_file ) = Perl::Tidy::streamhandle( $output_file, 'w' );
        unless ($fh) { Perl::Tidy::Die "Cannot write to output stream\n"; }
        $output_file_open = 1;
        if ($binmode) {
            if ( ref($fh) eq 'IO::File' ) {
                binmode $fh;
            }
            if ( $output_file eq '-' ) { binmode STDOUT }
        }
    }

    # in order to check output syntax when standard output is used,
    # or when it is an object, we have to make a copy of the file
    if ( $output_file eq '-' || ref $output_file ) {
        if ( $rOpts->{'check-syntax'} ) {

            # Turning off syntax check when standard output is used.
            # The reason is that temporary files cause problems on
            # on many systems.
            $rOpts->{'check-syntax'} = 0;
            $$rpending_logfile_message .= <<EOM;
Note: --syntax check will be skipped because standard output is used
EOM

        }
    }

    bless {
        _fh               => $fh,
        _fh_tee           => $fh_tee,
        _output_file      => $output_file,
        _output_file_open => $output_file_open,
        _tee_flag         => 0,
        _tee_file         => $tee_file,
        _tee_file_opened  => 0,
        _line_separator   => $line_separator,
        _binmode          => $binmode,
    }, $class;
}

sub write_line {

    my $self = shift;
    my $fh   = $self->{_fh};

    my $output_file_open = $self->{_output_file_open};
    chomp $_[0];
    $_[0] .= $self->{_line_separator};

    $fh->print( $_[0] ) if ( $self->{_output_file_open} );

    if ( $self->{_tee_flag} ) {
        unless ( $self->{_tee_file_opened} ) { $self->really_open_tee_file() }
        my $fh_tee = $self->{_fh_tee};
        print $fh_tee $_[0];
    }
}

sub tee_on {
    my $self = shift;
    $self->{_tee_flag} = 1;
}

sub tee_off {
    my $self = shift;
    $self->{_tee_flag} = 0;
}

sub really_open_tee_file {
    my $self     = shift;
    my $tee_file = $self->{_tee_file};
    my $fh_tee;
    $fh_tee = IO::File->new(">$tee_file")
      or Perl::Tidy::Die("couldn't open TEE file $tee_file: $!\n");
    binmode $fh_tee if $self->{_binmode};
    $self->{_tee_file_opened} = 1;
    $self->{_fh_tee}          = $fh_tee;
}

sub close_output_file {
    my $self = shift;

    # Only close physical files, not STDOUT and other objects
    my $output_file = $self->{_output_file};
    if ( $output_file ne '-' && !ref $output_file ) {
        eval { $self->{_fh}->close() } if $self->{_output_file_open};
    }
    $self->close_tee_file();
}

sub close_tee_file {
    my $self = shift;

    # Only close physical files, not STDOUT and other objects
    if ( $self->{_tee_file_opened} ) {
        my $tee_file = $self->{_tee_file};
        if ( $tee_file ne '-' && !ref $tee_file ) {
            eval { $self->{_fh_tee}->close() };
            $self->{_tee_file_opened} = 0;
        }
    }
}

#####################################################################
#
# The Perl::Tidy::Diagnostics class writes the DIAGNOSTICS file, which is
# useful for program development.
#
# Only one such file is created regardless of the number of input
# files processed.  This allows the results of processing many files
# to be summarized in a single file.
#
#####################################################################

package Perl::Tidy::Diagnostics;

sub new {

    my $class = shift;
    bless {
        _write_diagnostics_count => 0,
        _last_diagnostic_file    => "",
        _input_file              => "",
        _fh                      => undef,
    }, $class;
}

sub set_input_file {
    my $self = shift;
    $self->{_input_file} = $_[0];
}

# This is a diagnostic routine which is useful for program development.
# Output from debug messages go to a file named DIAGNOSTICS, where
# they are labeled by file and line.  This allows many files to be
# scanned at once for some particular condition of interest.
sub write_diagnostics {
    my $self = shift;

    unless ( $self->{_write_diagnostics_count} ) {
        open DIAGNOSTICS, ">DIAGNOSTICS"
          or death("couldn't open DIAGNOSTICS: $!\n");
    }

    my $last_diagnostic_file = $self->{_last_diagnostic_file};
    my $input_file           = $self->{_input_file};
    if ( $last_diagnostic_file ne $input_file ) {
        print DIAGNOSTICS "\nFILE:$input_file\n";
    }
    $self->{_last_diagnostic_file} = $input_file;
    my $input_line_number = Perl::Tidy::Tokenizer::get_input_line_number();
    print DIAGNOSTICS "$input_line_number:\t@_";
    $self->{_write_diagnostics_count}++;
}

#####################################################################
#
# The Perl::Tidy::Logger class writes the .LOG and .ERR files
#
#####################################################################

package Perl::Tidy::Logger;

sub new {
    my $class = shift;
    my $fh;
    my ( $rOpts, $log_file, $warning_file, $fh_stderr, $saw_extrude, ) = @_;

    my $fh_warnings = $rOpts->{'standard-error-output'} ? $fh_stderr : undef;

    # remove any old error output file if we might write a new one
    unless ( $fh_warnings || ref($warning_file) ) {
        if ( -e $warning_file ) { unlink($warning_file) }
    }

    bless {
        _log_file                      => $log_file,
        _rOpts                         => $rOpts,
        _fh_warnings                   => $fh_warnings,
        _last_input_line_written       => 0,
        _at_end_of_file                => 0,
        _use_prefix                    => 1,
        _block_log_output              => 0,
        _line_of_tokens                => undef,
        _output_line_number            => undef,
        _wrote_line_information_string => 0,
        _wrote_column_headings         => 0,
        _warning_file                  => $warning_file,
        _warning_count                 => 0,
        _complaint_count               => 0,
        _saw_code_bug    => -1,             # -1=no 0=maybe 1=for sure
        _saw_brace_error => 0,
        _saw_extrude     => $saw_extrude,
        _output_array    => [],
    }, $class;
}

sub get_warning_count {
    my $self = shift;
    return $self->{_warning_count};
}

sub get_use_prefix {
    my $self = shift;
    return $self->{_use_prefix};
}

sub block_log_output {
    my $self = shift;
    $self->{_block_log_output} = 1;
}

sub unblock_log_output {
    my $self = shift;
    $self->{_block_log_output} = 0;
}

sub interrupt_logfile {
    my $self = shift;
    $self->{_use_prefix} = 0;
    $self->warning("\n");
    $self->write_logfile_entry( '#' x 24 . "  WARNING  " . '#' x 25 . "\n" );
}

sub resume_logfile {
    my $self = shift;
    $self->write_logfile_entry( '#' x 60 . "\n" );
    $self->{_use_prefix} = 1;
}

sub we_are_at_the_last_line {
    my $self = shift;
    unless ( $self->{_wrote_line_information_string} ) {
        $self->write_logfile_entry("Last line\n\n");
    }
    $self->{_at_end_of_file} = 1;
}

# record some stuff in case we go down in flames
sub black_box {
    my $self = shift;
    my ( $line_of_tokens, $output_line_number ) = @_;
    my $input_line        = $line_of_tokens->{_line_text};
    my $input_line_number = $line_of_tokens->{_line_number};

    # save line information in case we have to write a logfile message
    $self->{_line_of_tokens}                = $line_of_tokens;
    $self->{_output_line_number}            = $output_line_number;
    $self->{_wrote_line_information_string} = 0;

    my $last_input_line_written = $self->{_last_input_line_written};
    my $rOpts                   = $self->{_rOpts};
    if (
        (
            ( $input_line_number - $last_input_line_written ) >=
            $rOpts->{'logfile-gap'}
        )
        || ( $input_line =~ /^\s*(sub|package)\s+(\w+)/ )
      )
    {
        my $rlevels                      = $line_of_tokens->{_rlevels};
        my $structural_indentation_level = $$rlevels[0];
        $self->{_last_input_line_written} = $input_line_number;
        ( my $out_str = $input_line ) =~ s/^\s*//;
        chomp $out_str;

        $out_str = ( '.' x $structural_indentation_level ) . $out_str;

        if ( length($out_str) > 35 ) {
            $out_str = substr( $out_str, 0, 35 ) . " ....";
        }
        $self->logfile_output( "", "$out_str\n" );
    }
}

sub write_logfile_entry {
    my $self = shift;

    # add leading >>> to avoid confusing error mesages and code
    $self->logfile_output( ">>>", "@_" );
}

sub write_column_headings {
    my $self = shift;

    $self->{_wrote_column_headings} = 1;
    my $routput_array = $self->{_output_array};
    push @{$routput_array}, <<EOM;
The nesting depths in the table below are at the start of the lines.
The indicated output line numbers are not always exact.
ci = levels of continuation indentation; bk = 1 if in BLOCK, 0 if not.

in:out indent c b  nesting   code + messages; (messages begin with >>>)
lines  levels i k            (code begins with one '.' per indent level)
------  ----- - - --------   -------------------------------------------
EOM
}

sub make_line_information_string {

    # make columns of information when a logfile message needs to go out
    my $self                    = shift;
    my $line_of_tokens          = $self->{_line_of_tokens};
    my $input_line_number       = $line_of_tokens->{_line_number};
    my $line_information_string = "";
    if ($input_line_number) {

        my $output_line_number   = $self->{_output_line_number};
        my $brace_depth          = $line_of_tokens->{_curly_brace_depth};
        my $paren_depth          = $line_of_tokens->{_paren_depth};
        my $square_bracket_depth = $line_of_tokens->{_square_bracket_depth};
        my $guessed_indentation_level =
          $line_of_tokens->{_guessed_indentation_level};
        my $rlevels         = $line_of_tokens->{_rlevels};
        my $rnesting_tokens = $line_of_tokens->{_rnesting_tokens};
        my $rci_levels      = $line_of_tokens->{_rci_levels};
        my $rnesting_blocks = $line_of_tokens->{_rnesting_blocks};

        my $structural_indentation_level = $$rlevels[0];

        $self->write_column_headings() unless $self->{_wrote_column_headings};

        # keep logfile columns aligned for scripts up to 999 lines;
        # for longer scripts it doesn't really matter
        my $extra_space = "";
        $extra_space .=
            ( $input_line_number < 10 )  ? "  "
          : ( $input_line_number < 100 ) ? " "
          :                                "";
        $extra_space .=
            ( $output_line_number < 10 )  ? "  "
          : ( $output_line_number < 100 ) ? " "
          :                                 "";

        # there are 2 possible nesting strings:
        # the original which looks like this:  (0 [1 {2
        # the new one, which looks like this:  {{[
        # the new one is easier to read, and shows the order, but
        # could be arbitrarily long, so we use it unless it is too long
        my $nesting_string =
          "($paren_depth [$square_bracket_depth {$brace_depth";
        my $nesting_string_new = $$rnesting_tokens[0];

        my $ci_level = $$rci_levels[0];
        if ( $ci_level > 9 ) { $ci_level = '*' }
        my $bk = ( $$rnesting_blocks[0] =~ /1$/ ) ? '1' : '0';

        if ( length($nesting_string_new) <= 8 ) {
            $nesting_string =
              $nesting_string_new . " " x ( 8 - length($nesting_string_new) );
        }
        $line_information_string =
"L$input_line_number:$output_line_number$extra_space i$guessed_indentation_level:$structural_indentation_level $ci_level $bk $nesting_string";
    }
    return $line_information_string;
}

sub logfile_output {
    my $self = shift;
    my ( $prompt, $msg ) = @_;
    return if ( $self->{_block_log_output} );

    my $routput_array = $self->{_output_array};
    if ( $self->{_at_end_of_file} || !$self->{_use_prefix} ) {
        push @{$routput_array}, "$msg";
    }
    else {
        my $line_information_string = $self->make_line_information_string();
        $self->{_wrote_line_information_string} = 1;

        if ($line_information_string) {
            push @{$routput_array}, "$line_information_string   $prompt$msg";
        }
        else {
            push @{$routput_array}, "$msg";
        }
    }
}

sub get_saw_brace_error {
    my $self = shift;
    return $self->{_saw_brace_error};
}

sub increment_brace_error {
    my $self = shift;
    $self->{_saw_brace_error}++;
}

sub brace_warning {
    my $self = shift;
    use constant BRACE_WARNING_LIMIT => 10;
    my $saw_brace_error = $self->{_saw_brace_error};

    if ( $saw_brace_error < BRACE_WARNING_LIMIT ) {
        $self->warning(@_);
    }
    $saw_brace_error++;
    $self->{_saw_brace_error} = $saw_brace_error;

    if ( $saw_brace_error == BRACE_WARNING_LIMIT ) {
        $self->warning("No further warnings of this type will be given\n");
    }
}

sub complain {

    # handle non-critical warning messages based on input flag
    my $self  = shift;
    my $rOpts = $self->{_rOpts};

    # these appear in .ERR output only if -w flag is used
    if ( $rOpts->{'warning-output'} ) {
        $self->warning(@_);
    }

    # otherwise, they go to the .LOG file
    else {
        $self->{_complaint_count}++;
        $self->write_logfile_entry(@_);
    }
}

sub warning {

    # report errors to .ERR file (or stdout)
    my $self = shift;
    use constant WARNING_LIMIT => 50;

    my $rOpts = $self->{_rOpts};
    unless ( $rOpts->{'quiet'} ) {

        my $warning_count = $self->{_warning_count};
        my $fh_warnings   = $self->{_fh_warnings};
        if ( !$fh_warnings ) {
            my $warning_file = $self->{_warning_file};
            ( $fh_warnings, my $filename ) =
              Perl::Tidy::streamhandle( $warning_file, 'w' );
            $fh_warnings or Perl::Tidy::Die("couldn't open $filename $!\n");
            Perl::Tidy::Warn "## Please see file $filename\n"
              unless ref($warning_file);
            $self->{_fh_warnings} = $fh_warnings;
        }

        if ( $warning_count < WARNING_LIMIT ) {
            if ( $self->get_use_prefix() > 0 ) {
                my $input_line_number =
                  Perl::Tidy::Tokenizer::get_input_line_number();
                $fh_warnings->print("$input_line_number:\t@_");
                $self->write_logfile_entry("WARNING: @_");
            }
            else {
                $fh_warnings->print(@_);
                $self->write_logfile_entry(@_);
            }
        }
        $warning_count++;
        $self->{_warning_count} = $warning_count;

        if ( $warning_count == WARNING_LIMIT ) {
            $fh_warnings->print("No further warnings will be given\n");
        }
    }
}

# programming bug codes:
#   -1 = no bug
#    0 = maybe, not sure.
#    1 = definitely
sub report_possible_bug {
    my $self         = shift;
    my $saw_code_bug = $self->{_saw_code_bug};
    $self->{_saw_code_bug} = ( $saw_code_bug < 0 ) ? 0 : $saw_code_bug;
}

sub report_definite_bug {
    my $self = shift;
    $self->{_saw_code_bug} = 1;
}

sub ask_user_for_bug_report {
    my $self = shift;

    my ( $infile_syntax_ok, $formatter ) = @_;
    my $saw_code_bug = $self->{_saw_code_bug};
    if ( ( $saw_code_bug == 0 ) && ( $infile_syntax_ok == 1 ) ) {
        $self->warning(<<EOM);

You may have encountered a code bug in perltidy.  If you think so, and
the problem is not listed in the BUGS file at
http://perltidy.sourceforge.net, please report it so that it can be
corrected.  Include the smallest possible script which has the problem,
along with the .LOG file. See the manual pages for contact information.
Thank you!
EOM

    }
    elsif ( $saw_code_bug == 1 ) {
        if ( $self->{_saw_extrude} ) {
            $self->warning(<<EOM);

You may have encountered a bug in perltidy.  However, since you are using the
-extrude option, the problem may be with perl or one of its modules, which have
occasional problems with this type of file.  If you believe that the
problem is with perltidy, and the problem is not listed in the BUGS file at
http://perltidy.sourceforge.net, please report it so that it can be corrected.
Include the smallest possible script which has the problem, along with the .LOG
file. See the manual pages for contact information.
Thank you!
EOM
        }
        else {
            $self->warning(<<EOM);

Oops, you seem to have encountered a bug in perltidy.  Please check the
BUGS file at http://perltidy.sourceforge.net.  If the problem is not
listed there, please report it so that it can be corrected.  Include the
smallest possible script which produces this message, along with the
.LOG file if appropriate.  See the manual pages for contact information.
Your efforts are appreciated.
Thank you!
EOM
            my $added_semicolon_count = 0;
            eval {
                $added_semicolon_count =
                  $formatter->get_added_semicolon_count();
            };
            if ( $added_semicolon_count > 0 ) {
                $self->warning(<<EOM);

The log file shows that perltidy added $added_semicolon_count semicolons.
Please rerun with -nasc to see if that is the cause of the syntax error.  Even
if that is the problem, please report it so that it can be fixed.
EOM

            }
        }
    }
}

sub finish {

    # called after all formatting to summarize errors
    my $self = shift;
    my ( $infile_syntax_ok, $formatter ) = @_;

    my $rOpts         = $self->{_rOpts};
    my $warning_count = $self->{_warning_count};
    my $saw_code_bug  = $self->{_saw_code_bug};

    my $save_logfile =
         ( $saw_code_bug == 0 && $infile_syntax_ok == 1 )
      || $saw_code_bug == 1
      || $rOpts->{'logfile'};
    my $log_file = $self->{_log_file};
    if ($warning_count) {
        if ($save_logfile) {
            $self->block_log_output();    # avoid echoing this to the logfile
            $self->warning(
                "The logfile $log_file may contain useful information\n");
            $self->unblock_log_output();
        }

        if ( $self->{_complaint_count} > 0 ) {
            $self->warning(
"To see $self->{_complaint_count} non-critical warnings rerun with -w\n"
            );
        }

        if ( $self->{_saw_brace_error}
            && ( $rOpts->{'logfile-gap'} > 1 || !$save_logfile ) )
        {
            $self->warning("To save a full .LOG file rerun with -g\n");
        }
    }
    $self->ask_user_for_bug_report( $infile_syntax_ok, $formatter );

    if ($save_logfile) {
        my $log_file = $self->{_log_file};
        my ( $fh, $filename ) = Perl::Tidy::streamhandle( $log_file, 'w' );
        if ($fh) {
            my $routput_array = $self->{_output_array};
            foreach ( @{$routput_array} ) { $fh->print($_) }
            if ( $log_file ne '-' && !ref $log_file ) {
                eval { $fh->close() };
            }
        }
    }
}

#####################################################################
#
# The Perl::Tidy::DevNull class supplies a dummy print method
#
#####################################################################

package Perl::Tidy::DevNull;
sub new { return bless {}, $_[0] }
sub print { return }
sub close { return }

#####################################################################
#
# The Perl::Tidy::HtmlWriter class writes a copy of the input stream in html
#
#####################################################################

package Perl::Tidy::HtmlWriter;

use File::Basename;

# class variables
use vars qw{
  %html_color
  %html_bold
  %html_italic
  %token_short_names
  %short_to_long_names
  $rOpts
  $css_filename
  $css_linkname
  $missing_html_entities
};

# replace unsafe characters with HTML entity representation if HTML::Entities
# is available
{ eval "use HTML::Entities"; $missing_html_entities = $@; }

sub new {

    my ( $class, $input_file, $html_file, $extension, $html_toc_extension,
        $html_src_extension )
      = @_;

    my $html_file_opened = 0;
    my $html_fh;
    ( $html_fh, my $html_filename ) =
      Perl::Tidy::streamhandle( $html_file, 'w' );
    unless ($html_fh) {
        Perl::Tidy::Warn("can't open $html_file: $!\n");
        return undef;
    }
    $html_file_opened = 1;

    if ( !$input_file || $input_file eq '-' || ref($input_file) ) {
        $input_file = "NONAME";
    }

    # write the table of contents to a string
    my $toc_string;
    my $html_toc_fh = Perl::Tidy::IOScalar->new( \$toc_string, 'w' );

    my $html_pre_fh;
    my @pre_string_stack;
    if ( $rOpts->{'html-pre-only'} ) {

        # pre section goes directly to the output stream
        $html_pre_fh = $html_fh;
        $html_pre_fh->print( <<"PRE_END");
<pre>
PRE_END
    }
    else {

        # pre section go out to a temporary string
        my $pre_string;
        $html_pre_fh = Perl::Tidy::IOScalar->new( \$pre_string, 'w' );
        push @pre_string_stack, \$pre_string;
    }

    # pod text gets diverted if the 'pod2html' is used
    my $html_pod_fh;
    my $pod_string;
    if ( $rOpts->{'pod2html'} ) {
        if ( $rOpts->{'html-pre-only'} ) {
            undef $rOpts->{'pod2html'};
        }
        else {
            eval "use Pod::Html";
            if ($@) {
                Perl::Tidy::Warn
"unable to find Pod::Html; cannot use pod2html\n-npod disables this message\n";
                undef $rOpts->{'pod2html'};
            }
            else {
                $html_pod_fh = Perl::Tidy::IOScalar->new( \$pod_string, 'w' );
            }
        }
    }

    my $toc_filename;
    my $src_filename;
    if ( $rOpts->{'frames'} ) {
        unless ($extension) {
            Perl::Tidy::Warn
"cannot use frames without a specified output extension; ignoring -frm\n";
            undef $rOpts->{'frames'};
        }
        else {
            $toc_filename = $input_file . $html_toc_extension . $extension;
            $src_filename = $input_file . $html_src_extension . $extension;
        }
    }

    # ----------------------------------------------------------
    # Output is now directed as follows:
    # html_toc_fh <-- table of contents items
    # html_pre_fh <-- the <pre> section of formatted code, except:
    # html_pod_fh <-- pod goes here with the pod2html option
    # ----------------------------------------------------------

    my $title = $rOpts->{'title'};
    unless ($title) {
        ( $title, my $path ) = fileparse($input_file);
    }
    my $toc_item_count = 0;
    my $in_toc_package = "";
    my $last_level     = 0;
    bless {
        _input_file        => $input_file,          # name of input file
        _title             => $title,               # title, unescaped
        _html_file         => $html_file,           # name of .html output file
        _toc_filename      => $toc_filename,        # for frames option
        _src_filename      => $src_filename,        # for frames option
        _html_file_opened  => $html_file_opened,    # a flag
        _html_fh           => $html_fh,             # the output stream
        _html_pre_fh       => $html_pre_fh,         # pre section goes here
        _rpre_string_stack => \@pre_string_stack,   # stack of pre sections
        _html_pod_fh       => $html_pod_fh,         # pod goes here if pod2html
        _rpod_string       => \$pod_string,         # string holding pod
        _pod_cut_count     => 0,                    # how many =cut's?
        _html_toc_fh       => $html_toc_fh,         # fh for table of contents
        _rtoc_string       => \$toc_string,         # string holding toc
        _rtoc_item_count   => \$toc_item_count,     # how many toc items
        _rin_toc_package   => \$in_toc_package,     # package name
        _rtoc_name_count   => {},                   # hash to track unique names
        _rpackage_stack    => [],                   # stack to check for package
                                                    # name changes
        _rlast_level       => \$last_level,         # brace indentation level
    }, $class;
}

sub add_toc_item {

    # Add an item to the html table of contents.
    # This is called even if no table of contents is written,
    # because we still want to put the anchors in the <pre> text.
    # We are given an anchor name and its type; types are:
    #      'package', 'sub', '__END__', '__DATA__', 'EOF'
    # There must be an 'EOF' call at the end to wrap things up.
    my $self = shift;
    my ( $name, $type ) = @_;
    my $html_toc_fh     = $self->{_html_toc_fh};
    my $html_pre_fh     = $self->{_html_pre_fh};
    my $rtoc_name_count = $self->{_rtoc_name_count};
    my $rtoc_item_count = $self->{_rtoc_item_count};
    my $rlast_level     = $self->{_rlast_level};
    my $rin_toc_package = $self->{_rin_toc_package};
    my $rpackage_stack  = $self->{_rpackage_stack};

    # packages contain sublists of subs, so to avoid errors all package
    # items are written and finished with the following routines
    my $end_package_list = sub {
        if ($$rin_toc_package) {
            $html_toc_fh->print("</ul>\n</li>\n");
            $$rin_toc_package = "";
        }
    };

    my $start_package_list = sub {
        my ( $unique_name, $package ) = @_;
        if ($$rin_toc_package) { $end_package_list->() }
        $html_toc_fh->print(<<EOM);
<li><a href=\"#$unique_name\">package $package</a>
<ul>
EOM
        $$rin_toc_package = $package;
    };

    # start the table of contents on the first item
    unless ($$rtoc_item_count) {

        # but just quit if we hit EOF without any other entries
        # in this case, there will be no toc
        return if ( $type eq 'EOF' );
        $html_toc_fh->print( <<"TOC_END");
<!-- BEGIN CODE INDEX --><a name="code-index"></a>
<ul>
TOC_END
    }
    $$rtoc_item_count++;

    # make a unique anchor name for this location:
    #   - packages get a 'package-' prefix
    #   - subs use their names
    my $unique_name = $name;
    if ( $type eq 'package' ) { $unique_name = "package-$name" }

    # append '-1', '-2', etc if necessary to make unique; this will
    # be unique because subs and packages cannot have a '-'
    if ( my $count = $rtoc_name_count->{ lc $unique_name }++ ) {
        $unique_name .= "-$count";
    }

    #   - all names get terminal '-' if pod2html is used, to avoid
    #     conflicts with anchor names created by pod2html
    if ( $rOpts->{'pod2html'} ) { $unique_name .= '-' }

    # start/stop lists of subs
    if ( $type eq 'sub' ) {
        my $package = $rpackage_stack->[$$rlast_level];
        unless ($package) { $package = 'main' }

        # if we're already in a package/sub list, be sure its the right
        # package or else close it
        if ( $$rin_toc_package && $$rin_toc_package ne $package ) {
            $end_package_list->();
        }

        # start a package/sub list if necessary
        unless ($$rin_toc_package) {
            $start_package_list->( $unique_name, $package );
        }
    }

    # now write an entry in the toc for this item
    if ( $type eq 'package' ) {
        $start_package_list->( $unique_name, $name );
    }
    elsif ( $type eq 'sub' ) {
        $html_toc_fh->print("<li><a href=\"#$unique_name\">$name</a></li>\n");
    }
    else {
        $end_package_list->();
        $html_toc_fh->print("<li><a href=\"#$unique_name\">$name</a></li>\n");
    }

    # write the anchor in the <pre> section
    $html_pre_fh->print("<a name=\"$unique_name\"></a>");

    # end the table of contents, if any, on the end of file
    if ( $type eq 'EOF' ) {
        $html_toc_fh->print( <<"TOC_END");
</ul>
<!-- END CODE INDEX -->
TOC_END
    }
}

BEGIN {

    # This is the official list of tokens which may be identified by the
    # user.  Long names are used as getopt keys.  Short names are
    # convenient short abbreviations for specifying input.  Short names
    # somewhat resemble token type characters, but are often different
    # because they may only be alphanumeric, to allow command line
    # input.  Also, note that because of case insensitivity of html,
    # this table must be in a single case only (I've chosen to use all
    # lower case).
    # When adding NEW_TOKENS: update this hash table
    # short names => long names
    %short_to_long_names = (
        'n'  => 'numeric',
        'p'  => 'paren',
        'q'  => 'quote',
        's'  => 'structure',
        'c'  => 'comment',
        'v'  => 'v-string',
        'cm' => 'comma',
        'w'  => 'bareword',
        'co' => 'colon',
        'pu' => 'punctuation',
        'i'  => 'identifier',
        'j'  => 'label',
        'h'  => 'here-doc-target',
        'hh' => 'here-doc-text',
        'k'  => 'keyword',
        'sc' => 'semicolon',
        'm'  => 'subroutine',
        'pd' => 'pod-text',
    );

    # Now we have to map actual token types into one of the above short
    # names; any token types not mapped will get 'punctuation'
    # properties.

    # The values of this hash table correspond to the keys of the
    # previous hash table.
    # The keys of this hash table are token types and can be seen
    # by running with --dump-token-types (-dtt).

    # When adding NEW_TOKENS: update this hash table
    # $type => $short_name
    %token_short_names = (
        '#'  => 'c',
        'n'  => 'n',
        'v'  => 'v',
        'k'  => 'k',
        'F'  => 'k',
        'Q'  => 'q',
        'q'  => 'q',
        'J'  => 'j',
        'j'  => 'j',
        'h'  => 'h',
        'H'  => 'hh',
        'w'  => 'w',
        ','  => 'cm',
        '=>' => 'cm',
        ';'  => 'sc',
        ':'  => 'co',
        'f'  => 'sc',
        '('  => 'p',
        ')'  => 'p',
        'M'  => 'm',
        'P'  => 'pd',
        'A'  => 'co',
    );

    # These token types will all be called identifiers for now
    # FIXME: could separate user defined modules as separate type
    my @identifier = qw" i t U C Y Z G :: CORE::";
    @token_short_names{@identifier} = ('i') x scalar(@identifier);

    # These token types will be called 'structure'
    my @structure = qw" { } ";
    @token_short_names{@structure} = ('s') x scalar(@structure);

    # OLD NOTES: save for reference
    # Any of these could be added later if it would be useful.
    # For now, they will by default become punctuation
    #    my @list = qw" L R [ ] ";
    #    @token_long_names{@list} = ('non-structure') x scalar(@list);
    #
    #    my @list = qw"
    #      / /= * *= ** **= + += - -= % %= = ++ -- << <<= >> >>= pp p m mm
    #      ";
    #    @token_long_names{@list} = ('math') x scalar(@list);
    #
    #    my @list = qw" & &= ~ ~= ^ ^= | |= ";
    #    @token_long_names{@list} = ('bit') x scalar(@list);
    #
    #    my @list = qw" == != < > <= <=> ";
    #    @token_long_names{@list} = ('numerical-comparison') x scalar(@list);
    #
    #    my @list = qw" && || ! &&= ||= //= ";
    #    @token_long_names{@list} = ('logical') x scalar(@list);
    #
    #    my @list = qw" . .= =~ !~ x x= ";
    #    @token_long_names{@list} = ('string-operators') x scalar(@list);
    #
    #    # Incomplete..
    #    my @list = qw" .. -> <> ... \ ? ";
    #    @token_long_names{@list} = ('misc-operators') x scalar(@list);

}

sub make_getopt_long_names {
    my $class = shift;
    my ($rgetopt_names) = @_;
    while ( my ( $short_name, $name ) = each %short_to_long_names ) {
        push @$rgetopt_names, "html-color-$name=s";
        push @$rgetopt_names, "html-italic-$name!";
        push @$rgetopt_names, "html-bold-$name!";
    }
    push @$rgetopt_names, "html-color-background=s";
    push @$rgetopt_names, "html-linked-style-sheet=s";
    push @$rgetopt_names, "nohtml-style-sheets";
    push @$rgetopt_names, "html-pre-only";
    push @$rgetopt_names, "html-line-numbers";
    push @$rgetopt_names, "html-entities!";
    push @$rgetopt_names, "stylesheet";
    push @$rgetopt_names, "html-table-of-contents!";
    push @$rgetopt_names, "pod2html!";
    push @$rgetopt_names, "frames!";
    push @$rgetopt_names, "html-toc-extension=s";
    push @$rgetopt_names, "html-src-extension=s";

    # Pod::Html parameters:
    push @$rgetopt_names, "backlink=s";
    push @$rgetopt_names, "cachedir=s";
    push @$rgetopt_names, "htmlroot=s";
    push @$rgetopt_names, "libpods=s";
    push @$rgetopt_names, "podpath=s";
    push @$rgetopt_names, "podroot=s";
    push @$rgetopt_names, "title=s";

    # Pod::Html parameters with leading 'pod' which will be removed
    # before the call to Pod::Html
    push @$rgetopt_names, "podquiet!";
    push @$rgetopt_names, "podverbose!";
    push @$rgetopt_names, "podrecurse!";
    push @$rgetopt_names, "podflush";
    push @$rgetopt_names, "podheader!";
    push @$rgetopt_names, "podindex!";
}

sub make_abbreviated_names {

    # We're appending things like this to the expansion list:
    #      'hcc'    => [qw(html-color-comment)],
    #      'hck'    => [qw(html-color-keyword)],
    #  etc
    my $class = shift;
    my ($rexpansion) = @_;

    # abbreviations for color/bold/italic properties
    while ( my ( $short_name, $long_name ) = each %short_to_long_names ) {
        ${$rexpansion}{"hc$short_name"}  = ["html-color-$long_name"];
        ${$rexpansion}{"hb$short_name"}  = ["html-bold-$long_name"];
        ${$rexpansion}{"hi$short_name"}  = ["html-italic-$long_name"];
        ${$rexpansion}{"nhb$short_name"} = ["nohtml-bold-$long_name"];
        ${$rexpansion}{"nhi$short_name"} = ["nohtml-italic-$long_name"];
    }

    # abbreviations for all other html options
    ${$rexpansion}{"hcbg"}  = ["html-color-background"];
    ${$rexpansion}{"pre"}   = ["html-pre-only"];
    ${$rexpansion}{"toc"}   = ["html-table-of-contents"];
    ${$rexpansion}{"ntoc"}  = ["nohtml-table-of-contents"];
    ${$rexpansion}{"nnn"}   = ["html-line-numbers"];
    ${$rexpansion}{"hent"}  = ["html-entities"];
    ${$rexpansion}{"nhent"} = ["nohtml-entities"];
    ${$rexpansion}{"css"}   = ["html-linked-style-sheet"];
    ${$rexpansion}{"nss"}   = ["nohtml-style-sheets"];
    ${$rexpansion}{"ss"}    = ["stylesheet"];
    ${$rexpansion}{"pod"}   = ["pod2html"];
    ${$rexpansion}{"npod"}  = ["nopod2html"];
    ${$rexpansion}{"frm"}   = ["frames"];
    ${$rexpansion}{"nfrm"}  = ["noframes"];
    ${$rexpansion}{"text"}  = ["html-toc-extension"];
    ${$rexpansion}{"sext"}  = ["html-src-extension"];
}

sub check_options {

    # This will be called once after options have been parsed
    my $class = shift;
    $rOpts = shift;

    # X11 color names for default settings that seemed to look ok
    # (these color names are only used for programming clarity; the hex
    # numbers are actually written)
    use constant ForestGreen   => "#228B22";
    use constant SaddleBrown   => "#8B4513";
    use constant magenta4      => "#8B008B";
    use constant IndianRed3    => "#CD5555";
    use constant DeepSkyBlue4  => "#00688B";
    use constant MediumOrchid3 => "#B452CD";
    use constant black         => "#000000";
    use constant white         => "#FFFFFF";
    use constant red           => "#FF0000";

    # set default color, bold, italic properties
    # anything not listed here will be given the default (punctuation) color --
    # these types currently not listed and get default: ws pu s sc cm co p
    # When adding NEW_TOKENS: add an entry here if you don't want defaults

    # set_default_properties( $short_name, default_color, bold?, italic? );
    set_default_properties( 'c',  ForestGreen,   0, 0 );
    set_default_properties( 'pd', ForestGreen,   0, 1 );
    set_default_properties( 'k',  magenta4,      1, 0 );    # was SaddleBrown
    set_default_properties( 'q',  IndianRed3,    0, 0 );
    set_default_properties( 'hh', IndianRed3,    0, 1 );
    set_default_properties( 'h',  IndianRed3,    1, 0 );
    set_default_properties( 'i',  DeepSkyBlue4,  0, 0 );
    set_default_properties( 'w',  black,         0, 0 );
    set_default_properties( 'n',  MediumOrchid3, 0, 0 );
    set_default_properties( 'v',  MediumOrchid3, 0, 0 );
    set_default_properties( 'j',  IndianRed3,    1, 0 );
    set_default_properties( 'm',  red,           1, 0 );

    set_default_color( 'html-color-background',  white );
    set_default_color( 'html-color-punctuation', black );

    # setup property lookup tables for tokens based on their short names
    # every token type has a short name, and will use these tables
    # to do the html markup
    while ( my ( $short_name, $long_name ) = each %short_to_long_names ) {
        $html_color{$short_name}  = $rOpts->{"html-color-$long_name"};
        $html_bold{$short_name}   = $rOpts->{"html-bold-$long_name"};
        $html_italic{$short_name} = $rOpts->{"html-italic-$long_name"};
    }

    # write style sheet to STDOUT and die if requested
    if ( defined( $rOpts->{'stylesheet'} ) ) {
        write_style_sheet_file('-');
        Perl::Tidy::Exit 0;
    }

    # make sure user gives a file name after -css
    if ( defined( $rOpts->{'html-linked-style-sheet'} ) ) {
        $css_linkname = $rOpts->{'html-linked-style-sheet'};
        if ( $css_linkname =~ /^-/ ) {
            Perl::Tidy::Die "You must specify a valid filename after -css\n";
        }
    }

    # check for conflict
    if ( $css_linkname && $rOpts->{'nohtml-style-sheets'} ) {
        $rOpts->{'nohtml-style-sheets'} = 0;
        warning("You can't specify both -css and -nss; -nss ignored\n");
    }

    # write a style sheet file if necessary
    if ($css_linkname) {

        # if the selected filename exists, don't write, because user may
        # have done some work by hand to create it; use backup name instead
        # Also, this will avoid a potential disaster in which the user
        # forgets to specify the style sheet, like this:
        #    perltidy -html -css myfile1.pl myfile2.pl
        # This would cause myfile1.pl to parsed as the style sheet by GetOpts
        my $css_filename = $css_linkname;
        unless ( -e $css_filename ) {
            write_style_sheet_file($css_filename);
        }
    }
    $missing_html_entities = 1 unless $rOpts->{'html-entities'};
}

sub write_style_sheet_file {

    my $css_filename = shift;
    my $fh;
    unless ( $fh = IO::File->new("> $css_filename") ) {
        Perl::Tidy::Die "can't open $css_filename: $!\n";
    }
    write_style_sheet_data($fh);
    eval { $fh->close };
}

sub write_style_sheet_data {

    # write the style sheet data to an open file handle
    my $fh = shift;

    my $bg_color   = $rOpts->{'html-color-background'};
    my $text_color = $rOpts->{'html-color-punctuation'};

    # pre-bgcolor is new, and may not be defined
    my $pre_bg_color = $rOpts->{'html-pre-color-background'};
    $pre_bg_color = $bg_color unless $pre_bg_color;

    $fh->print(<<"EOM");
/* default style sheet generated by perltidy */
body {background: $bg_color; color: $text_color}
pre { color: $text_color;
      background: $pre_bg_color;
      font-family: courier;
    }

EOM

    foreach my $short_name ( sort keys %short_to_long_names ) {
        my $long_name = $short_to_long_names{$short_name};

        my $abbrev = '.' . $short_name;
        if ( length($short_name) == 1 ) { $abbrev .= ' ' }    # for alignment
        my $color = $html_color{$short_name};
        if ( !defined($color) ) { $color = $text_color }
        $fh->print("$abbrev \{ color: $color;");

        if ( $html_bold{$short_name} ) {
            $fh->print(" font-weight:bold;");
        }

        if ( $html_italic{$short_name} ) {
            $fh->print(" font-style:italic;");
        }
        $fh->print("} /* $long_name */\n");
    }
}

sub set_default_color {

    # make sure that options hash $rOpts->{$key} contains a valid color
    my ( $key, $color ) = @_;
    if ( $rOpts->{$key} ) { $color = $rOpts->{$key} }
    $rOpts->{$key} = check_RGB($color);
}

sub check_RGB {

    # if color is a 6 digit hex RGB value, prepend a #, otherwise
    # assume that it is a valid ascii color name
    my ($color) = @_;
    if ( $color =~ /^[0-9a-fA-F]{6,6}$/ ) { $color = "#$color" }
    return $color;
}

sub set_default_properties {
    my ( $short_name, $color, $bold, $italic ) = @_;

    set_default_color( "html-color-$short_to_long_names{$short_name}", $color );
    my $key;
    $key = "html-bold-$short_to_long_names{$short_name}";
    $rOpts->{$key} = ( defined $rOpts->{$key} ) ? $rOpts->{$key} : $bold;
    $key = "html-italic-$short_to_long_names{$short_name}";
    $rOpts->{$key} = ( defined $rOpts->{$key} ) ? $rOpts->{$key} : $italic;
}

sub pod_to_html {

    # Use Pod::Html to process the pod and make the page
    # then merge the perltidy code sections into it.
    # return 1 if success, 0 otherwise
    my $self = shift;
    my ( $pod_string, $css_string, $toc_string, $rpre_string_stack ) = @_;
    my $input_file   = $self->{_input_file};
    my $title        = $self->{_title};
    my $success_flag = 0;

    # don't try to use pod2html if no pod
    unless ($pod_string) {
        return $success_flag;
    }

    # Pod::Html requires a real temporary filename
    # If we are making a frame, we have a name available
    # Otherwise, we have to fine one
    my $tmpfile;
    if ( $rOpts->{'frames'} ) {
        $tmpfile = $self->{_toc_filename};
    }
    else {
        $tmpfile = Perl::Tidy::make_temporary_filename();
    }
    my $fh_tmp = IO::File->new( $tmpfile, 'w' );
    unless ($fh_tmp) {
        Perl::Tidy::Warn
          "unable to open temporary file $tmpfile; cannot use pod2html\n";
        return $success_flag;
    }

    #------------------------------------------------------------------
    # Warning: a temporary file is open; we have to clean up if
    # things go bad.  From here on all returns should be by going to
    # RETURN so that the temporary file gets unlinked.
    #------------------------------------------------------------------

    # write the pod text to the temporary file
    $fh_tmp->print($pod_string);
    $fh_tmp->close();

    # Hand off the pod to pod2html.
    # Note that we can use the same temporary filename for input and output
    # because of the way pod2html works.
    {

        my @args;
        push @args, "--infile=$tmpfile", "--outfile=$tmpfile", "--title=$title";
        my $kw;

        # Flags with string args:
        # "backlink=s", "cachedir=s", "htmlroot=s", "libpods=s",
        # "podpath=s", "podroot=s"
        # Note: -css=s is handled by perltidy itself
        foreach $kw (qw(backlink cachedir htmlroot libpods podpath podroot)) {
            if ( $rOpts->{$kw} ) { push @args, "--$kw=$rOpts->{$kw}" }
        }

        # Toggle switches; these have extra leading 'pod'
        # "header!", "index!", "recurse!", "quiet!", "verbose!"
        foreach $kw (qw(podheader podindex podrecurse podquiet podverbose)) {
            my $kwd = $kw;    # allows us to strip 'pod'
            if ( $rOpts->{$kw} ) { $kwd =~ s/^pod//; push @args, "--$kwd" }
            elsif ( defined( $rOpts->{$kw} ) ) {
                $kwd =~ s/^pod//;
                push @args, "--no$kwd";
            }
        }

        # "flush",
        $kw = 'podflush';
        if ( $rOpts->{$kw} ) { $kw =~ s/^pod//; push @args, "--$kw" }

        # Must clean up if pod2html dies (it can);
        # Be careful not to overwrite callers __DIE__ routine
        local $SIG{__DIE__} = sub {
            unlink $tmpfile if -e $tmpfile;
            Perl::Tidy::Die $_[0];
        };

        pod2html(@args);
    }
    $fh_tmp = IO::File->new( $tmpfile, 'r' );
    unless ($fh_tmp) {

        # this error shouldn't happen ... we just used this filename
        Perl::Tidy::Warn
          "unable to open temporary file $tmpfile; cannot use pod2html\n";
        goto RETURN;
    }

    my $html_fh = $self->{_html_fh};
    my @toc;
    my $in_toc;
    my $no_print;

    # This routine will write the html selectively and store the toc
    my $html_print = sub {
        foreach (@_) {
            $html_fh->print($_) unless ($no_print);
            if ($in_toc) { push @toc, $_ }
        }
    };

    # loop over lines of html output from pod2html and merge in
    # the necessary perltidy html sections
    my ( $saw_body, $saw_index, $saw_body_end );
    while ( my $line = $fh_tmp->getline() ) {

        if ( $line =~ /^\s*<html>\s*$/i ) {
            my $date = localtime;
            $html_print->("<!-- Generated by perltidy on $date -->\n");
            $html_print->($line);
        }

        # Copy the perltidy css, if any, after <body> tag
        elsif ( $line =~ /^\s*<body.*>\s*$/i ) {
            $saw_body = 1;
            $html_print->($css_string) if $css_string;
            $html_print->($line);

            # add a top anchor and heading
            $html_print->("<a name=\"-top-\"></a>\n");
            $title = escape_html($title);
            $html_print->("<h1>$title</h1>\n");
        }
        elsif ( $line =~ /^\s*<!-- INDEX BEGIN -->\s*$/i ) {
            $in_toc = 1;

            # when frames are used, an extra table of contents in the
            # contents panel is confusing, so don't print it
            $no_print = $rOpts->{'frames'}
              || !$rOpts->{'html-table-of-contents'};
            $html_print->("<h2>Doc Index:</h2>\n") if $rOpts->{'frames'};
            $html_print->($line);
        }

        # Copy the perltidy toc, if any, after the Pod::Html toc
        elsif ( $line =~ /^\s*<!-- INDEX END -->\s*$/i ) {
            $saw_index = 1;
            $html_print->($line);
            if ($toc_string) {
                $html_print->("<hr />\n") if $rOpts->{'frames'};
                $html_print->("<h2>Code Index:</h2>\n");
                my @toc = map { $_ .= "\n" } split /\n/, $toc_string;
                $html_print->(@toc);
            }
            $in_toc   = 0;
            $no_print = 0;
        }

        # Copy one perltidy section after each marker
        elsif ( $line =~ /^(.*)<!-- pERLTIDY sECTION -->(.*)$/ ) {
            $line = $2;
            $html_print->($1) if $1;

            # Intermingle code and pod sections if we saw multiple =cut's.
            if ( $self->{_pod_cut_count} > 1 ) {
                my $rpre_string = shift(@$rpre_string_stack);
                if ($$rpre_string) {
                    $html_print->('<pre>');
                    $html_print->($$rpre_string);
                    $html_print->('</pre>');
                }
                else {

                    # shouldn't happen: we stored a string before writing
                    # each marker.
                    Perl::Tidy::Warn
"Problem merging html stream with pod2html; order may be wrong\n";
                }
                $html_print->($line);
            }

            # If didn't see multiple =cut lines, we'll put the pod out first
            # and then the code, because it's less confusing.
            else {

                # since we are not intermixing code and pod, we don't need
                # or want any <hr> lines which separated pod and code
                $html_print->($line) unless ( $line =~ /^\s*<hr>\s*$/i );
            }
        }

        # Copy any remaining code section before the </body> tag
        elsif ( $line =~ /^\s*<\/body>\s*$/i ) {
            $saw_body_end = 1;
            if (@$rpre_string_stack) {
                unless ( $self->{_pod_cut_count} > 1 ) {
                    $html_print->('<hr />');
                }
                while ( my $rpre_string = shift(@$rpre_string_stack) ) {
                    $html_print->('<pre>');
                    $html_print->($$rpre_string);
                    $html_print->('</pre>');
                }
            }
            $html_print->($line);
        }
        else {
            $html_print->($line);
        }
    }

    $success_flag = 1;
    unless ($saw_body) {
        Perl::Tidy::Warn "Did not see <body> in pod2html output\n";
        $success_flag = 0;
    }
    unless ($saw_body_end) {
        Perl::Tidy::Warn "Did not see </body> in pod2html output\n";
        $success_flag = 0;
    }
    unless ($saw_index) {
        Perl::Tidy::Warn "Did not find INDEX END in pod2html output\n";
        $success_flag = 0;
    }

  RETURN:
    eval { $html_fh->close() };

    # note that we have to unlink tmpfile before making frames
    # because the tmpfile may be one of the names used for frames
    unlink $tmpfile if -e $tmpfile;
    if ( $success_flag && $rOpts->{'frames'} ) {
        $self->make_frame( \@toc );
    }
    return $success_flag;
}

sub make_frame {

    # Make a frame with table of contents in the left panel
    # and the text in the right panel.
    # On entry:
    #  $html_filename contains the no-frames html output
    #  $rtoc is a reference to an array with the table of contents
    my $self          = shift;
    my ($rtoc)        = @_;
    my $input_file    = $self->{_input_file};
    my $html_filename = $self->{_html_file};
    my $toc_filename  = $self->{_toc_filename};
    my $src_filename  = $self->{_src_filename};
    my $title         = $self->{_title};
    $title = escape_html($title);

    # FUTURE input parameter:
    my $top_basename = "";

    # We need to produce 3 html files:
    # 1. - the table of contents
    # 2. - the contents (source code) itself
    # 3. - the frame which contains them

    # get basenames for relative links
    my ( $toc_basename, $toc_path ) = fileparse($toc_filename);
    my ( $src_basename, $src_path ) = fileparse($src_filename);

    # 1. Make the table of contents panel, with appropriate changes
    # to the anchor names
    my $src_frame_name = 'SRC';
    my $first_anchor =
      write_toc_html( $title, $toc_filename, $src_basename, $rtoc,
        $src_frame_name );

    # 2. The current .html filename is renamed to be the contents panel
    rename( $html_filename, $src_filename )
      or Perl::Tidy::Die "Cannot rename $html_filename to $src_filename:$!\n";

    # 3. Then use the original html filename for the frame
    write_frame_html(
        $title,        $html_filename, $top_basename,
        $toc_basename, $src_basename,  $src_frame_name
    );
}

sub write_toc_html {

    # write a separate html table of contents file for frames
    my ( $title, $toc_filename, $src_basename, $rtoc, $src_frame_name ) = @_;
    my $fh = IO::File->new( $toc_filename, 'w' )
      or Perl::Tidy::Die "Cannot open $toc_filename:$!\n";
    $fh->print(<<EOM);
<html>
<head>
<title>$title</title>
</head>
<body>
<h1><a href=\"$src_basename#-top-" target="$src_frame_name">$title</a></h1>
EOM

    my $first_anchor =
      change_anchor_names( $rtoc, $src_basename, "$src_frame_name" );
    $fh->print( join "", @$rtoc );

    $fh->print(<<EOM);
</body>
</html>
EOM

}

sub write_frame_html {

    # write an html file to be the table of contents frame
    my (
        $title,        $frame_filename, $top_basename,
        $toc_basename, $src_basename,   $src_frame_name
    ) = @_;

    my $fh = IO::File->new( $frame_filename, 'w' )
      or Perl::Tidy::Die "Cannot open $toc_basename:$!\n";

    $fh->print(<<EOM);
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN"
    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd">
<?xml version="1.0" encoding="iso-8859-1" ?>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>$title</title>
</head>
EOM

    # two left panels, one right, if master index file
    if ($top_basename) {
        $fh->print(<<EOM);
<frameset cols="20%,80%">
<frameset rows="30%,70%">
<frame src = "$top_basename" />
<frame src = "$toc_basename" />
</frameset>
EOM
    }

    # one left panels, one right, if no master index file
    else {
        $fh->print(<<EOM);
<frameset cols="20%,*">
<frame src = "$toc_basename" />
EOM
    }
    $fh->print(<<EOM);
<frame src = "$src_basename" name = "$src_frame_name" />
<noframes>
<body>
<p>If you see this message, you are using a non-frame-capable web client.</p>
<p>This document contains:</p>
<ul>
<li><a href="$toc_basename">A table of contents</a></li>
<li><a href="$src_basename">The source code</a></li>
</ul>
</body>
</noframes>
</frameset>
</html>
EOM
}

sub change_anchor_names {

    # add a filename and target to anchors
    # also return the first anchor
    my ( $rlines, $filename, $target ) = @_;
    my $first_anchor;
    foreach my $line (@$rlines) {

        #  We're looking for lines like this:
        #  <LI><A HREF="#synopsis">SYNOPSIS</A></LI>
        #  ----  -       --------  -----------------
        #  $1              $4            $5
        if ( $line =~ /^(.*)<a(.*)href\s*=\s*"([^#]*)#([^"]+)"[^>]*>(.*)$/i ) {
            my $pre  = $1;
            my $name = $4;
            my $post = $5;
            my $href = "$filename#$name";
            $line = "$pre<a href=\"$href\" target=\"$target\">$post\n";
            unless ($first_anchor) { $first_anchor = $href }
        }
    }
    return $first_anchor;
}

sub close_html_file {
    my $self = shift;
    return unless $self->{_html_file_opened};

    my $html_fh     = $self->{_html_fh};
    my $rtoc_string = $self->{_rtoc_string};

    # There are 3 basic paths to html output...

    # ---------------------------------
    # Path 1: finish up if in -pre mode
    # ---------------------------------
    if ( $rOpts->{'html-pre-only'} ) {
        $html_fh->print( <<"PRE_END");
</pre>
PRE_END
        eval { $html_fh->close() };
        return;
    }

    # Finish the index
    $self->add_toc_item( 'EOF', 'EOF' );

    my $rpre_string_stack = $self->{_rpre_string_stack};

    # Patch to darken the <pre> background color in case of pod2html and
    # interleaved code/documentation.  Otherwise, the distinction
    # between code and documentation is blurred.
    if (   $rOpts->{pod2html}
        && $self->{_pod_cut_count} >= 1
        && $rOpts->{'html-color-background'} eq '#FFFFFF' )
    {
        $rOpts->{'html-pre-color-background'} = '#F0F0F0';
    }

    # put the css or its link into a string, if used
    my $css_string;
    my $fh_css = Perl::Tidy::IOScalar->new( \$css_string, 'w' );

    # use css linked to another file
    if ( $rOpts->{'html-linked-style-sheet'} ) {
        $fh_css->print(
            qq(<link rel="stylesheet" href="$css_linkname" type="text/css" />)
        );
    }

    # use css embedded in this file
    elsif ( !$rOpts->{'nohtml-style-sheets'} ) {
        $fh_css->print( <<'ENDCSS');
<style type="text/css">
<!--
ENDCSS
        write_style_sheet_data($fh_css);
        $fh_css->print( <<"ENDCSS");
-->
</style>
ENDCSS
    }

    # -----------------------------------------------------------
    # path 2: use pod2html if requested
    #         If we fail for some reason, continue on to path 3
    # -----------------------------------------------------------
    if ( $rOpts->{'pod2html'} ) {
        my $rpod_string = $self->{_rpod_string};
        $self->pod_to_html( $$rpod_string, $css_string, $$rtoc_string,
            $rpre_string_stack )
          && return;
    }

    # --------------------------------------------------
    # path 3: write code in html, with pod only in italics
    # --------------------------------------------------
    my $input_file = $self->{_input_file};
    my $title      = escape_html($input_file);
    my $date       = localtime;
    $html_fh->print( <<"HTML_START");
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
   "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<!-- Generated by perltidy on $date -->
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>$title</title>
HTML_START

    # output the css, if used
    if ($css_string) {
        $html_fh->print($css_string);
        $html_fh->print( <<"ENDCSS");
</head>
<body>
ENDCSS
    }
    else {

        $html_fh->print( <<"HTML_START");
</head>
<body bgcolor=\"$rOpts->{'html-color-background'}\" text=\"$rOpts->{'html-color-punctuation'}\">
HTML_START
    }

    $html_fh->print("<a name=\"-top-\"></a>\n");
    $html_fh->print( <<"EOM");
<h1>$title</h1>
EOM

    # copy the table of contents
    if (   $$rtoc_string
        && !$rOpts->{'frames'}
        && $rOpts->{'html-table-of-contents'} )
    {
        $html_fh->print($$rtoc_string);
    }

    # copy the pre section(s)
    my $fname_comment = $input_file;
    $fname_comment =~ s/--+/-/g;    # protect HTML comment tags
    $html_fh->print( <<"END_PRE");
<hr />
<!-- contents of filename: $fname_comment -->
<pre>
END_PRE

    foreach my $rpre_string (@$rpre_string_stack) {
        $html_fh->print($$rpre_string);
    }

    # and finish the html page
    $html_fh->print( <<"HTML_END");
</pre>
</body>
</html>
HTML_END
    eval { $html_fh->close() };    # could be object without close method

    if ( $rOpts->{'frames'} ) {
        my @toc = map { $_ .= "\n" } split /\n/, $$rtoc_string;
        $self->make_frame( \@toc );
    }
}

sub markup_tokens {
    my $self = shift;
    my ( $rtokens, $rtoken_type, $rlevels ) = @_;
    my ( @colored_tokens, $j, $string, $type, $token, $level );
    my $rlast_level    = $self->{_rlast_level};
    my $rpackage_stack = $self->{_rpackage_stack};

    for ( $j = 0 ; $j < @$rtoken_type ; $j++ ) {
        $type  = $$rtoken_type[$j];
        $token = $$rtokens[$j];
        $level = $$rlevels[$j];
        $level = 0 if ( $level < 0 );

        #-------------------------------------------------------
        # Update the package stack.  The package stack is needed to keep
        # the toc correct because some packages may be declared within
        # blocks and go out of scope when we leave the block.
        #-------------------------------------------------------
        if ( $level > $$rlast_level ) {
            unless ( $rpackage_stack->[ $level - 1 ] ) {
                $rpackage_stack->[ $level - 1 ] = 'main';
            }
            $rpackage_stack->[$level] = $rpackage_stack->[ $level - 1 ];
        }
        elsif ( $level < $$rlast_level ) {
            my $package = $rpackage_stack->[$level];
            unless ($package) { $package = 'main' }

            # if we change packages due to a nesting change, we
            # have to make an entry in the toc
            if ( $package ne $rpackage_stack->[ $level + 1 ] ) {
                $self->add_toc_item( $package, 'package' );
            }
        }
        $$rlast_level = $level;

        #-------------------------------------------------------
        # Intercept a sub name here; split it
        # into keyword 'sub' and sub name; and add an
        # entry in the toc
        #-------------------------------------------------------
        if ( $type eq 'i' && $token =~ /^(sub\s+)(\w.*)$/ ) {
            $token = $self->markup_html_element( $1, 'k' );
            push @colored_tokens, $token;
            $token = $2;
            $type  = 'M';

            # but don't include sub declarations in the toc;
            # these wlll have leading token types 'i;'
            my $signature = join "", @$rtoken_type;
            unless ( $signature =~ /^i;/ ) {
                my $subname = $token;
                $subname =~ s/[\s\(].*$//; # remove any attributes and prototype
                $self->add_toc_item( $subname, 'sub' );
            }
        }

        #-------------------------------------------------------
        # Intercept a package name here; split it
        # into keyword 'package' and name; add to the toc,
        # and update the package stack
        #-------------------------------------------------------
        if ( $type eq 'i' && $token =~ /^(package\s+)(\w.*)$/ ) {
            $token = $self->markup_html_element( $1, 'k' );
            push @colored_tokens, $token;
            $token = $2;
            $type  = 'i';
            $self->add_toc_item( "$token", 'package' );
            $rpackage_stack->[$level] = $token;
        }

        $token = $self->markup_html_element( $token, $type );
        push @colored_tokens, $token;
    }
    return ( \@colored_tokens );
}

sub markup_html_element {
    my $self = shift;
    my ( $token, $type ) = @_;

    return $token if ( $type eq 'b' );         # skip a blank token
    return $token if ( $token =~ /^\s*$/ );    # skip a blank line
    $token = escape_html($token);

    # get the short abbreviation for this token type
    my $short_name = $token_short_names{$type};
    if ( !defined($short_name) ) {
        $short_name = "pu";                    # punctuation is default
    }

    # handle style sheets..
    if ( !$rOpts->{'nohtml-style-sheets'} ) {
        if ( $short_name ne 'pu' ) {
            $token = qq(<span class="$short_name">) . $token . "</span>";
        }
    }

    # handle no style sheets..
    else {
        my $color = $html_color{$short_name};

        if ( $color && ( $color ne $rOpts->{'html-color-punctuation'} ) ) {
            $token = qq(<font color="$color">) . $token . "</font>";
        }
        if ( $html_italic{$short_name} ) { $token = "<i>$token</i>" }
        if ( $html_bold{$short_name} )   { $token = "<b>$token</b>" }
    }
    return $token;
}

sub escape_html {

    my $token = shift;
    if ($missing_html_entities) {
        $token =~ s/\&/&amp;/g;
        $token =~ s/\</&lt;/g;
        $token =~ s/\>/&gt;/g;
        $token =~ s/\"/&quot;/g;
    }
    else {
        HTML::Entities::encode_entities($token);
    }
    return $token;
}

sub finish_formatting {

    # called after last line
    my $self = shift;
    $self->close_html_file();
    return;
}

sub write_line {

    my $self = shift;
    return unless $self->{_html_file_opened};
    my $html_pre_fh      = $self->{_html_pre_fh};
    my ($line_of_tokens) = @_;
    my $line_type        = $line_of_tokens->{_line_type};
    my $input_line       = $line_of_tokens->{_line_text};
    my $line_number      = $line_of_tokens->{_line_number};
    chomp $input_line;

    # markup line of code..
    my $html_line;
    if ( $line_type eq 'CODE' ) {
        my $rtoken_type = $line_of_tokens->{_rtoken_type};
        my $rtokens     = $line_of_tokens->{_rtokens};
        my $rlevels     = $line_of_tokens->{_rlevels};

        if ( $input_line =~ /(^\s*)/ ) {
            $html_line = $1;
        }
        else {
            $html_line = "";
        }
        my ($rcolored_tokens) =
          $self->markup_tokens( $rtokens, $rtoken_type, $rlevels );
        $html_line .= join '', @$rcolored_tokens;
    }

    # markup line of non-code..
    else {
        my $line_character;
        if    ( $line_type eq 'HERE' )       { $line_character = 'H' }
        elsif ( $line_type eq 'HERE_END' )   { $line_character = 'h' }
        elsif ( $line_type eq 'FORMAT' )     { $line_character = 'H' }
        elsif ( $line_type eq 'FORMAT_END' ) { $line_character = 'h' }
        elsif ( $line_type eq 'SYSTEM' )     { $line_character = 'c' }
        elsif ( $line_type eq 'END_START' ) {
            $line_character = 'k';
            $self->add_toc_item( '__END__', '__END__' );
        }
        elsif ( $line_type eq 'DATA_START' ) {
            $line_character = 'k';
            $self->add_toc_item( '__DATA__', '__DATA__' );
        }
        elsif ( $line_type =~ /^POD/ ) {
            $line_character = 'P';
            if ( $rOpts->{'pod2html'} ) {
                my $html_pod_fh = $self->{_html_pod_fh};
                if ( $line_type eq 'POD_START' ) {

                    my $rpre_string_stack = $self->{_rpre_string_stack};
                    my $rpre_string       = $rpre_string_stack->[-1];

                    # if we have written any non-blank lines to the
                    # current pre section, start writing to a new output
                    # string
                    if ( $$rpre_string =~ /\S/ ) {
                        my $pre_string;
                        $html_pre_fh =
                          Perl::Tidy::IOScalar->new( \$pre_string, 'w' );
                        $self->{_html_pre_fh} = $html_pre_fh;
                        push @$rpre_string_stack, \$pre_string;

                        # leave a marker in the pod stream so we know
                        # where to put the pre section we just
                        # finished.
                        my $for_html = '=for html';    # don't confuse pod utils
                        $html_pod_fh->print(<<EOM);

$for_html
<!-- pERLTIDY sECTION -->

EOM
                    }

                    # otherwise, just clear the current string and start
                    # over
                    else {
                        $$rpre_string = "";
                        $html_pod_fh->print("\n");
                    }
                }
                $html_pod_fh->print( $input_line . "\n" );
                if ( $line_type eq 'POD_END' ) {
                    $self->{_pod_cut_count}++;
                    $html_pod_fh->print("\n");
                }
                return;
            }
        }
        else { $line_character = 'Q' }
        $html_line = $self->markup_html_element( $input_line, $line_character );
    }

    # add the line number if requested
    if ( $rOpts->{'html-line-numbers'} ) {
        my $extra_space .=
            ( $line_number < 10 )   ? "   "
          : ( $line_number < 100 )  ? "  "
          : ( $line_number < 1000 ) ? " "
          :                           "";
        $html_line = $extra_space . $line_number . " " . $html_line;
    }

    # write the line
    $html_pre_fh->print("$html_line\n");
}

#####################################################################
#
# The Perl::Tidy::Formatter package adds indentation, whitespace, and
# line breaks to the token stream
#
# WARNING: This is not a real class for speed reasons.  Only one
# Formatter may be used.
#
#####################################################################

package Perl::Tidy::Formatter;

BEGIN {

    # Caution: these debug flags produce a lot of output
    # They should all be 0 except when debugging small scripts
    use constant FORMATTER_DEBUG_FLAG_RECOMBINE   => 0;
    use constant FORMATTER_DEBUG_FLAG_BOND_TABLES => 0;
    use constant FORMATTER_DEBUG_FLAG_BOND        => 0;
    use constant FORMATTER_DEBUG_FLAG_BREAK       => 0;
    use constant FORMATTER_DEBUG_FLAG_CI          => 0;
    use constant FORMATTER_DEBUG_FLAG_FLUSH       => 0;
    use constant FORMATTER_DEBUG_FLAG_FORCE       => 0;
    use constant FORMATTER_DEBUG_FLAG_LIST        => 0;
    use constant FORMATTER_DEBUG_FLAG_NOBREAK     => 0;
    use constant FORMATTER_DEBUG_FLAG_OUTPUT      => 0;
    use constant FORMATTER_DEBUG_FLAG_SPARSE      => 0;
    use constant FORMATTER_DEBUG_FLAG_STORE       => 0;
    use constant FORMATTER_DEBUG_FLAG_UNDOBP      => 0;
    use constant FORMATTER_DEBUG_FLAG_WHITE       => 0;

    my $debug_warning = sub {
        print STDOUT "FORMATTER_DEBUGGING with key $_[0]\n";
    };

    FORMATTER_DEBUG_FLAG_RECOMBINE   && $debug_warning->('RECOMBINE');
    FORMATTER_DEBUG_FLAG_BOND_TABLES && $debug_warning->('BOND_TABLES');
    FORMATTER_DEBUG_FLAG_BOND        && $debug_warning->('BOND');
    FORMATTER_DEBUG_FLAG_BREAK       && $debug_warning->('BREAK');
    FORMATTER_DEBUG_FLAG_CI          && $debug_warning->('CI');
    FORMATTER_DEBUG_FLAG_FLUSH       && $debug_warning->('FLUSH');
    FORMATTER_DEBUG_FLAG_FORCE       && $debug_warning->('FORCE');
    FORMATTER_DEBUG_FLAG_LIST        && $debug_warning->('LIST');
    FORMATTER_DEBUG_FLAG_NOBREAK     && $debug_warning->('NOBREAK');
    FORMATTER_DEBUG_FLAG_OUTPUT      && $debug_warning->('OUTPUT');
    FORMATTER_DEBUG_FLAG_SPARSE      && $debug_warning->('SPARSE');
    FORMATTER_DEBUG_FLAG_STORE       && $debug_warning->('STORE');
    FORMATTER_DEBUG_FLAG_UNDOBP      && $debug_warning->('UNDOBP');
    FORMATTER_DEBUG_FLAG_WHITE       && $debug_warning->('WHITE');
}

use Carp;
use vars qw{

  @gnu_stack
  $max_gnu_stack_index
  $gnu_position_predictor
  $line_start_index_to_go
  $last_indentation_written
  $last_unadjusted_indentation
  $last_leading_token
  $last_output_short_opening_token

  $saw_VERSION_in_this_file
  $saw_END_or_DATA_

  @gnu_item_list
  $max_gnu_item_index
  $gnu_sequence_number
  $last_output_indentation
  %last_gnu_equals
  %gnu_comma_count
  %gnu_arrow_count

  @block_type_to_go
  @type_sequence_to_go
  @container_environment_to_go
  @bond_strength_to_go
  @forced_breakpoint_to_go
  @token_lengths_to_go
  @summed_lengths_to_go
  @levels_to_go
  @leading_spaces_to_go
  @reduced_spaces_to_go
  @matching_token_to_go
  @mate_index_to_go
  @nesting_blocks_to_go
  @ci_levels_to_go
  @nesting_depth_to_go
  @nobreak_to_go
  @old_breakpoint_to_go
  @tokens_to_go
  @types_to_go
  @inext_to_go
  @iprev_to_go

  %saved_opening_indentation

  $max_index_to_go
  $comma_count_in_batch
  $old_line_count_in_batch
  $last_nonblank_index_to_go
  $last_nonblank_type_to_go
  $last_nonblank_token_to_go
  $last_last_nonblank_index_to_go
  $last_last_nonblank_type_to_go
  $last_last_nonblank_token_to_go
  @nonblank_lines_at_depth
  $starting_in_quote
  $ending_in_quote
  @whitespace_level_stack
  $whitespace_last_level

  $in_format_skipping_section
  $format_skipping_pattern_begin
  $format_skipping_pattern_end

  $forced_breakpoint_count
  $forced_breakpoint_undo_count
  @forced_breakpoint_undo_stack
  %postponed_breakpoint

  $tabbing
  $embedded_tab_count
  $first_embedded_tab_at
  $last_embedded_tab_at
  $deleted_semicolon_count
  $first_deleted_semicolon_at
  $last_deleted_semicolon_at
  $added_semicolon_count
  $first_added_semicolon_at
  $last_added_semicolon_at
  $first_tabbing_disagreement
  $last_tabbing_disagreement
  $in_tabbing_disagreement
  $tabbing_disagreement_count
  $input_line_tabbing

  $last_line_type
  $last_line_leading_type
  $last_line_leading_level
  $last_last_line_leading_level

  %block_leading_text
  %block_opening_line_number
  $csc_new_statement_ok
  $csc_last_label
  %csc_block_label
  $accumulating_text_for_block
  $leading_block_text
  $rleading_block_if_elsif_text
  $leading_block_text_level
  $leading_block_text_length_exceeded
  $leading_block_text_line_length
  $leading_block_text_line_number
  $closing_side_comment_prefix_pattern
  $closing_side_comment_list_pattern

  $last_nonblank_token
  $last_nonblank_type
  $last_last_nonblank_token
  $last_last_nonblank_type
  $last_nonblank_block_type
  $last_output_level
  %is_do_follower
  %is_if_brace_follower
  %space_after_keyword
  $rbrace_follower
  $looking_for_else
  %is_last_next_redo_return
  %is_other_brace_follower
  %is_else_brace_follower
  %is_anon_sub_brace_follower
  %is_anon_sub_1_brace_follower
  %is_sort_map_grep
  %is_sort_map_grep_eval
  %is_sort_map_grep_eval_do
  %is_block_without_semicolon
  %is_if_unless
  %is_and_or
  %is_assignment
  %is_chain_operator
  %is_if_unless_and_or_last_next_redo_return

  @has_broken_sublist
  @dont_align
  @want_comma_break

  $is_static_block_comment
  $index_start_one_line_block
  $semicolons_before_block_self_destruct
  $index_max_forced_break
  $input_line_number
  $diagnostics_object
  $vertical_aligner_object
  $logger_object
  $file_writer_object
  $formatter_self
  @ci_stack
  $last_line_had_side_comment
  %want_break_before
  %outdent_keyword
  $static_block_comment_pattern
  $static_side_comment_pattern
  %opening_vertical_tightness
  %closing_vertical_tightness
  %closing_token_indentation
  $some_closing_token_indentation

  %opening_token_right
  %stack_opening_token
  %stack_closing_token

  $block_brace_vertical_tightness_pattern

  $rOpts_add_newlines
  $rOpts_add_whitespace
  $rOpts_block_brace_tightness
  $rOpts_block_brace_vertical_tightness
  $rOpts_brace_left_and_indent
  $rOpts_comma_arrow_breakpoints
  $rOpts_break_at_old_keyword_breakpoints
  $rOpts_break_at_old_comma_breakpoints
  $rOpts_break_at_old_logical_breakpoints
  $rOpts_break_at_old_ternary_breakpoints
  $rOpts_break_at_old_attribute_breakpoints
  $rOpts_closing_side_comment_else_flag
  $rOpts_closing_side_comment_maximum_text
  $rOpts_continuation_indentation
  $rOpts_cuddled_else
  $rOpts_delete_old_whitespace
  $rOpts_fuzzy_line_length
  $rOpts_indent_columns
  $rOpts_line_up_parentheses
  $rOpts_maximum_fields_per_table
  $rOpts_maximum_line_length
  $rOpts_variable_maximum_line_length
  $rOpts_short_concatenation_item_length
  $rOpts_keep_old_blank_lines
  $rOpts_ignore_old_breakpoints
  $rOpts_format_skipping
  $rOpts_space_function_paren
  $rOpts_space_keyword_paren
  $rOpts_keep_interior_semicolons
  $rOpts_ignore_side_comment_lengths
  $rOpts_stack_closing_block_brace
  $rOpts_whitespace_cycle
  $rOpts_tight_secret_operators

  %is_opening_type
  %is_closing_type
  %is_keyword_returning_list
  %tightness
  %matching_token
  $rOpts
  %right_bond_strength
  %left_bond_strength
  %binary_ws_rules
  %want_left_space
  %want_right_space
  %is_digraph
  %is_trigraph
  $bli_pattern
  $bli_list_string
  %is_closing_type
  %is_opening_type
  %is_closing_token
  %is_opening_token
};

BEGIN {

    # default list of block types for which -bli would apply
    $bli_list_string = 'if else elsif unless while for foreach do : sub';

    @_ = qw(
      .. :: << >> ** && .. || // -> => += -= .= %= &= |= ^= *= <>
      <= >= == =~ !~ != ++ -- /= x=
    );
    @is_digraph{@_} = (1) x scalar(@_);

    @_ = qw( ... **= <<= >>= &&= ||= //= <=> );
    @is_trigraph{@_} = (1) x scalar(@_);

    @_ = qw(
      = **= += *= &= <<= &&=
      -= /= |= >>= ||= //=
      .= %= ^=
      x=
    );
    @is_assignment{@_} = (1) x scalar(@_);

    @_ = qw(
      grep
      keys
      map
      reverse
      sort
      split
    );
    @is_keyword_returning_list{@_} = (1) x scalar(@_);

    @_ = qw(is if unless and or err last next redo return);
    @is_if_unless_and_or_last_next_redo_return{@_} = (1) x scalar(@_);

    @_ = qw(last next redo return);
    @is_last_next_redo_return{@_} = (1) x scalar(@_);

    @_ = qw(sort map grep);
    @is_sort_map_grep{@_} = (1) x scalar(@_);

    @_ = qw(sort map grep eval);
    @is_sort_map_grep_eval{@_} = (1) x scalar(@_);

    @_ = qw(sort map grep eval do);
    @is_sort_map_grep_eval_do{@_} = (1) x scalar(@_);

    @_ = qw(if unless);
    @is_if_unless{@_} = (1) x scalar(@_);

    @_ = qw(and or err);
    @is_and_or{@_} = (1) x scalar(@_);

    # Identify certain operators which often occur in chains.
    # Note: the minus (-) causes a side effect of padding of the first line in
    # something like this (by sub set_logical_padding):
    #    Checkbutton => 'Transmission checked',
    #   -variable    => \$TRANS
    # This usually improves appearance so it seems ok.
    @_ = qw(&& || and or : ? . + - * /);
    @is_chain_operator{@_} = (1) x scalar(@_);

    # We can remove semicolons after blocks preceded by these keywords
    @_ =
      qw(BEGIN END CHECK INIT AUTOLOAD DESTROY UNITCHECK continue if elsif else
      unless while until for foreach given when default);
    @is_block_without_semicolon{@_} = (1) x scalar(@_);

    # 'L' is token for opening { at hash key
    @_ = qw" L { ( [ ";
    @is_opening_type{@_} = (1) x scalar(@_);

    # 'R' is token for closing } at hash key
    @_ = qw" R } ) ] ";
    @is_closing_type{@_} = (1) x scalar(@_);

    @_ = qw" { ( [ ";
    @is_opening_token{@_} = (1) x scalar(@_);

    @_ = qw" } ) ] ";
    @is_closing_token{@_} = (1) x scalar(@_);
}

# whitespace codes
use constant WS_YES      => 1;
use constant WS_OPTIONAL => 0;
use constant WS_NO       => -1;

# Token bond strengths.
use constant NO_BREAK    => 10000;
use constant VERY_STRONG => 100;
use constant STRONG      => 2.1;
use constant NOMINAL     => 1.1;
use constant WEAK        => 0.8;
use constant VERY_WEAK   => 0.55;

# values for testing indexes in output array
use constant UNDEFINED_INDEX => -1;

# Maximum number of little messages; probably need not be changed.
use constant MAX_NAG_MESSAGES => 6;

# increment between sequence numbers for each type
# For example, ?: pairs might have numbers 7,11,15,...
use constant TYPE_SEQUENCE_INCREMENT => 4;

{

    # methods to count instances
    my $_count = 0;
    sub get_count        { $_count; }
    sub _increment_count { ++$_count }
    sub _decrement_count { --$_count }
}

sub trim {

    # trim leading and trailing whitespace from a string
    $_[0] =~ s/\s+$//;
    $_[0] =~ s/^\s+//;
    return $_[0];
}

sub max {
    my $max = shift;
    foreach (@_) {
        $max = ( $max < $_ ) ? $_ : $max;
    }
    return $max;
}

sub min {
    my $min = shift;
    foreach (@_) {
        $min = ( $min > $_ ) ? $_ : $min;
    }
    return $min;
}

sub split_words {

    # given a string containing words separated by whitespace,
    # return the list of words
    my ($str) = @_;
    return unless $str;
    $str =~ s/\s+$//;
    $str =~ s/^\s+//;
    return split( /\s+/, $str );
}

# interface to Perl::Tidy::Logger routines
sub warning {
    if ($logger_object) {
        $logger_object->warning(@_);
    }
}

sub complain {
    if ($logger_object) {
        $logger_object->complain(@_);
    }
}

sub write_logfile_entry {
    if ($logger_object) {
        $logger_object->write_logfile_entry(@_);
    }
}

sub black_box {
    if ($logger_object) {
        $logger_object->black_box(@_);
    }
}

sub report_definite_bug {
    if ($logger_object) {
        $logger_object->report_definite_bug();
    }
}

sub get_saw_brace_error {
    if ($logger_object) {
        $logger_object->get_saw_brace_error();
    }
}

sub we_are_at_the_last_line {
    if ($logger_object) {
        $logger_object->we_are_at_the_last_line();
    }
}

# interface to Perl::Tidy::Diagnostics routine
sub write_diagnostics {

    if ($diagnostics_object) {
        $diagnostics_object->write_diagnostics(@_);
    }
}

sub get_added_semicolon_count {
    my $self = shift;
    return $added_semicolon_count;
}

sub DESTROY {
    $_[0]->_decrement_count();
}

sub new {

    my $class = shift;

    # we are given an object with a write_line() method to take lines
    my %defaults = (
        sink_object        => undef,
        diagnostics_object => undef,
        logger_object      => undef,
    );
    my %args = ( %defaults, @_ );

    $logger_object      = $args{logger_object};
    $diagnostics_object = $args{diagnostics_object};

    # we create another object with a get_line() and peek_ahead() method
    my $sink_object = $args{sink_object};
    $file_writer_object =
      Perl::Tidy::FileWriter->new( $sink_object, $rOpts, $logger_object );

    # initialize the leading whitespace stack to negative levels
    # so that we can never run off the end of the stack
    $gnu_position_predictor = 0;    # where the current token is predicted to be
    $max_gnu_stack_index    = 0;
    $max_gnu_item_index     = -1;
    $gnu_stack[0] = new_lp_indentation_item( 0, -1, -1, 0, 0 );
    @gnu_item_list                   = ();
    $last_output_indentation         = 0;
    $last_indentation_written        = 0;
    $last_unadjusted_indentation     = 0;
    $last_leading_token              = "";
    $last_output_short_opening_token = 0;

    $saw_VERSION_in_this_file = !$rOpts->{'pass-version-line'};
    $saw_END_or_DATA_         = 0;

    @block_type_to_go            = ();
    @type_sequence_to_go         = ();
    @container_environment_to_go = ();
    @bond_strength_to_go         = ();
    @forced_breakpoint_to_go     = ();
    @summed_lengths_to_go        = ();    # line length to start of ith token
    @token_lengths_to_go         = ();
    @levels_to_go                = ();
    @matching_token_to_go        = ();
    @mate_index_to_go            = ();
    @nesting_blocks_to_go        = ();
    @ci_levels_to_go             = ();
    @nesting_depth_to_go         = (0);
    @nobreak_to_go               = ();
    @old_breakpoint_to_go        = ();
    @tokens_to_go                = ();
    @types_to_go                 = ();
    @leading_spaces_to_go        = ();
    @reduced_spaces_to_go        = ();
    @inext_to_go                 = ();
    @iprev_to_go                 = ();

    @whitespace_level_stack = ();
    $whitespace_last_level  = -1;

    @dont_align         = ();
    @has_broken_sublist = ();
    @want_comma_break   = ();

    @ci_stack                   = ("");
    $first_tabbing_disagreement = 0;
    $last_tabbing_disagreement  = 0;
    $tabbing_disagreement_count = 0;
    $in_tabbing_disagreement    = 0;
    $input_line_tabbing         = undef;

    $last_line_type               = "";
    $last_last_line_leading_level = 0;
    $last_line_leading_level      = 0;
    $last_line_leading_type       = '#';

    $last_nonblank_token        = ';';
    $last_nonblank_type         = ';';
    $last_last_nonblank_token   = ';';
    $last_last_nonblank_type    = ';';
    $last_nonblank_block_type   = "";
    $last_output_level          = 0;
    $looking_for_else           = 0;
    $embedded_tab_count         = 0;
    $first_embedded_tab_at      = 0;
    $last_embedded_tab_at       = 0;
    $deleted_semicolon_count    = 0;
    $first_deleted_semicolon_at = 0;
    $last_deleted_semicolon_at  = 0;
    $added_semicolon_count      = 0;
    $first_added_semicolon_at   = 0;
    $last_added_semicolon_at    = 0;
    $last_line_had_side_comment = 0;
    $is_static_block_comment    = 0;
    %postponed_breakpoint       = ();

    # variables for adding side comments
    %block_leading_text        = ();
    %block_opening_line_number = ();
    $csc_new_statement_ok      = 1;
    %csc_block_label           = ();

    %saved_opening_indentation  = ();
    $in_format_skipping_section = 0;

    reset_block_text_accumulator();

    prepare_for_new_input_lines();

    $vertical_aligner_object =
      Perl::Tidy::VerticalAligner->initialize( $rOpts, $file_writer_object,
        $logger_object, $diagnostics_object );

    if ( $rOpts->{'entab-leading-whitespace'} ) {
        write_logfile_entry(
"Leading whitespace will be entabbed with $rOpts->{'entab-leading-whitespace'} spaces per tab\n"
        );
    }
    elsif ( $rOpts->{'tabs'} ) {
        write_logfile_entry("Indentation will be with a tab character\n");
    }
    else {
        write_logfile_entry(
            "Indentation will be with $rOpts->{'indent-columns'} spaces\n");
    }

    # This was the start of a formatter referent, but object-oriented
    # coding has turned out to be too slow here.
    $formatter_self = {};

    bless $formatter_self, $class;

    # Safety check..this is not a class yet
    if ( _increment_count() > 1 ) {
        confess
"Attempt to create more than 1 object in $class, which is not a true class yet\n";
    }
    return $formatter_self;
}

sub prepare_for_new_input_lines {

    $gnu_sequence_number++;    # increment output batch counter
    %last_gnu_equals                = ();
    %gnu_comma_count                = ();
    %gnu_arrow_count                = ();
    $line_start_index_to_go         = 0;
    $max_gnu_item_index             = UNDEFINED_INDEX;
    $index_max_forced_break         = UNDEFINED_INDEX;
    $max_index_to_go                = UNDEFINED_INDEX;
    $last_nonblank_index_to_go      = UNDEFINED_INDEX;
    $last_nonblank_type_to_go       = '';
    $last_nonblank_token_to_go      = '';
    $last_last_nonblank_index_to_go = UNDEFINED_INDEX;
    $last_last_nonblank_type_to_go  = '';
    $last_last_nonblank_token_to_go = '';
    $forced_breakpoint_count        = 0;
    $forced_breakpoint_undo_count   = 0;
    $rbrace_follower                = undef;
    $summed_lengths_to_go[0]        = 0;
    $old_line_count_in_batch        = 1;
    $comma_count_in_batch           = 0;
    $starting_in_quote              = 0;

    destroy_one_line_block();
}

sub write_line {

    my $self = shift;
    my ($line_of_tokens) = @_;

    my $line_type  = $line_of_tokens->{_line_type};
    my $input_line = $line_of_tokens->{_line_text};

    if ( $rOpts->{notidy} ) {
        write_unindented_line($input_line);
        $last_line_type = $line_type;
        return;
    }

    # _line_type codes are:
    #   SYSTEM         - system-specific code before hash-bang line
    #   CODE           - line of perl code (including comments)
    #   POD_START      - line starting pod, such as '=head'
    #   POD            - pod documentation text
    #   POD_END        - last line of pod section, '=cut'
    #   HERE           - text of here-document
    #   HERE_END       - last line of here-doc (target word)
    #   FORMAT         - format section
    #   FORMAT_END     - last line of format section, '.'
    #   DATA_START     - __DATA__ line
    #   DATA           - unidentified text following __DATA__
    #   END_START      - __END__ line
    #   END            - unidentified text following __END__
    #   ERROR          - we are in big trouble, probably not a perl script

    # put a blank line after an =cut which comes before __END__ and __DATA__
    # (required by podchecker)
    if ( $last_line_type eq 'POD_END' && !$saw_END_or_DATA_ ) {
        $file_writer_object->reset_consecutive_blank_lines();
        if ( $input_line !~ /^\s*$/ ) { want_blank_line() }
    }

    # handle line of code..
    if ( $line_type eq 'CODE' ) {

        # let logger see all non-blank lines of code
        if ( $input_line !~ /^\s*$/ ) {
            my $output_line_number =
              $vertical_aligner_object->get_output_line_number();
            black_box( $line_of_tokens, $output_line_number );
        }
        print_line_of_tokens($line_of_tokens);
    }

    # handle line of non-code..
    else {

        # set special flags
        my $skip_line = 0;
        my $tee_line  = 0;
        if ( $line_type =~ /^POD/ ) {

            # Pod docs should have a preceding blank line.  But stay
            # out of __END__ and __DATA__ sections, because
            # the user may be using this section for any purpose whatsoever
            if ( $rOpts->{'delete-pod'} ) { $skip_line = 1; }
            if ( $rOpts->{'tee-pod'} )    { $tee_line  = 1; }
            if (  !$skip_line
                && $line_type eq 'POD_START'
                && !$saw_END_or_DATA_ )
            {
                want_blank_line();
            }
        }

        # leave the blank counters in a predictable state
        # after __END__ or __DATA__
        elsif ( $line_type =~ /^(END_START|DATA_START)$/ ) {
            $file_writer_object->reset_consecutive_blank_lines();
            $saw_END_or_DATA_ = 1;
        }

        # write unindented non-code line
        if ( !$skip_line ) {
            if ($tee_line) { $file_writer_object->tee_on() }
            write_unindented_line($input_line);
            if ($tee_line) { $file_writer_object->tee_off() }
        }
    }
    $last_line_type = $line_type;
}

sub create_one_line_block {
    $index_start_one_line_block            = $_[0];
    $semicolons_before_block_self_destruct = $_[1];
}

sub destroy_one_line_block {
    $index_start_one_line_block            = UNDEFINED_INDEX;
    $semicolons_before_block_self_destruct = 0;
}

sub leading_spaces_to_go {

    # return the number of indentation spaces for a token in the output stream;
    # these were previously stored by 'set_leading_whitespace'.

    my $ii = shift;
    if ( $ii < 0 ) { $ii = 0 }
    return get_SPACES( $leading_spaces_to_go[$ii] );

}

sub get_SPACES {

    # return the number of leading spaces associated with an indentation
    # variable $indentation is either a constant number of spaces or an object
    # with a get_SPACES method.
    my $indentation = shift;
    return ref($indentation) ? $indentation->get_SPACES() : $indentation;
}

sub get_RECOVERABLE_SPACES {

    # return the number of spaces (+ means shift right, - means shift left)
    # that we would like to shift a group of lines with the same indentation
    # to get them to line up with their opening parens
    my $indentation = shift;
    return ref($indentation) ? $indentation->get_RECOVERABLE_SPACES() : 0;
}

sub get_AVAILABLE_SPACES_to_go {

    my $item = $leading_spaces_to_go[ $_[0] ];

    # return the number of available leading spaces associated with an
    # indentation variable.  $indentation is either a constant number of
    # spaces or an object with a get_AVAILABLE_SPACES method.
    return ref($item) ? $item->get_AVAILABLE_SPACES() : 0;
}

sub new_lp_indentation_item {

    # this is an interface to the IndentationItem class
    my ( $spaces, $level, $ci_level, $available_spaces, $align_paren ) = @_;

    # A negative level implies not to store the item in the item_list
    my $index = 0;
    if ( $level >= 0 ) { $index = ++$max_gnu_item_index; }

    my $item = Perl::Tidy::IndentationItem->new(
        $spaces,      $level,
        $ci_level,    $available_spaces,
        $index,       $gnu_sequence_number,
        $align_paren, $max_gnu_stack_index,
        $line_start_index_to_go,
    );

    if ( $level >= 0 ) {
        $gnu_item_list[$max_gnu_item_index] = $item;
    }

    return $item;
}

sub set_leading_whitespace {

    # This routine defines leading whitespace
    # given: the level and continuation_level of a token,
    # define: space count of leading string which would apply if it
    # were the first token of a new line.

    my ( $level_abs, $ci_level, $in_continued_quote ) = @_;

    # Adjust levels if necessary to recycle whitespace:
    # given $level_abs, the absolute level
    # define $level, a possibly reduced level for whitespace
    my $level = $level_abs;
    if ( $rOpts_whitespace_cycle && $rOpts_whitespace_cycle > 0 ) {
        if ( $level_abs < $whitespace_last_level ) {
            pop(@whitespace_level_stack);
        }
        if ( !@whitespace_level_stack ) {
            push @whitespace_level_stack, $level_abs;
        }
        elsif ( $level_abs > $whitespace_last_level ) {
            $level = $whitespace_level_stack[-1] +
              ( $level_abs - $whitespace_last_level );

            if (
                # 1 Try to break at a block brace
                (
                       $level > $rOpts_whitespace_cycle
                    && $last_nonblank_type eq '{'
                    && $last_nonblank_token eq '{'
                )

                # 2 Then either a brace or bracket
                || (   $level > $rOpts_whitespace_cycle + 1
                    && $last_nonblank_token =~ /^[\{\[]$/ )

                # 3 Then a paren too
                || $level > $rOpts_whitespace_cycle + 2
              )
            {
                $level = 1;
            }
            push @whitespace_level_stack, $level;
        }
        $level = $whitespace_level_stack[-1];
    }
    $whitespace_last_level = $level_abs;

    # modify for -bli, which adds one continuation indentation for
    # opening braces
    if (   $rOpts_brace_left_and_indent
        && $max_index_to_go == 0
        && $block_type_to_go[$max_index_to_go] =~ /$bli_pattern/o )
    {
        $ci_level++;
    }

    # patch to avoid trouble when input file has negative indentation.
    # other logic should catch this error.
    if ( $level < 0 ) { $level = 0 }

    #-------------------------------------------
    # handle the standard indentation scheme
    #-------------------------------------------
    unless ($rOpts_line_up_parentheses) {
        my $space_count =
          $ci_level * $rOpts_continuation_indentation +
          $level * $rOpts_indent_columns;
        my $ci_spaces =
          ( $ci_level == 0 ) ? 0 : $rOpts_continuation_indentation;

        if ($in_continued_quote) {
            $space_count = 0;
            $ci_spaces   = 0;
        }
        $leading_spaces_to_go[$max_index_to_go] = $space_count;
        $reduced_spaces_to_go[$max_index_to_go] = $space_count - $ci_spaces;
        return;
    }

    #-------------------------------------------------------------
    # handle case of -lp indentation..
    #-------------------------------------------------------------

    # The continued_quote flag means that this is the first token of a
    # line, and it is the continuation of some kind of multi-line quote
    # or pattern.  It requires special treatment because it must have no
    # added leading whitespace. So we create a special indentation item
    # which is not in the stack.
    if ($in_continued_quote) {
        my $space_count     = 0;
        my $available_space = 0;
        $level = -1;    # flag to prevent storing in item_list
        $leading_spaces_to_go[$max_index_to_go] =
          $reduced_spaces_to_go[$max_index_to_go] =
          new_lp_indentation_item( $space_count, $level, $ci_level,
            $available_space, 0 );
        return;
    }

    # get the top state from the stack
    my $space_count      = $gnu_stack[$max_gnu_stack_index]->get_SPACES();
    my $current_level    = $gnu_stack[$max_gnu_stack_index]->get_LEVEL();
    my $current_ci_level = $gnu_stack[$max_gnu_stack_index]->get_CI_LEVEL();

    my $type        = $types_to_go[$max_index_to_go];
    my $token       = $tokens_to_go[$max_index_to_go];
    my $total_depth = $nesting_depth_to_go[$max_index_to_go];

    if ( $type eq '{' || $type eq '(' ) {

        $gnu_comma_count{ $total_depth + 1 } = 0;
        $gnu_arrow_count{ $total_depth + 1 } = 0;

        # If we come to an opening token after an '=' token of some type,
        # see if it would be helpful to 'break' after the '=' to save space
        my $last_equals = $last_gnu_equals{$total_depth};
        if ( $last_equals && $last_equals > $line_start_index_to_go ) {

            # find the position if we break at the '='
            my $i_test = $last_equals;
            if ( $types_to_go[ $i_test + 1 ] eq 'b' ) { $i_test++ }

            # TESTING
            ##my $too_close = ($i_test==$max_index_to_go-1);

            my $test_position = total_line_length( $i_test, $max_index_to_go );
            my $mll = maximum_line_length($i_test);

            if (

                # the equals is not just before an open paren (testing)
                ##!$too_close &&

                # if we are beyond the midpoint
                $gnu_position_predictor > $mll - $rOpts_maximum_line_length / 2

                # or we are beyont the 1/4 point and there was an old
                # break at the equals
                || (
                    $gnu_position_predictor >
                    $mll - $rOpts_maximum_line_length * 3 / 4
                    && (
                        $old_breakpoint_to_go[$last_equals]
                        || (   $last_equals > 0
                            && $old_breakpoint_to_go[ $last_equals - 1 ] )
                        || (   $last_equals > 1
                            && $types_to_go[ $last_equals - 1 ] eq 'b'
                            && $old_breakpoint_to_go[ $last_equals - 2 ] )
                    )
                )
              )
            {

                # then make the switch -- note that we do not set a real
                # breakpoint here because we may not really need one; sub
                # scan_list will do that if necessary
                $line_start_index_to_go = $i_test + 1;
                $gnu_position_predictor = $test_position;
            }
        }
    }

    my $halfway =
      maximum_line_length_for_level($level) - $rOpts_maximum_line_length / 2;

    # Check for decreasing depth ..
    # Note that one token may have both decreasing and then increasing
    # depth. For example, (level, ci) can go from (1,1) to (2,0).  So,
    # in this example we would first go back to (1,0) then up to (2,0)
    # in a single call.
    if ( $level < $current_level || $ci_level < $current_ci_level ) {

        # loop to find the first entry at or completely below this level
        my ( $lev, $ci_lev );
        while (1) {
            if ($max_gnu_stack_index) {

                # save index of token which closes this level
                $gnu_stack[$max_gnu_stack_index]->set_CLOSED($max_index_to_go);

                # Undo any extra indentation if we saw no commas
                my $available_spaces =
                  $gnu_stack[$max_gnu_stack_index]->get_AVAILABLE_SPACES();

                my $comma_count = 0;
                my $arrow_count = 0;
                if ( $type eq '}' || $type eq ')' ) {
                    $comma_count = $gnu_comma_count{$total_depth};
                    $arrow_count = $gnu_arrow_count{$total_depth};
                    $comma_count = 0 unless $comma_count;
                    $arrow_count = 0 unless $arrow_count;
                }
                $gnu_stack[$max_gnu_stack_index]->set_COMMA_COUNT($comma_count);
                $gnu_stack[$max_gnu_stack_index]->set_ARROW_COUNT($arrow_count);

                if ( $available_spaces > 0 ) {

                    if ( $comma_count <= 0 || $arrow_count > 0 ) {

                        my $i = $gnu_stack[$max_gnu_stack_index]->get_INDEX();
                        my $seqno =
                          $gnu_stack[$max_gnu_stack_index]
                          ->get_SEQUENCE_NUMBER();

                        # Be sure this item was created in this batch.  This
                        # should be true because we delete any available
                        # space from open items at the end of each batch.
                        if (   $gnu_sequence_number != $seqno
                            || $i > $max_gnu_item_index )
                        {
                            warning(
"Program bug with -lp.  seqno=$seqno should be $gnu_sequence_number and i=$i should be less than max=$max_gnu_item_index\n"
                            );
                            report_definite_bug();
                        }

                        else {
                            if ( $arrow_count == 0 ) {
                                $gnu_item_list[$i]
                                  ->permanently_decrease_AVAILABLE_SPACES(
                                    $available_spaces);
                            }
                            else {
                                $gnu_item_list[$i]
                                  ->tentatively_decrease_AVAILABLE_SPACES(
                                    $available_spaces);
                            }

                            my $j;
                            for (
                                $j = $i + 1 ;
                                $j <= $max_gnu_item_index ;
                                $j++
                              )
                            {
                                $gnu_item_list[$j]
                                  ->decrease_SPACES($available_spaces);
                            }
                        }
                    }
                }

                # go down one level
                --$max_gnu_stack_index;
                $lev    = $gnu_stack[$max_gnu_stack_index]->get_LEVEL();
                $ci_lev = $gnu_stack[$max_gnu_stack_index]->get_CI_LEVEL();

                # stop when we reach a level at or below the current level
                if ( $lev <= $level && $ci_lev <= $ci_level ) {
                    $space_count =
                      $gnu_stack[$max_gnu_stack_index]->get_SPACES();
                    $current_level    = $lev;
                    $current_ci_level = $ci_lev;
                    last;
                }
            }

            # reached bottom of stack .. should never happen because
            # only negative levels can get here, and $level was forced
            # to be positive above.
            else {
                warning(
"program bug with -lp: stack_error. level=$level; lev=$lev; ci_level=$ci_level; ci_lev=$ci_lev; rerun with -nlp\n"
                );
                report_definite_bug();
                last;
            }
        }
    }

    # handle increasing depth
    if ( $level > $current_level || $ci_level > $current_ci_level ) {

        # Compute the standard incremental whitespace.  This will be
        # the minimum incremental whitespace that will be used.  This
        # choice results in a smooth transition between the gnu-style
        # and the standard style.
        my $standard_increment =
          ( $level - $current_level ) * $rOpts_indent_columns +
          ( $ci_level - $current_ci_level ) * $rOpts_continuation_indentation;

        # Now we have to define how much extra incremental space
        # ("$available_space") we want.  This extra space will be
        # reduced as necessary when long lines are encountered or when
        # it becomes clear that we do not have a good list.
        my $available_space = 0;
        my $align_paren     = 0;
        my $excess          = 0;

        # initialization on empty stack..
        if ( $max_gnu_stack_index == 0 ) {
            $space_count = $level * $rOpts_indent_columns;
        }

        # if this is a BLOCK, add the standard increment
        elsif ($last_nonblank_block_type) {
            $space_count += $standard_increment;
        }

        # if last nonblank token was not structural indentation,
        # just use standard increment
        elsif ( $last_nonblank_type ne '{' ) {
            $space_count += $standard_increment;
        }

        # otherwise use the space to the first non-blank level change token
        else {

            $space_count = $gnu_position_predictor;

            my $min_gnu_indentation =
              $gnu_stack[$max_gnu_stack_index]->get_SPACES();

            $available_space = $space_count - $min_gnu_indentation;
            if ( $available_space >= $standard_increment ) {
                $min_gnu_indentation += $standard_increment;
            }
            elsif ( $available_space > 1 ) {
                $min_gnu_indentation += $available_space + 1;
            }
            elsif ( $last_nonblank_token =~ /^[\{\[\(]$/ ) {
                if ( ( $tightness{$last_nonblank_token} < 2 ) ) {
                    $min_gnu_indentation += 2;
                }
                else {
                    $min_gnu_indentation += 1;
                }
            }
            else {
                $min_gnu_indentation += $standard_increment;
            }
            $available_space = $space_count - $min_gnu_indentation;

            if ( $available_space < 0 ) {
                $space_count     = $min_gnu_indentation;
                $available_space = 0;
            }
            $align_paren = 1;
        }

        # update state, but not on a blank token
        if ( $types_to_go[$max_index_to_go] ne 'b' ) {

            $gnu_stack[$max_gnu_stack_index]->set_HAVE_CHILD(1);

            ++$max_gnu_stack_index;
            $gnu_stack[$max_gnu_stack_index] =
              new_lp_indentation_item( $space_count, $level, $ci_level,
                $available_space, $align_paren );

            # If the opening paren is beyond the half-line length, then
            # we will use the minimum (standard) indentation.  This will
            # help avoid problems associated with running out of space
            # near the end of a line.  As a result, in deeply nested
            # lists, there will be some indentations which are limited
            # to this minimum standard indentation. But the most deeply
            # nested container will still probably be able to shift its
            # parameters to the right for proper alignment, so in most
            # cases this will not be noticeable.
            if ( $available_space > 0 && $space_count > $halfway ) {
                $gnu_stack[$max_gnu_stack_index]
                  ->tentatively_decrease_AVAILABLE_SPACES($available_space);
            }
        }
    }

    # Count commas and look for non-list characters.  Once we see a
    # non-list character, we give up and don't look for any more commas.
    if ( $type eq '=>' ) {
        $gnu_arrow_count{$total_depth}++;

        # tentatively treating '=>' like '=' for estimating breaks
        # TODO: this could use some experimentation
        $last_gnu_equals{$total_depth} = $max_index_to_go;
    }

    elsif ( $type eq ',' ) {
        $gnu_comma_count{$total_depth}++;
    }

    elsif ( $is_assignment{$type} ) {
        $last_gnu_equals{$total_depth} = $max_index_to_go;
    }

    # this token might start a new line
    # if this is a non-blank..
    if ( $type ne 'b' ) {

        # and if ..
        if (

            # this is the first nonblank token of the line
            $max_index_to_go == 1 && $types_to_go[0] eq 'b'

            # or previous character was one of these:
            || $last_nonblank_type_to_go =~ /^([\:\?\,f])$/

            # or previous character was opening and this does not close it
            || ( $last_nonblank_type_to_go eq '{' && $type ne '}' )
            || ( $last_nonblank_type_to_go eq '(' and $type ne ')' )

            # or this token is one of these:
            || $type =~ /^([\.]|\|\||\&\&)$/

            # or this is a closing structure
            || (   $last_nonblank_type_to_go eq '}'
                && $last_nonblank_token_to_go eq $last_nonblank_type_to_go )

            # or previous token was keyword 'return'
            || ( $last_nonblank_type_to_go eq 'k'
                && ( $last_nonblank_token_to_go eq 'return' && $type ne '{' ) )

            # or starting a new line at certain keywords is fine
            || (   $type eq 'k'
                && $is_if_unless_and_or_last_next_redo_return{$token} )

            # or this is after an assignment after a closing structure
            || (
                $is_assignment{$last_nonblank_type_to_go}
                && (
                    $last_last_nonblank_type_to_go =~ /^[\}\)\]]$/

                    # and it is significantly to the right
                    || $gnu_position_predictor > $halfway
                )
            )
          )
        {
            check_for_long_gnu_style_lines();
            $line_start_index_to_go = $max_index_to_go;

            # back up 1 token if we want to break before that type
            # otherwise, we may strand tokens like '?' or ':' on a line
            if ( $line_start_index_to_go > 0 ) {
                if ( $last_nonblank_type_to_go eq 'k' ) {

                    if ( $want_break_before{$last_nonblank_token_to_go} ) {
                        $line_start_index_to_go--;
                    }
                }
                elsif ( $want_break_before{$last_nonblank_type_to_go} ) {
                    $line_start_index_to_go--;
                }
            }
        }
    }

    # remember the predicted position of this token on the output line
    if ( $max_index_to_go > $line_start_index_to_go ) {
        $gnu_position_predictor =
          total_line_length( $line_start_index_to_go, $max_index_to_go );
    }
    else {
        $gnu_position_predictor =
          $space_count + $token_lengths_to_go[$max_index_to_go];
    }

    # store the indentation object for this token
    # this allows us to manipulate the leading whitespace
    # (in case we have to reduce indentation to fit a line) without
    # having to change any token values
    $leading_spaces_to_go[$max_index_to_go] = $gnu_stack[$max_gnu_stack_index];
    $reduced_spaces_to_go[$max_index_to_go] =
      ( $max_gnu_stack_index > 0 && $ci_level )
      ? $gnu_stack[ $max_gnu_stack_index - 1 ]
      : $gnu_stack[$max_gnu_stack_index];
    return;
}

sub check_for_long_gnu_style_lines {

    # look at the current estimated maximum line length, and
    # remove some whitespace if it exceeds the desired maximum

    # this is only for the '-lp' style
    return unless ($rOpts_line_up_parentheses);

    # nothing can be done if no stack items defined for this line
    return if ( $max_gnu_item_index == UNDEFINED_INDEX );

    # see if we have exceeded the maximum desired line length
    # keep 2 extra free because they are needed in some cases
    # (result of trial-and-error testing)
    my $spaces_needed =
      $gnu_position_predictor - maximum_line_length($max_index_to_go) + 2;

    return if ( $spaces_needed <= 0 );

    # We are over the limit, so try to remove a requested number of
    # spaces from leading whitespace.  We are only allowed to remove
    # from whitespace items created on this batch, since others have
    # already been used and cannot be undone.
    my @candidates = ();
    my $i;

    # loop over all whitespace items created for the current batch
    for ( $i = 0 ; $i <= $max_gnu_item_index ; $i++ ) {
        my $item = $gnu_item_list[$i];

        # item must still be open to be a candidate (otherwise it
        # cannot influence the current token)
        next if ( $item->get_CLOSED() >= 0 );

        my $available_spaces = $item->get_AVAILABLE_SPACES();

        if ( $available_spaces > 0 ) {
            push( @candidates, [ $i, $available_spaces ] );
        }
    }

    return unless (@candidates);

    # sort by available whitespace so that we can remove whitespace
    # from the maximum available first
    @candidates = sort { $b->[1] <=> $a->[1] } @candidates;

    # keep removing whitespace until we are done or have no more
    my $candidate;
    foreach $candidate (@candidates) {
        my ( $i, $available_spaces ) = @{$candidate};
        my $deleted_spaces =
          ( $available_spaces > $spaces_needed )
          ? $spaces_needed
          : $available_spaces;

        # remove the incremental space from this item
        $gnu_item_list[$i]->decrease_AVAILABLE_SPACES($deleted_spaces);

        my $i_debug = $i;

        # update the leading whitespace of this item and all items
        # that came after it
        for ( ; $i <= $max_gnu_item_index ; $i++ ) {

            my $old_spaces = $gnu_item_list[$i]->get_SPACES();
            if ( $old_spaces >= $deleted_spaces ) {
                $gnu_item_list[$i]->decrease_SPACES($deleted_spaces);
            }

            # shouldn't happen except for code bug:
            else {
                my $level        = $gnu_item_list[$i_debug]->get_LEVEL();
                my $ci_level     = $gnu_item_list[$i_debug]->get_CI_LEVEL();
                my $old_level    = $gnu_item_list[$i]->get_LEVEL();
                my $old_ci_level = $gnu_item_list[$i]->get_CI_LEVEL();
                warning(
"program bug with -lp: want to delete $deleted_spaces from item $i, but old=$old_spaces deleted: lev=$level ci=$ci_level  deleted: level=$old_level ci=$ci_level\n"
                );
                report_definite_bug();
            }
        }
        $gnu_position_predictor -= $deleted_spaces;
        $spaces_needed          -= $deleted_spaces;
        last unless ( $spaces_needed > 0 );
    }
}

sub finish_lp_batch {

    # This routine is called once after each each output stream batch is
    # finished to undo indentation for all incomplete -lp
    # indentation levels.  It is too risky to leave a level open,
    # because then we can't backtrack in case of a long line to follow.
    # This means that comments and blank lines will disrupt this
    # indentation style.  But the vertical aligner may be able to
    # get the space back if there are side comments.

    # this is only for the 'lp' style
    return unless ($rOpts_line_up_parentheses);

    # nothing can be done if no stack items defined for this line
    return if ( $max_gnu_item_index == UNDEFINED_INDEX );

    # loop over all whitespace items created for the current batch
    my $i;
    for ( $i = 0 ; $i <= $max_gnu_item_index ; $i++ ) {
        my $item = $gnu_item_list[$i];

        # only look for open items
        next if ( $item->get_CLOSED() >= 0 );

        # Tentatively remove all of the available space
        # (The vertical aligner will try to get it back later)
        my $available_spaces = $item->get_AVAILABLE_SPACES();
        if ( $available_spaces > 0 ) {

            # delete incremental space for this item
            $gnu_item_list[$i]
              ->tentatively_decrease_AVAILABLE_SPACES($available_spaces);

            # Reduce the total indentation space of any nodes that follow
            # Note that any such nodes must necessarily be dependents
            # of this node.
            foreach ( $i + 1 .. $max_gnu_item_index ) {
                $gnu_item_list[$_]->decrease_SPACES($available_spaces);
            }
        }
    }
    return;
}

sub reduce_lp_indentation {

    # reduce the leading whitespace at token $i if possible by $spaces_needed
    # (a large value of $spaces_needed will remove all excess space)
    # NOTE: to be called from scan_list only for a sequence of tokens
    # contained between opening and closing parens/braces/brackets

    my ( $i, $spaces_wanted ) = @_;
    my $deleted_spaces = 0;

    my $item             = $leading_spaces_to_go[$i];
    my $available_spaces = $item->get_AVAILABLE_SPACES();

    if (
        $available_spaces > 0
        && ( ( $spaces_wanted <= $available_spaces )
            || !$item->get_HAVE_CHILD() )
      )
    {

        # we'll remove these spaces, but mark them as recoverable
        $deleted_spaces =
          $item->tentatively_decrease_AVAILABLE_SPACES($spaces_wanted);
    }

    return $deleted_spaces;
}

sub token_sequence_length {

    # return length of tokens ($ibeg .. $iend) including $ibeg & $iend
    # returns 0 if $ibeg > $iend (shouldn't happen)
    my ( $ibeg, $iend ) = @_;
    return 0 if ( $iend < 0 || $ibeg > $iend );
    return $summed_lengths_to_go[ $iend + 1 ] if ( $ibeg < 0 );
    return $summed_lengths_to_go[ $iend + 1 ] - $summed_lengths_to_go[$ibeg];
}

sub total_line_length {

    # return length of a line of tokens ($ibeg .. $iend)
    my ( $ibeg, $iend ) = @_;
    return leading_spaces_to_go($ibeg) + token_sequence_length( $ibeg, $iend );
}

sub maximum_line_length_for_level {

    # return maximum line length for line starting with a given level
    my $maximum_line_length = $rOpts_maximum_line_length;

    # Modify if -vmll option is selected
    if ($rOpts_variable_maximum_line_length) {
        my $level = shift;
        if ( $level < 0 ) { $level = 0 }
        $maximum_line_length += $level * $rOpts_indent_columns;
    }
    return $maximum_line_length;
}

sub maximum_line_length {

    # return maximum line length for line starting with the token at given index
    return maximum_line_length_for_level( $levels_to_go[ $_[0] ] );

}

sub excess_line_length {

    # return number of characters by which a line of tokens ($ibeg..$iend)
    # exceeds the allowable line length.
    my ( $ibeg, $iend ) = @_;
    return total_line_length( $ibeg, $iend ) - maximum_line_length($ibeg);
}

sub finish_formatting {

    # flush buffer and write any informative messages
    my $self = shift;

    flush();
    $file_writer_object->decrement_output_line_number()
      ;    # fix up line number since it was incremented
    we_are_at_the_last_line();
    if ( $added_semicolon_count > 0 ) {
        my $first = ( $added_semicolon_count > 1 ) ? "First" : "";
        my $what =
          ( $added_semicolon_count > 1 ) ? "semicolons were" : "semicolon was";
        write_logfile_entry("$added_semicolon_count $what added:\n");
        write_logfile_entry(
            "  $first at input line $first_added_semicolon_at\n");

        if ( $added_semicolon_count > 1 ) {
            write_logfile_entry(
                "   Last at input line $last_added_semicolon_at\n");
        }
        write_logfile_entry("  (Use -nasc to prevent semicolon addition)\n");
        write_logfile_entry("\n");
    }

    if ( $deleted_semicolon_count > 0 ) {
        my $first = ( $deleted_semicolon_count > 1 ) ? "First" : "";
        my $what =
          ( $deleted_semicolon_count > 1 )
          ? "semicolons were"
          : "semicolon was";
        write_logfile_entry(
            "$deleted_semicolon_count unnecessary $what deleted:\n");
        write_logfile_entry(
            "  $first at input line $first_deleted_semicolon_at\n");

        if ( $deleted_semicolon_count > 1 ) {
            write_logfile_entry(
                "   Last at input line $last_deleted_semicolon_at\n");
        }
        write_logfile_entry("  (Use -ndsc to prevent semicolon deletion)\n");
        write_logfile_entry("\n");
    }

    if ( $embedded_tab_count > 0 ) {
        my $first = ( $embedded_tab_count > 1 ) ? "First" : "";
        my $what =
          ( $embedded_tab_count > 1 )
          ? "quotes or patterns"
          : "quote or pattern";
        write_logfile_entry("$embedded_tab_count $what had embedded tabs:\n");
        write_logfile_entry(
"This means the display of this script could vary with device or software\n"
        );
        write_logfile_entry("  $first at input line $first_embedded_tab_at\n");

        if ( $embedded_tab_count > 1 ) {
            write_logfile_entry(
                "   Last at input line $last_embedded_tab_at\n");
        }
        write_logfile_entry("\n");
    }

    if ($first_tabbing_disagreement) {
        write_logfile_entry(
"First indentation disagreement seen at input line $first_tabbing_disagreement\n"
        );
    }

    if ($in_tabbing_disagreement) {
        write_logfile_entry(
"Ending with indentation disagreement which started at input line $in_tabbing_disagreement\n"
        );
    }
    else {

        if ($last_tabbing_disagreement) {

            write_logfile_entry(
"Last indentation disagreement seen at input line $last_tabbing_disagreement\n"
            );
        }
        else {
            write_logfile_entry("No indentation disagreement seen\n");
        }
    }
    if ($first_tabbing_disagreement) {
        write_logfile_entry(
"Note: Indentation disagreement detection is not accurate for outdenting and -lp.\n"
        );
    }
    write_logfile_entry("\n");

    $vertical_aligner_object->report_anything_unusual();

    $file_writer_object->report_line_length_errors();
}

sub check_options {

    # This routine is called to check the Opts hash after it is defined

    ($rOpts) = @_;

    make_static_block_comment_pattern();
    make_static_side_comment_pattern();
    make_closing_side_comment_prefix();
    make_closing_side_comment_list_pattern();
    $format_skipping_pattern_begin =
      make_format_skipping_pattern( 'format-skipping-begin', '#<<<' );
    $format_skipping_pattern_end =
      make_format_skipping_pattern( 'format-skipping-end', '#>>>' );

    # If closing side comments ARE selected, then we can safely
    # delete old closing side comments unless closing side comment
    # warnings are requested.  This is a good idea because it will
    # eliminate any old csc's which fall below the line count threshold.
    # We cannot do this if warnings are turned on, though, because we
    # might delete some text which has been added.  So that must
    # be handled when comments are created.
    if ( $rOpts->{'closing-side-comments'} ) {
        if ( !$rOpts->{'closing-side-comment-warnings'} ) {
            $rOpts->{'delete-closing-side-comments'} = 1;
        }
    }

    # If closing side comments ARE NOT selected, but warnings ARE
    # selected and we ARE DELETING csc's, then we will pretend to be
    # adding with a huge interval.  This will force the comments to be
    # generated for comparison with the old comments, but not added.
    elsif ( $rOpts->{'closing-side-comment-warnings'} ) {
        if ( $rOpts->{'delete-closing-side-comments'} ) {
            $rOpts->{'delete-closing-side-comments'}  = 0;
            $rOpts->{'closing-side-comments'}         = 1;
            $rOpts->{'closing-side-comment-interval'} = 100000000;
        }
    }

    make_bli_pattern();
    make_block_brace_vertical_tightness_pattern();

    if ( $rOpts->{'line-up-parentheses'} ) {

        if (   $rOpts->{'indent-only'}
            || !$rOpts->{'add-newlines'}
            || !$rOpts->{'delete-old-newlines'} )
        {
            Perl::Tidy::Warn <<EOM;
-----------------------------------------------------------------------
Conflict: -lp  conflicts with -io, -fnl, -nanl, or -ndnl; ignoring -lp

The -lp indentation logic requires that perltidy be able to coordinate
arbitrarily large numbers of line breakpoints.  This isn't possible
with these flags. Sometimes an acceptable workaround is to use -wocb=3
-----------------------------------------------------------------------
EOM
            $rOpts->{'line-up-parentheses'} = 0;
        }
    }

    # At present, tabs are not compatible with the line-up-parentheses style
    # (it would be possible to entab the total leading whitespace
    # just prior to writing the line, if desired).
    if ( $rOpts->{'line-up-parentheses'} && $rOpts->{'tabs'} ) {
        Perl::Tidy::Warn <<EOM;
Conflict: -t (tabs) cannot be used with the -lp  option; ignoring -t; see -et.
EOM
        $rOpts->{'tabs'} = 0;
    }

    # Likewise, tabs are not compatible with outdenting..
    if ( $rOpts->{'outdent-keywords'} && $rOpts->{'tabs'} ) {
        Perl::Tidy::Warn <<EOM;
Conflict: -t (tabs) cannot be used with the -okw options; ignoring -t; see -et.
EOM
        $rOpts->{'tabs'} = 0;
    }

    if ( $rOpts->{'outdent-labels'} && $rOpts->{'tabs'} ) {
        Perl::Tidy::Warn <<EOM;
Conflict: -t (tabs) cannot be used with the -ola  option; ignoring -t; see -et.
EOM
        $rOpts->{'tabs'} = 0;
    }

    if ( !$rOpts->{'space-for-semicolon'} ) {
        $want_left_space{'f'} = -1;
    }

    if ( $rOpts->{'space-terminal-semicolon'} ) {
        $want_left_space{';'} = 1;
    }

    # implement outdenting preferences for keywords
    %outdent_keyword = ();
    unless ( @_ = split_words( $rOpts->{'outdent-keyword-okl'} ) ) {
        @_ = qw(next last redo goto return);    # defaults
    }

    # FUTURE: if not a keyword, assume that it is an identifier
    foreach (@_) {
        if ( $Perl::Tidy::Tokenizer::is_keyword{$_} ) {
            $outdent_keyword{$_} = 1;
        }
        else {
            Perl::Tidy::Warn "ignoring '$_' in -okwl list; not a perl keyword";
        }
    }

    # implement user whitespace preferences
    if ( @_ = split_words( $rOpts->{'want-left-space'} ) ) {
        @want_left_space{@_} = (1) x scalar(@_);
    }

    if ( @_ = split_words( $rOpts->{'want-right-space'} ) ) {
        @want_right_space{@_} = (1) x scalar(@_);
    }

    if ( @_ = split_words( $rOpts->{'nowant-left-space'} ) ) {
        @want_left_space{@_} = (-1) x scalar(@_);
    }

    if ( @_ = split_words( $rOpts->{'nowant-right-space'} ) ) {
        @want_right_space{@_} = (-1) x scalar(@_);
    }
    if ( $rOpts->{'dump-want-left-space'} ) {
        dump_want_left_space(*STDOUT);
        Perl::Tidy::Exit 0;
    }

    if ( $rOpts->{'dump-want-right-space'} ) {
        dump_want_right_space(*STDOUT);
        Perl::Tidy::Exit 0;
    }

    # default keywords for which space is introduced before an opening paren
    # (at present, including them messes up vertical alignment)
    @_ = qw(my local our and or err eq ne if else elsif until
      unless while for foreach return switch case given when);
    @space_after_keyword{@_} = (1) x scalar(@_);

    # first remove any or all of these if desired
    if ( @_ = split_words( $rOpts->{'nospace-after-keyword'} ) ) {

        # -nsak='*' selects all the above keywords
        if ( @_ == 1 && $_[0] eq '*' ) { @_ = keys(%space_after_keyword) }
        @space_after_keyword{@_} = (0) x scalar(@_);
    }

    # then allow user to add to these defaults
    if ( @_ = split_words( $rOpts->{'space-after-keyword'} ) ) {
        @space_after_keyword{@_} = (1) x scalar(@_);
    }

    # implement user break preferences
    my @all_operators = qw(% + - * / x != == >= <= =~ !~ < > | &
      = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
      . : ? && || and or err xor
    );

    my $break_after = sub {
        foreach my $tok (@_) {
            if ( $tok eq '?' ) { $tok = ':' }    # patch to coordinate ?/:
            my $lbs = $left_bond_strength{$tok};
            my $rbs = $right_bond_strength{$tok};
            if ( defined($lbs) && defined($rbs) && $lbs < $rbs ) {
                ( $right_bond_strength{$tok}, $left_bond_strength{$tok} ) =
                  ( $lbs, $rbs );
            }
        }
    };

    my $break_before = sub {
        foreach my $tok (@_) {
            my $lbs = $left_bond_strength{$tok};
            my $rbs = $right_bond_strength{$tok};
            if ( defined($lbs) && defined($rbs) && $rbs < $lbs ) {
                ( $right_bond_strength{$tok}, $left_bond_strength{$tok} ) =
                  ( $lbs, $rbs );
            }
        }
    };

    $break_after->(@all_operators) if ( $rOpts->{'break-after-all-operators'} );
    $break_before->(@all_operators)
      if ( $rOpts->{'break-before-all-operators'} );

    $break_after->( split_words( $rOpts->{'want-break-after'} ) );
    $break_before->( split_words( $rOpts->{'want-break-before'} ) );

    # make note if breaks are before certain key types
    %want_break_before = ();
    foreach my $tok ( @all_operators, ',' ) {
        $want_break_before{$tok} =
          $left_bond_strength{$tok} < $right_bond_strength{$tok};
    }

    # Coordinate ?/: breaks, which must be similar
    if ( !$want_break_before{':'} ) {
        $want_break_before{'?'}   = $want_break_before{':'};
        $right_bond_strength{'?'} = $right_bond_strength{':'} + 0.01;
        $left_bond_strength{'?'}  = NO_BREAK;
    }

    # Define here tokens which may follow the closing brace of a do statement
    # on the same line, as in:
    #   } while ( $something);
    @_ = qw(until while unless if ; : );
    push @_, ',';
    @is_do_follower{@_} = (1) x scalar(@_);

    # These tokens may follow the closing brace of an if or elsif block.
    # In other words, for cuddled else we want code to look like:
    #   } elsif ( $something) {
    #   } else {
    if ( $rOpts->{'cuddled-else'} ) {
        @_ = qw(else elsif);
        @is_if_brace_follower{@_} = (1) x scalar(@_);
    }
    else {
        %is_if_brace_follower = ();
    }

    # nothing can follow the closing curly of an else { } block:
    %is_else_brace_follower = ();

    # what can follow a multi-line anonymous sub definition closing curly:
    @_ = qw# ; : => or and  && || ~~ !~~ ) #;
    push @_, ',';
    @is_anon_sub_brace_follower{@_} = (1) x scalar(@_);

    # what can follow a one-line anonynomous sub closing curly:
    # one-line anonumous subs also have ']' here...
    # see tk3.t and PP.pm
    @_ = qw#  ; : => or and  && || ) ] ~~ !~~ #;
    push @_, ',';
    @is_anon_sub_1_brace_follower{@_} = (1) x scalar(@_);

    # What can follow a closing curly of a block
    # which is not an if/elsif/else/do/sort/map/grep/eval/sub
    # Testfiles: 'Toolbar.pm', 'Menubar.pm', bless.t, '3rules.pl'
    @_ = qw#  ; : => or and  && || ) #;
    push @_, ',';

    # allow cuddled continue if cuddled else is specified
    if ( $rOpts->{'cuddled-else'} ) { push @_, 'continue'; }

    @is_other_brace_follower{@_} = (1) x scalar(@_);

    $right_bond_strength{'{'} = WEAK;
    $left_bond_strength{'{'}  = VERY_STRONG;

    # make -l=0  equal to -l=infinite
    if ( !$rOpts->{'maximum-line-length'} ) {
        $rOpts->{'maximum-line-length'} = 1000000;
    }

    # make -lbl=0  equal to -lbl=infinite
    if ( !$rOpts->{'long-block-line-count'} ) {
        $rOpts->{'long-block-line-count'} = 1000000;
    }

    my $ole = $rOpts->{'output-line-ending'};
    if ($ole) {
        my %endings = (
            dos  => "\015\012",
            win  => "\015\012",
            mac  => "\015",
            unix => "\012",
        );
        $ole = lc $ole;
        unless ( $rOpts->{'output-line-ending'} = $endings{$ole} ) {
            my $str = join " ", keys %endings;
            Perl::Tidy::Die <<EOM;
Unrecognized line ending '$ole'; expecting one of: $str
EOM
        }
        if ( $rOpts->{'preserve-line-endings'} ) {
            Perl::Tidy::Warn "Ignoring -ple; conflicts with -ole\n";
            $rOpts->{'preserve-line-endings'} = undef;
        }
    }

    # hashes used to simplify setting whitespace
    %tightness = (
        '{' => $rOpts->{'brace-tightness'},
        '}' => $rOpts->{'brace-tightness'},
        '(' => $rOpts->{'paren-tightness'},
        ')' => $rOpts->{'paren-tightness'},
        '[' => $rOpts->{'square-bracket-tightness'},
        ']' => $rOpts->{'square-bracket-tightness'},
    );
    %matching_token = (
        '{' => '}',
        '(' => ')',
        '[' => ']',
        '?' => ':',
    );

    # frequently used parameters
    $rOpts_add_newlines          = $rOpts->{'add-newlines'};
    $rOpts_add_whitespace        = $rOpts->{'add-whitespace'};
    $rOpts_block_brace_tightness = $rOpts->{'block-brace-tightness'};
    $rOpts_block_brace_vertical_tightness =
      $rOpts->{'block-brace-vertical-tightness'};
    $rOpts_brace_left_and_indent   = $rOpts->{'brace-left-and-indent'};
    $rOpts_comma_arrow_breakpoints = $rOpts->{'comma-arrow-breakpoints'};
    $rOpts_break_at_old_ternary_breakpoints =
      $rOpts->{'break-at-old-ternary-breakpoints'};
    $rOpts_break_at_old_attribute_breakpoints =
      $rOpts->{'break-at-old-attribute-breakpoints'};
    $rOpts_break_at_old_comma_breakpoints =
      $rOpts->{'break-at-old-comma-breakpoints'};
    $rOpts_break_at_old_keyword_breakpoints =
      $rOpts->{'break-at-old-keyword-breakpoints'};
    $rOpts_break_at_old_logical_breakpoints =
      $rOpts->{'break-at-old-logical-breakpoints'};
    $rOpts_closing_side_comment_else_flag =
      $rOpts->{'closing-side-comment-else-flag'};
    $rOpts_closing_side_comment_maximum_text =
      $rOpts->{'closing-side-comment-maximum-text'};
    $rOpts_continuation_indentation = $rOpts->{'continuation-indentation'};
    $rOpts_cuddled_else             = $rOpts->{'cuddled-else'};
    $rOpts_delete_old_whitespace    = $rOpts->{'delete-old-whitespace'};
    $rOpts_fuzzy_line_length        = $rOpts->{'fuzzy-line-length'};
    $rOpts_indent_columns           = $rOpts->{'indent-columns'};
    $rOpts_line_up_parentheses      = $rOpts->{'line-up-parentheses'};
    $rOpts_maximum_fields_per_table = $rOpts->{'maximum-fields-per-table'};
    $rOpts_maximum_line_length      = $rOpts->{'maximum-line-length'};
    $rOpts_whitespace_cycle         = $rOpts->{'whitespace-cycle'};

    $rOpts_variable_maximum_line_length =
      $rOpts->{'variable-maximum-line-length'};
    $rOpts_short_concatenation_item_length =
      $rOpts->{'short-concatenation-item-length'};

    $rOpts_keep_old_blank_lines     = $rOpts->{'keep-old-blank-lines'};
    $rOpts_ignore_old_breakpoints   = $rOpts->{'ignore-old-breakpoints'};
    $rOpts_format_skipping          = $rOpts->{'format-skipping'};
    $rOpts_space_function_paren     = $rOpts->{'space-function-paren'};
    $rOpts_space_keyword_paren      = $rOpts->{'space-keyword-paren'};
    $rOpts_keep_interior_semicolons = $rOpts->{'keep-interior-semicolons'};
    $rOpts_ignore_side_comment_lengths =
      $rOpts->{'ignore-side-comment-lengths'};

    # Note that both opening and closing tokens can access the opening
    # and closing flags of their container types.
    %opening_vertical_tightness = (
        '(' => $rOpts->{'paren-vertical-tightness'},
        '{' => $rOpts->{'brace-vertical-tightness'},
        '[' => $rOpts->{'square-bracket-vertical-tightness'},
        ')' => $rOpts->{'paren-vertical-tightness'},
        '}' => $rOpts->{'brace-vertical-tightness'},
        ']' => $rOpts->{'square-bracket-vertical-tightness'},
    );

    %closing_vertical_tightness = (
        '(' => $rOpts->{'paren-vertical-tightness-closing'},
        '{' => $rOpts->{'brace-vertical-tightness-closing'},
        '[' => $rOpts->{'square-bracket-vertical-tightness-closing'},
        ')' => $rOpts->{'paren-vertical-tightness-closing'},
        '}' => $rOpts->{'brace-vertical-tightness-closing'},
        ']' => $rOpts->{'square-bracket-vertical-tightness-closing'},
    );

    $rOpts_tight_secret_operators = $rOpts->{'tight-secret-operators'};

    # assume flag for '>' same as ')' for closing qw quotes
    %closing_token_indentation = (
        ')' => $rOpts->{'closing-paren-indentation'},
        '}' => $rOpts->{'closing-brace-indentation'},
        ']' => $rOpts->{'closing-square-bracket-indentation'},
        '>' => $rOpts->{'closing-paren-indentation'},
    );

    # flag indicating if any closing tokens are indented
    $some_closing_token_indentation =
         $rOpts->{'closing-paren-indentation'}
      || $rOpts->{'closing-brace-indentation'}
      || $rOpts->{'closing-square-bracket-indentation'}
      || $rOpts->{'indent-closing-brace'};

    %opening_token_right = (
        '(' => $rOpts->{'opening-paren-right'},
        '{' => $rOpts->{'opening-hash-brace-right'},
        '[' => $rOpts->{'opening-square-bracket-right'},
    );

    %stack_opening_token = (
        '(' => $rOpts->{'stack-opening-paren'},
        '{' => $rOpts->{'stack-opening-hash-brace'},
        '[' => $rOpts->{'stack-opening-square-bracket'},
    );

    %stack_closing_token = (
        ')' => $rOpts->{'stack-closing-paren'},
        '}' => $rOpts->{'stack-closing-hash-brace'},
        ']' => $rOpts->{'stack-closing-square-bracket'},
    );
    $rOpts_stack_closing_block_brace = $rOpts->{'stack-closing-block-brace'};
}

sub make_static_block_comment_pattern {

    # create the pattern used to identify static block comments
    $static_block_comment_pattern = '^\s*##';

    # allow the user to change it
    if ( $rOpts->{'static-block-comment-prefix'} ) {
        my $prefix = $rOpts->{'static-block-comment-prefix'};
        $prefix =~ s/^\s*//;
        my $pattern = $prefix;

        # user may give leading caret to force matching left comments only
        if ( $prefix !~ /^\^#/ ) {
            if ( $prefix !~ /^#/ ) {
                Perl::Tidy::Die
"ERROR: the -sbcp prefix is '$prefix' but must begin with '#' or '^#'\n";
            }
            $pattern = '^\s*' . $prefix;
        }
        eval "'##'=~/$pattern/";
        if ($@) {
            Perl::Tidy::Die
"ERROR: the -sbc prefix '$prefix' causes the invalid regex '$pattern'\n";
        }
        $static_block_comment_pattern = $pattern;
    }
}

sub make_format_skipping_pattern {
    my ( $opt_name, $default ) = @_;
    my $param = $rOpts->{$opt_name};
    unless ($param) { $param = $default }
    $param =~ s/^\s*//;
    if ( $param !~ /^#/ ) {
        Perl::Tidy::Die
          "ERROR: the $opt_name parameter '$param' must begin with '#'\n";
    }
    my $pattern = '^' . $param . '\s';
    eval "'#'=~/$pattern/";
    if ($@) {
        Perl::Tidy::Die
"ERROR: the $opt_name parameter '$param' causes the invalid regex '$pattern'\n";
    }
    return $pattern;
}

sub make_closing_side_comment_list_pattern {

    # turn any input list into a regex for recognizing selected block types
    $closing_side_comment_list_pattern = '^\w+';
    if ( defined( $rOpts->{'closing-side-comment-list'} )
        && $rOpts->{'closing-side-comment-list'} )
    {
        $closing_side_comment_list_pattern =
          make_block_pattern( '-cscl', $rOpts->{'closing-side-comment-list'} );
    }
}

sub make_bli_pattern {

    if ( defined( $rOpts->{'brace-left-and-indent-list'} )
        && $rOpts->{'brace-left-and-indent-list'} )
    {
        $bli_list_string = $rOpts->{'brace-left-and-indent-list'};
    }

    $bli_pattern = make_block_pattern( '-blil', $bli_list_string );
}

sub make_block_brace_vertical_tightness_pattern {

    # turn any input list into a regex for recognizing selected block types
    $block_brace_vertical_tightness_pattern =
      '^((if|else|elsif|unless|while|for|foreach|do|\w+:)$|sub)';
    if ( defined( $rOpts->{'block-brace-vertical-tightness-list'} )
        && $rOpts->{'block-brace-vertical-tightness-list'} )
    {
        $block_brace_vertical_tightness_pattern =
          make_block_pattern( '-bbvtl',
            $rOpts->{'block-brace-vertical-tightness-list'} );
    }
}

sub make_block_pattern {

    #  given a string of block-type keywords, return a regex to match them
    #  The only tricky part is that labels are indicated with a single ':'
    #  and the 'sub' token text may have additional text after it (name of
    #  sub).
    #
    #  Example:
    #
    #   input string: "if else elsif unless while for foreach do : sub";
    #   pattern:  '^((if|else|elsif|unless|while|for|foreach|do|\w+:)$|sub)';

    my ( $abbrev, $string ) = @_;
    my @list  = split_words($string);
    my @words = ();
    my %seen;
    for my $i (@list) {
        if ( $i eq '*' ) { my $pattern = '^.*'; return $pattern }
        next if $seen{$i};
        $seen{$i} = 1;
        if ( $i eq 'sub' ) {
        }
        elsif ( $i eq ';' ) {
            push @words, ';';
        }
        elsif ( $i eq '{' ) {
            push @words, '\{';
        }
        elsif ( $i eq ':' ) {
            push @words, '\w+:';
        }
        elsif ( $i =~ /^\w/ ) {
            push @words, $i;
        }
        else {
            Perl::Tidy::Warn
              "unrecognized block type $i after $abbrev, ignoring\n";
        }
    }
    my $pattern = '(' . join( '|', @words ) . ')$';
    if ( $seen{'sub'} ) {
        $pattern = '(' . $pattern . '|sub)';
    }
    $pattern = '^' . $pattern;
    return $pattern;
}

sub make_static_side_comment_pattern {

    # create the pattern used to identify static side comments
    $static_side_comment_pattern = '^##';

    # allow the user to change it
    if ( $rOpts->{'static-side-comment-prefix'} ) {
        my $prefix = $rOpts->{'static-side-comment-prefix'};
        $prefix =~ s/^\s*//;
        my $pattern = '^' . $prefix;
        eval "'##'=~/$pattern/";
        if ($@) {
            Perl::Tidy::Die
"ERROR: the -sscp prefix '$prefix' causes the invalid regex '$pattern'\n";
        }
        $static_side_comment_pattern = $pattern;
    }
}

sub make_closing_side_comment_prefix {

    # Be sure we have a valid closing side comment prefix
    my $csc_prefix = $rOpts->{'closing-side-comment-prefix'};
    my $csc_prefix_pattern;
    if ( !defined($csc_prefix) ) {
        $csc_prefix         = '## end';
        $csc_prefix_pattern = '^##\s+end';
    }
    else {
        my $test_csc_prefix = $csc_prefix;
        if ( $test_csc_prefix !~ /^#/ ) {
            $test_csc_prefix = '#' . $test_csc_prefix;
        }

        # make a regex to recognize the prefix
        my $test_csc_prefix_pattern = $test_csc_prefix;

        # escape any special characters
        $test_csc_prefix_pattern =~ s/([^#\s\w])/\\$1/g;

        $test_csc_prefix_pattern = '^' . $test_csc_prefix_pattern;

        # allow exact number of intermediate spaces to vary
        $test_csc_prefix_pattern =~ s/\s+/\\s\+/g;

        # make sure we have a good pattern
        # if we fail this we probably have an error in escaping
        # characters.
        eval "'##'=~/$test_csc_prefix_pattern/";
        if ($@) {

            # shouldn't happen..must have screwed up escaping, above
            report_definite_bug();
            Perl::Tidy::Warn
"Program Error: the -cscp prefix '$csc_prefix' caused the invalid regex '$csc_prefix_pattern'\n";

            # just warn and keep going with defaults
            Perl::Tidy::Warn "Please consider using a simpler -cscp prefix\n";
            Perl::Tidy::Warn
              "Using default -cscp instead; please check output\n";
        }
        else {
            $csc_prefix         = $test_csc_prefix;
            $csc_prefix_pattern = $test_csc_prefix_pattern;
        }
    }
    $rOpts->{'closing-side-comment-prefix'} = $csc_prefix;
    $closing_side_comment_prefix_pattern = $csc_prefix_pattern;
}

sub dump_want_left_space {
    my $fh = shift;
    local $" = "\n";
    print $fh <<EOM;
These values are the main control of whitespace to the left of a token type;
They may be altered with the -wls parameter.
For a list of token types, use perltidy --dump-token-types (-dtt)
 1 means the token wants a space to its left
-1 means the token does not want a space to its left
------------------------------------------------------------------------
EOM
    foreach ( sort keys %want_left_space ) {
        print $fh "$_\t$want_left_space{$_}\n";
    }
}

sub dump_want_right_space {
    my $fh = shift;
    local $" = "\n";
    print $fh <<EOM;
These values are the main control of whitespace to the right of a token type;
They may be altered with the -wrs parameter.
For a list of token types, use perltidy --dump-token-types (-dtt)
 1 means the token wants a space to its right
-1 means the token does not want a space to its right
------------------------------------------------------------------------
EOM
    foreach ( sort keys %want_right_space ) {
        print $fh "$_\t$want_right_space{$_}\n";
    }
}

{    # begin is_essential_whitespace

    my %is_sort_grep_map;
    my %is_for_foreach;

    BEGIN {

        @_ = qw(sort grep map);
        @is_sort_grep_map{@_} = (1) x scalar(@_);

        @_ = qw(for foreach);
        @is_for_foreach{@_} = (1) x scalar(@_);

    }

    sub is_essential_whitespace {

        # Essential whitespace means whitespace which cannot be safely deleted
        # without risking the introduction of a syntax error.
        # We are given three tokens and their types:
        # ($tokenl, $typel) is the token to the left of the space in question
        # ($tokenr, $typer) is the token to the right of the space in question
        # ($tokenll, $typell) is previous nonblank token to the left of $tokenl
        #
        # This is a slow routine but is not needed too often except when -mangle
        # is used.
        #
        # Note: This routine should almost never need to be changed.  It is
        # for avoiding syntax problems rather than for formatting.
        my ( $tokenll, $typell, $tokenl, $typel, $tokenr, $typer ) = @_;

        my $result =

          # never combine two bare words or numbers
          # examples:  and ::ok(1)
          #            return ::spw(...)
          #            for bla::bla:: abc
          # example is "%overload:: and" in files Dumpvalue.pm or colonbug.pl
          #            $input eq"quit" to make $inputeq"quit"
          #            my $size=-s::SINK if $file;  <==OK but we won't do it
          # don't join something like: for bla::bla:: abc
          # example is "%overload:: and" in files Dumpvalue.pm or colonbug.pl
          (      ( $tokenl =~ /([\'\w]|\:\:)$/ && $typel ne 'CORE::' )
              && ( $tokenr =~ /^([\'\w]|\:\:)/ ) )

          # do not combine a number with a concatination dot
          # example: pom.caputo:
          # $vt100_compatible ? "\e[0;0H" : ('-' x 78 . "\n");
          || ( ( $typel eq 'n' ) && ( $tokenr eq '.' ) )
          || ( ( $typer eq 'n' ) && ( $tokenl eq '.' ) )

          # do not join a minus with a bare word, because you might form
          # a file test operator.  Example from Complex.pm:
          # if (CORE::abs($z - i) < $eps); "z-i" would be taken as a file test.
          || ( ( $tokenl eq '-' ) && ( $tokenr =~ /^[_A-Za-z]$/ ) )

          # and something like this could become ambiguous without space
          # after the '-':
          #   use constant III=>1;
          #   $a = $b - III;
          # and even this:
          #   $a = - III;
          || ( ( $tokenl eq '-' )
            && ( $typer =~ /^[wC]$/ && $tokenr =~ /^[_A-Za-z]/ ) )

          # '= -' should not become =- or you will get a warning
          # about reversed -=
          # || ($tokenr eq '-')

          # keep a space between a quote and a bareword to prevent the
          # bareword from becoming a quote modifier.
          || ( ( $typel eq 'Q' ) && ( $tokenr =~ /^[a-zA-Z_]/ ) )

          # keep a space between a token ending in '$' and any word;
          # this caused trouble:  "die @$ if $@"
          || ( ( $typel eq 'i' && $tokenl =~ /\$$/ )
            && ( $tokenr =~ /^[a-zA-Z_]/ ) )

          # perl is very fussy about spaces before <<
          || ( $tokenr =~ /^\<\</ )

          # avoid combining tokens to create new meanings. Example:
          #     $a+ +$b must not become $a++$b
          || ( $is_digraph{ $tokenl . $tokenr } )
          || ( $is_trigraph{ $tokenl . $tokenr } )

          # another example: do not combine these two &'s:
          #     allow_options & &OPT_EXECCGI
          || ( $is_digraph{ $tokenl . substr( $tokenr, 0, 1 ) } )

          # don't combine $$ or $# with any alphanumeric
          # (testfile mangle.t with --mangle)
          || ( ( $tokenl =~ /^\$[\$\#]$/ ) && ( $tokenr =~ /^\w/ ) )

          # retain any space after possible filehandle
          # (testfiles prnterr1.t with --extrude and mangle.t with --mangle)
          || ( $typel eq 'Z' )

          # Perl is sensitive to whitespace after the + here:
          #  $b = xvals $a + 0.1 * yvals $a;
          || ( $typell eq 'Z' && $typel =~ /^[\/\?\+\-\*]$/ )

          # keep paren separate in 'use Foo::Bar ()'
          || ( $tokenr eq '('
            && $typel eq 'w'
            && $typell eq 'k'
            && $tokenll eq 'use' )

          # keep any space between filehandle and paren:
          # file mangle.t with --mangle:
          || ( $typel eq 'Y' && $tokenr eq '(' )

          # retain any space after here doc operator ( hereerr.t)
          || ( $typel eq 'h' )

          # be careful with a space around ++ and --, to avoid ambiguity as to
          # which token it applies
          || ( ( $typer =~ /^(pp|mm)$/ )     && ( $tokenl !~ /^[\;\{\(\[]/ ) )
          || ( ( $typel =~ /^(\+\+|\-\-)$/ ) && ( $tokenr !~ /^[\;\}\)\]]/ ) )

          # need space after foreach my; for example, this will fail in
          # older versions of Perl:
          # foreach my$ft(@filetypes)...
          || (
            $tokenl eq 'my'

            #  /^(for|foreach)$/
            && $is_for_foreach{$tokenll}
            && $tokenr =~ /^\$/
          )

          # must have space between grep and left paren; "grep(" will fail
          || ( $tokenr eq '(' && $is_sort_grep_map{$tokenl} )

          # don't stick numbers next to left parens, as in:
          #use Mail::Internet 1.28 (); (see Entity.pm, Head.pm, Test.pm)
          || ( ( $typel eq 'n' ) && ( $tokenr eq '(' ) )

          # We must be sure that a space between a ? and a quoted string
          # remains if the space before the ? remains.  [Loca.pm, lockarea]
          # ie,
          #    $b=join $comma ? ',' : ':', @_;  # ok
          #    $b=join $comma?',' : ':', @_;    # ok!
          #    $b=join $comma ?',' : ':', @_;   # error!
          # Not really required:
          ## || ( ( $typel eq '?' ) && ( $typer eq 'Q' ) )

          # do not remove space between an '&' and a bare word because
          # it may turn into a function evaluation, like here
          # between '&' and 'O_ACCMODE', producing a syntax error [File.pm]
          #    $opts{rdonly} = (($opts{mode} & O_ACCMODE) == O_RDONLY);
          || ( ( $typel eq '&' ) && ( $tokenr =~ /^[a-zA-Z_]/ ) )

          # space stacked labels  (TODO: check if really necessary)
          || ( $typel eq 'J' && $typer eq 'J' )

          ;    # the value of this long logic sequence is the result we want
        return $result;
    }
}

{
    my %secret_operators;
    my %is_leading_secret_token;

    BEGIN {

        # token lists for perl secret operators as compiled by Philippe Bruhat
        # at: https://metacpan.org/module/perlsecret
        %secret_operators = (
            'Goatse'            => [qw#= ( ) =#],        #=( )=
            'Venus1'            => [qw#0 +#],            # 0+
            'Venus2'            => [qw#+ 0#],            # +0
            'Enterprise'        => [qw#) x ! !#],        # ()x!!
            'Kite1'             => [qw#~ ~ <>#],         # ~~<>
            'Kite2'             => [qw#~~ <>#],          # ~~<>
            'Winking Fat Comma' => [ ( ',', '=>' ) ],    # ,=>
        );

        # The following operators and constants are not included because they
        # are normally kept tight by perltidy:
        # !!  ~~ <~>
        #

        # Make a lookup table indexed by the first token of each operator:
        # first token => [list, list, ...]
        foreach my $value ( values(%secret_operators) ) {
            my $tok = $value->[0];
            push @{ $is_leading_secret_token{$tok} }, $value;
        }
    }

    sub secret_operator_whitespace {

        my ( $jmax, $rtokens, $rtoken_type, $rwhite_space_flag ) = @_;

        # Loop over all tokens in this line
        my ( $j, $token, $type );
        for ( $j = 0 ; $j <= $jmax ; $j++ ) {

            $token = $$rtokens[$j];
            $type  = $$rtoken_type[$j];

            # Skip unless this token might start a secret operator
            next if ( $type eq 'b' );
            next unless ( $is_leading_secret_token{$token} );

            #      Loop over all secret operators with this leading token
            foreach my $rpattern ( @{ $is_leading_secret_token{$token} } ) {
                my $jend = $j - 1;
                foreach my $tok ( @{$rpattern} ) {
                    $jend++;
                    $jend++

                      if ( $jend <= $jmax && $$rtoken_type[$jend] eq 'b' );
                    if ( $jend > $jmax || $tok ne $$rtokens[$jend] ) {
                        $jend = undef;
                        last;
                    }
                }

                if ($jend) {

                    # set flags to prevent spaces within this operator
                    for ( my $jj = $j + 1 ; $jj <= $jend ; $jj++ ) {
                        $rwhite_space_flag->[$jj] = WS_NO;
                    }
                    $j = $jend;
                    last;
                }
            }    ##      End Loop over all operators
        }    ## End loop over all tokens
    }    # End sub
}

sub set_white_space_flag {

    #    This routine examines each pair of nonblank tokens and
    #    sets values for array @white_space_flag.
    #
    #    $white_space_flag[$j] is a flag indicating whether a white space
    #    BEFORE token $j is needed, with the following values:
    #
    #             WS_NO      = -1 do not want a space before token $j
    #             WS_OPTIONAL=  0 optional space or $j is a whitespace
    #             WS_YES     =  1 want a space before token $j
    #
    #
    #   The values for the first token will be defined based
    #   upon the contents of the "to_go" output array.
    #
    #   Note: retain debug print statements because they are usually
    #   required after adding new token types.

    BEGIN {

        # initialize these global hashes, which control the use of
        # whitespace around tokens:
        #
        # %binary_ws_rules
        # %want_left_space
        # %want_right_space
        # %space_after_keyword
        #
        # Many token types are identical to the tokens themselves.
        # See the tokenizer for a complete list. Here are some special types:
        #   k = perl keyword
        #   f = semicolon in for statement
        #   m = unary minus
        #   p = unary plus
        # Note that :: is excluded since it should be contained in an identifier
        # Note that '->' is excluded because it never gets space
        # parentheses and brackets are excluded since they are handled specially
        # curly braces are included but may be overridden by logic, such as
        # newline logic.

        # NEW_TOKENS: create a whitespace rule here.  This can be as
        # simple as adding your new letter to @spaces_both_sides, for
        # example.

        @_ = qw" L { ( [ ";
        @is_opening_type{@_} = (1) x scalar(@_);

        @_ = qw" R } ) ] ";
        @is_closing_type{@_} = (1) x scalar(@_);

        my @spaces_both_sides = qw"
          + - * / % ? = . : x < > | & ^ .. << >> ** && .. || // => += -=
          .= %= x= &= |= ^= *= <> <= >= == =~ !~ /= != ... <<= >>= ~~ !~~
          &&= ||= //= <=> A k f w F n C Y U G v
          ";

        my @spaces_left_side = qw"
          t ! ~ m p { \ h pp mm Z j
          ";
        push( @spaces_left_side, '#' );    # avoids warning message

        my @spaces_right_side = qw"
          ; } ) ] R J ++ -- **=
          ";
        push( @spaces_right_side, ',' );    # avoids warning message

        # Note that we are in a BEGIN block here.  Later in processing
        # the values of %want_left_space and  %want_right_space
        # may be overridden by any user settings specified by the
        # -wls and -wrs parameters.  However the binary_whitespace_rules
        # are hardwired and have priority.
        @want_left_space{@spaces_both_sides} = (1) x scalar(@spaces_both_sides);
        @want_right_space{@spaces_both_sides} =
          (1) x scalar(@spaces_both_sides);
        @want_left_space{@spaces_left_side}  = (1) x scalar(@spaces_left_side);
        @want_right_space{@spaces_left_side} = (-1) x scalar(@spaces_left_side);
        @want_left_space{@spaces_right_side} =
          (-1) x scalar(@spaces_right_side);
        @want_right_space{@spaces_right_side} =
          (1) x scalar(@spaces_right_side);
        $want_left_space{'->'}      = WS_NO;
        $want_right_space{'->'}     = WS_NO;
        $want_left_space{'**'}      = WS_NO;
        $want_right_space{'**'}     = WS_NO;
        $want_right_space{'CORE::'} = WS_NO;

        # These binary_ws_rules are hardwired and have priority over the above
        # settings.  It would be nice to allow adjustment by the user,
        # but it would be complicated to specify.
        #
        # hash type information must stay tightly bound
        # as in :  ${xxxx}
        $binary_ws_rules{'i'}{'L'} = WS_NO;
        $binary_ws_rules{'i'}{'{'} = WS_YES;
        $binary_ws_rules{'k'}{'{'} = WS_YES;
        $binary_ws_rules{'U'}{'{'} = WS_YES;
        $binary_ws_rules{'i'}{'['} = WS_NO;
        $binary_ws_rules{'R'}{'L'} = WS_NO;
        $binary_ws_rules{'R'}{'{'} = WS_NO;
        $binary_ws_rules{'t'}{'L'} = WS_NO;
        $binary_ws_rules{'t'}{'{'} = WS_NO;
        $binary_ws_rules{'}'}{'L'} = WS_NO;
        $binary_ws_rules{'}'}{'{'} = WS_NO;
        $binary_ws_rules{'$'}{'L'} = WS_NO;
        $binary_ws_rules{'$'}{'{'} = WS_NO;
        $binary_ws_rules{'@'}{'L'} = WS_NO;
        $binary_ws_rules{'@'}{'{'} = WS_NO;
        $binary_ws_rules{'='}{'L'} = WS_YES;
        $binary_ws_rules{'J'}{'J'} = WS_YES;

        # the following includes ') {'
        # as in :    if ( xxx ) { yyy }
        $binary_ws_rules{']'}{'L'} = WS_NO;
        $binary_ws_rules{']'}{'{'} = WS_NO;
        $binary_ws_rules{')'}{'{'} = WS_YES;
        $binary_ws_rules{')'}{'['} = WS_NO;
        $binary_ws_rules{']'}{'['} = WS_NO;
        $binary_ws_rules{']'}{'{'} = WS_NO;
        $binary_ws_rules{'}'}{'['} = WS_NO;
        $binary_ws_rules{'R'}{'['} = WS_NO;

        $binary_ws_rules{']'}{'++'} = WS_NO;
        $binary_ws_rules{']'}{'--'} = WS_NO;
        $binary_ws_rules{')'}{'++'} = WS_NO;
        $binary_ws_rules{')'}{'--'} = WS_NO;

        $binary_ws_rules{'R'}{'++'} = WS_NO;
        $binary_ws_rules{'R'}{'--'} = WS_NO;

        $binary_ws_rules{'i'}{'Q'} = WS_YES;
        $binary_ws_rules{'n'}{'('} = WS_YES;    # occurs in 'use package n ()'

        # FIXME: we could to split 'i' into variables and functions
        # and have no space for functions but space for variables.  For now,
        # I have a special patch in the special rules below
        $binary_ws_rules{'i'}{'('} = WS_NO;

        $binary_ws_rules{'w'}{'('} = WS_NO;
        $binary_ws_rules{'w'}{'{'} = WS_YES;
    } ## end BEGIN block

    my ( $jmax, $rtokens, $rtoken_type, $rblock_type ) = @_;
    my ( $last_token, $last_type, $last_block_type, $token, $type,
        $block_type );
    my (@white_space_flag);
    my $j_tight_closing_paren = -1;

    if ( $max_index_to_go >= 0 ) {
        $token      = $tokens_to_go[$max_index_to_go];
        $type       = $types_to_go[$max_index_to_go];
        $block_type = $block_type_to_go[$max_index_to_go];

        #---------------------------------------------------------------
        # Patch due to splitting of tokens with leading ->
        #---------------------------------------------------------------
        #
        # This routine is dealing with the raw tokens from the tokenizer,
        # but to get started it needs the previous token, which will
        # have been stored in the '_to_go' arrays.
        #
        # This patch avoids requiring two iterations to
        # converge for cases such as the following, where a paren
        # comes in on a line following a variable with leading arrow:
        #     $self->{main}->add_content_defer_opening
        #                         ($name, $wmkf, $self->{attrs}, $self);
        # In this case when we see the opening paren on line 2 we need
        # to know if the last token on the previous line had an arrow,
        # but it has already been split off so we have to add it back
        # in to avoid getting an unwanted space before the paren.
        if ( $type =~ /^[wi]$/ ) {
            my $im = $iprev_to_go[$max_index_to_go];
            my $tm = ( $im >= 0 ) ? $types_to_go[$im] : "";
            if ( $tm eq '->' ) { $token = $tm . $token }
        }

        #---------------------------------------------------------------
        # End patch due to splitting of tokens with leading ->
        #---------------------------------------------------------------
    }
    else {
        $token      = ' ';
        $type       = 'b';
        $block_type = '';
    }

    my ( $j, $ws );

    # main loop over all tokens to define the whitespace flags
    for ( $j = 0 ; $j <= $jmax ; $j++ ) {

        if ( $$rtoken_type[$j] eq 'b' ) {
            $white_space_flag[$j] = WS_OPTIONAL;
            next;
        }

        # set a default value, to be changed as needed
        $ws              = undef;
        $last_token      = $token;
        $last_type       = $type;
        $last_block_type = $block_type;
        $token           = $$rtokens[$j];
        $type            = $$rtoken_type[$j];
        $block_type      = $$rblock_type[$j];

        #---------------------------------------------------------------
        # Whitespace Rules Section 1:
        # Handle space on the inside of opening braces.
        #---------------------------------------------------------------

        #    /^[L\{\(\[]$/
        if ( $is_opening_type{$last_type} ) {

            $j_tight_closing_paren = -1;

            # let's keep empty matched braces together: () {} []
            # except for BLOCKS
            if ( $token eq $matching_token{$last_token} ) {
                if ($block_type) {
                    $ws = WS_YES;
                }
                else {
                    $ws = WS_NO;
                }
            }
            else {

                # we're considering the right of an opening brace
                # tightness = 0 means always pad inside with space
                # tightness = 1 means pad inside if "complex"
                # tightness = 2 means never pad inside with space

                my $tightness;
                if (   $last_type eq '{'
                    && $last_token eq '{'
                    && $last_block_type )
                {
                    $tightness = $rOpts_block_brace_tightness;
                }
                else { $tightness = $tightness{$last_token} }

               #=============================================================
               # Patch for test problem fabrice_bug.pl
               # We must always avoid spaces around a bare word beginning
               # with ^ as in:
               #    my $before = ${^PREMATCH};
               # Because all of the following cause an error in perl:
               #    my $before = ${ ^PREMATCH };
               #    my $before = ${ ^PREMATCH};
               #    my $before = ${^PREMATCH };
               # So if brace tightness flag is -bt=0 we must temporarily reset
               # to bt=1.  Note that here we must set tightness=1 and not 2 so
               # that the closing space
               # is also avoided (via the $j_tight_closing_paren flag in coding)
                if ( $type eq 'w' && $token =~ /^\^/ ) { $tightness = 1 }

                #=============================================================

                if ( $tightness <= 0 ) {
                    $ws = WS_YES;
                }
                elsif ( $tightness > 1 ) {
                    $ws = WS_NO;
                }
                else {

                    # Patch to count '-foo' as single token so that
                    # each of  $a{-foo} and $a{foo} and $a{'foo'} do
                    # not get spaces with default formatting.
                    my $j_here = $j;
                    ++$j_here
                      if ( $token eq '-'
                        && $last_token eq '{'
                        && $$rtoken_type[ $j + 1 ] eq 'w' );

                    # $j_next is where a closing token should be if
                    # the container has a single token
                    my $j_next =
                      ( $$rtoken_type[ $j_here + 1 ] eq 'b' )
                      ? $j_here + 2
                      : $j_here + 1;
                    my $tok_next  = $$rtokens[$j_next];
                    my $type_next = $$rtoken_type[$j_next];

                    # for tightness = 1, if there is just one token
                    # within the matching pair, we will keep it tight
                    if (
                        $tok_next eq $matching_token{$last_token}

                        # but watch out for this: [ [ ]    (misc.t)
                        && $last_token ne $token
                      )
                    {

                        # remember where to put the space for the closing paren
                        $j_tight_closing_paren = $j_next;
                        $ws                    = WS_NO;
                    }
                    else {
                        $ws = WS_YES;
                    }
                }
            }
        }    # end setting space flag inside opening tokens
        my $ws_1 = $ws
          if FORMATTER_DEBUG_FLAG_WHITE;

        #---------------------------------------------------------------
        # Whitespace Rules Section 2:
        # Handle space on inside of closing brace pairs.
        #---------------------------------------------------------------

        #   /[\}\)\]R]/
        if ( $is_closing_type{$type} ) {

            if ( $j == $j_tight_closing_paren ) {

                $j_tight_closing_paren = -1;
                $ws                    = WS_NO;
            }
            else {

                if ( !defined($ws) ) {

                    my $tightness;
                    if ( $type eq '}' && $token eq '}' && $block_type ) {
                        $tightness = $rOpts_block_brace_tightness;
                    }
                    else { $tightness = $tightness{$token} }

                    $ws = ( $tightness > 1 ) ? WS_NO : WS_YES;
                }
            }
        }    # end setting space flag inside closing tokens

        my $ws_2 = $ws
          if FORMATTER_DEBUG_FLAG_WHITE;

        #---------------------------------------------------------------
        # Whitespace Rules Section 3:
        # Use the binary rule table.
        #---------------------------------------------------------------
        if ( !defined($ws) ) {
            $ws = $binary_ws_rules{$last_type}{$type};
        }
        my $ws_3 = $ws
          if FORMATTER_DEBUG_FLAG_WHITE;

        #---------------------------------------------------------------
        # Whitespace Rules Section 4:
        # Handle some special cases.
        #---------------------------------------------------------------
        if ( $token eq '(' ) {

            # This will have to be tweaked as tokenization changes.
            # We usually want a space at '} (', for example:
            #     map { 1 * $_; } ( $y, $M, $w, $d, $h, $m, $s );
            #
            # But not others:
            #     &{ $_->[1] }( delete $_[$#_]{ $_->[0] } );
            # At present, the above & block is marked as type L/R so this case
            # won't go through here.
            if ( $last_type eq '}' ) { $ws = WS_YES }

            # NOTE: some older versions of Perl had occasional problems if
            # spaces are introduced between keywords or functions and opening
            # parens.  So the default is not to do this except is certain
            # cases.  The current Perl seems to tolerate spaces.

            # Space between keyword and '('
            elsif ( $last_type eq 'k' ) {
                $ws = WS_NO
                  unless ( $rOpts_space_keyword_paren
                    || $space_after_keyword{$last_token} );
            }

            # Space between function and '('
            # -----------------------------------------------------
            # 'w' and 'i' checks for something like:
            #   myfun(    &myfun(   ->myfun(
            # -----------------------------------------------------
            elsif (( $last_type =~ /^[wUG]$/ )
                || ( $last_type =~ /^[wi]$/ && $last_token =~ /^(\&|->)/ ) )
            {
                $ws = WS_NO unless ($rOpts_space_function_paren);
            }

            # space between something like $i and ( in
            # for $i ( 0 .. 20 ) {
            # FIXME: eventually, type 'i' needs to be split into multiple
            # token types so this can be a hardwired rule.
            elsif ( $last_type eq 'i' && $last_token =~ /^[\$\%\@]/ ) {
                $ws = WS_YES;
            }

            # allow constant function followed by '()' to retain no space
            elsif ( $last_type eq 'C' && $$rtokens[ $j + 1 ] eq ')' ) {
                $ws = WS_NO;
            }
        }

        # patch for SWITCH/CASE: make space at ']{' optional
        # since the '{' might begin a case or when block
        elsif ( ( $token eq '{' && $type ne 'L' ) && $last_token eq ']' ) {
            $ws = WS_OPTIONAL;
        }

        # keep space between 'sub' and '{' for anonymous sub definition
        if ( $type eq '{' ) {
            if ( $last_token eq 'sub' ) {
                $ws = WS_YES;
            }

            # this is needed to avoid no space in '){'
            if ( $last_token eq ')' && $token eq '{' ) { $ws = WS_YES }

            # avoid any space before the brace or bracket in something like
            #  @opts{'a','b',...}
            if ( $last_type eq 'i' && $last_token =~ /^\@/ ) {
                $ws = WS_NO;
            }
        }

        elsif ( $type eq 'i' ) {

            # never a space before ->
            if ( $token =~ /^\-\>/ ) {
                $ws = WS_NO;
            }
        }

        # retain any space between '-' and bare word
        elsif ( $type eq 'w' || $type eq 'C' ) {
            $ws = WS_OPTIONAL if $last_type eq '-';

            # never a space before ->
            if ( $token =~ /^\-\>/ ) {
                $ws = WS_NO;
            }
        }

        # retain any space between '-' and bare word
        # example: avoid space between 'USER' and '-' here:
        #   $myhash{USER-NAME}='steve';
        elsif ( $type eq 'm' || $type eq '-' ) {
            $ws = WS_OPTIONAL if ( $last_type eq 'w' );
        }

        # always space before side comment
        elsif ( $type eq '#' ) { $ws = WS_YES if $j > 0 }

        # always preserver whatever space was used after a possible
        # filehandle (except _) or here doc operator
        if (
            $type ne '#'
            && ( ( $last_type eq 'Z' && $last_token ne '_' )
                || $last_type eq 'h' )
          )
        {
            $ws = WS_OPTIONAL;
        }

        my $ws_4 = $ws
          if FORMATTER_DEBUG_FLAG_WHITE;

        #---------------------------------------------------------------
        # Whitespace Rules Section 5:
        # Apply default rules not covered above.
        #---------------------------------------------------------------

        # If we fall through to here, look at the pre-defined hash tables for
        # the two tokens, and:
        #  if (they are equal) use the common value
        #  if (either is zero or undef) use the other
        #  if (either is -1) use it
        # That is,
        # left  vs right
        #  1    vs    1     -->  1
        #  0    vs    0     -->  0
        # -1    vs   -1     --> -1
        #
        #  0    vs   -1     --> -1
        #  0    vs    1     -->  1
        #  1    vs    0     -->  1
        # -1    vs    0     --> -1
        #
        # -1    vs    1     --> -1
        #  1    vs   -1     --> -1
        if ( !defined($ws) ) {
            my $wl = $want_left_space{$type};
            my $wr = $want_right_space{$last_type};
            if ( !defined($wl) ) { $wl = 0 }
            if ( !defined($wr) ) { $wr = 0 }
            $ws = ( ( $wl == $wr ) || ( $wl == -1 ) || !$wr ) ? $wl : $wr;
        }

        if ( !defined($ws) ) {
            $ws = 0;
            write_diagnostics(
                "WS flag is undefined for tokens $last_token $token\n");
        }

        # Treat newline as a whitespace. Otherwise, we might combine
        # 'Send' and '-recipients' here according to the above rules:
        #    my $msg = new Fax::Send
        #      -recipients => $to,
        #      -data => $data;
        if ( $ws == 0 && $j == 0 ) { $ws = 1 }

        if (   ( $ws == 0 )
            && $j > 0
            && $j < $jmax
            && ( $last_type !~ /^[Zh]$/ ) )
        {

            # If this happens, we have a non-fatal but undesirable
            # hole in the above rules which should be patched.
            write_diagnostics(
                "WS flag is zero for tokens $last_token $token\n");
        }
        $white_space_flag[$j] = $ws;

        FORMATTER_DEBUG_FLAG_WHITE && do {
            my $str = substr( $last_token, 0, 15 );
            $str .= ' ' x ( 16 - length($str) );
            if ( !defined($ws_1) ) { $ws_1 = "*" }
            if ( !defined($ws_2) ) { $ws_2 = "*" }
            if ( !defined($ws_3) ) { $ws_3 = "*" }
            if ( !defined($ws_4) ) { $ws_4 = "*" }
            print STDOUT
"WHITE:  i=$j $str $last_type $type $ws_1 : $ws_2 : $ws_3 : $ws_4 : $ws \n";
        };
    } ## end main loop

    if ($rOpts_tight_secret_operators) {
        secret_operator_whitespace( $jmax, $rtokens, $rtoken_type,
            \@white_space_flag );
    }

    return \@white_space_flag;
} ## end sub set_white_space_flag

{    # begin print_line_of_tokens

    my $rtoken_type;
    my $rtokens;
    my $rlevels;
    my $rslevels;
    my $rblock_type;
    my $rcontainer_type;
    my $rcontainer_environment;
    my $rtype_sequence;
    my $input_line;
    my $rnesting_tokens;
    my $rci_levels;
    my $rnesting_blocks;

    my $in_quote;
    my $guessed_indentation_level;

    # These local token variables are stored by store_token_to_go:
    my $block_type;
    my $ci_level;
    my $container_environment;
    my $container_type;
    my $in_continued_quote;
    my $level;
    my $nesting_blocks;
    my $no_internal_newlines;
    my $slevel;
    my $token;
    my $type;
    my $type_sequence;

    # routine to pull the jth token from the line of tokens
    sub extract_token {
        my $j = shift;
        $token                 = $$rtokens[$j];
        $type                  = $$rtoken_type[$j];
        $block_type            = $$rblock_type[$j];
        $container_type        = $$rcontainer_type[$j];
        $container_environment = $$rcontainer_environment[$j];
        $type_sequence         = $$rtype_sequence[$j];
        $level                 = $$rlevels[$j];
        $slevel                = $$rslevels[$j];
        $nesting_blocks        = $$rnesting_blocks[$j];
        $ci_level              = $$rci_levels[$j];
    }

    {
        my @saved_token;

        sub save_current_token {

            @saved_token = (
                $block_type,            $ci_level,
                $container_environment, $container_type,
                $in_continued_quote,    $level,
                $nesting_blocks,        $no_internal_newlines,
                $slevel,                $token,
                $type,                  $type_sequence,
            );
        }

        sub restore_current_token {
            (
                $block_type,            $ci_level,
                $container_environment, $container_type,
                $in_continued_quote,    $level,
                $nesting_blocks,        $no_internal_newlines,
                $slevel,                $token,
                $type,                  $type_sequence,
            ) = @saved_token;
        }
    }

    sub token_length {

        # Returns the length of a token, given:
        #  $token=text of the token
        #  $type = type
        #  $not_first_token = should be TRUE if this is not the first token of
        #   the line.  It might the index of this token in an array.  It is
        #   used to test for a side comment vs a block commment.
        # Note: Eventually this should be the only routine determining the
        # length of a token in this package.
        my ( $token, $type, $not_first_token ) = @_;
        my $token_length = length($token);

        # We mark lengths of side comments as just 1 if we are
        # ignoring their lengths when setting line breaks.
        $token_length = 1
          if ( $rOpts_ignore_side_comment_lengths
            && $not_first_token
            && $type eq '#' );
        return $token_length;
    }

    sub rtoken_length {

        # return length of ith token in @{$rtokens}
        my ($i) = @_;
        return token_length( $$rtokens[$i], $$rtoken_type[$i], $i );
    }

    # Routine to place the current token into the output stream.
    # Called once per output token.
    sub store_token_to_go {

        my $flag = $no_internal_newlines;
        if ( $_[0] ) { $flag = 1 }

        $tokens_to_go[ ++$max_index_to_go ]            = $token;
        $types_to_go[$max_index_to_go]                 = $type;
        $nobreak_to_go[$max_index_to_go]               = $flag;
        $old_breakpoint_to_go[$max_index_to_go]        = 0;
        $forced_breakpoint_to_go[$max_index_to_go]     = 0;
        $block_type_to_go[$max_index_to_go]            = $block_type;
        $type_sequence_to_go[$max_index_to_go]         = $type_sequence;
        $container_environment_to_go[$max_index_to_go] = $container_environment;
        $nesting_blocks_to_go[$max_index_to_go]        = $nesting_blocks;
        $ci_levels_to_go[$max_index_to_go]             = $ci_level;
        $mate_index_to_go[$max_index_to_go]            = -1;
        $matching_token_to_go[$max_index_to_go]        = '';
        $bond_strength_to_go[$max_index_to_go]         = 0;

        # Note: negative levels are currently retained as a diagnostic so that
        # the 'final indentation level' is correctly reported for bad scripts.
        # But this means that every use of $level as an index must be checked.
        # If this becomes too much of a problem, we might give up and just clip
        # them at zero.
        ## $levels_to_go[$max_index_to_go] = ( $level > 0 ) ? $level : 0;
        $levels_to_go[$max_index_to_go] = $level;
        $nesting_depth_to_go[$max_index_to_go] = ( $slevel >= 0 ) ? $slevel : 0;

        # link the non-blank tokens
        my $iprev = $max_index_to_go - 1;
        $iprev-- if ( $iprev >= 0 && $types_to_go[$iprev] eq 'b' );
        $iprev_to_go[$max_index_to_go] = $iprev;
        $inext_to_go[$iprev]           = $max_index_to_go
          if ( $iprev >= 0 && $type ne 'b' );
        $inext_to_go[$max_index_to_go] = $max_index_to_go + 1;

        $token_lengths_to_go[$max_index_to_go] =
          token_length( $token, $type, $max_index_to_go );

        # We keep a running sum of token lengths from the start of this batch:
        #   summed_lengths_to_go[$i]   = total length to just before token $i
        #   summed_lengths_to_go[$i+1] = total length to just after token $i
        $summed_lengths_to_go[ $max_index_to_go + 1 ] =
          $summed_lengths_to_go[$max_index_to_go] +
          $token_lengths_to_go[$max_index_to_go];

        # Define the indentation that this token would have if it started
        # a new line.  We have to do this now because we need to know this
        # when considering one-line blocks.
        set_leading_whitespace( $level, $ci_level, $in_continued_quote );

        # remember previous nonblank tokens seen
        if ( $type ne 'b' ) {
            $last_last_nonblank_index_to_go = $last_nonblank_index_to_go;
            $last_last_nonblank_type_to_go  = $last_nonblank_type_to_go;
            $last_last_nonblank_token_to_go = $last_nonblank_token_to_go;
            $last_nonblank_index_to_go      = $max_index_to_go;
            $last_nonblank_type_to_go       = $type;
            $last_nonblank_token_to_go      = $token;
            if ( $type eq ',' ) {
                $comma_count_in_batch++;
            }
        }

        FORMATTER_DEBUG_FLAG_STORE && do {
            my ( $a, $b, $c ) = caller();
            print STDOUT
"STORE: from $a $c: storing token $token type $type lev=$level slev=$slevel at $max_index_to_go\n";
        };
    }

    sub insert_new_token_to_go {

        # insert a new token into the output stream.  use same level as
        # previous token; assumes a character at max_index_to_go.
        save_current_token();
        ( $token, $type, $slevel, $no_internal_newlines ) = @_;

        if ( $max_index_to_go == UNDEFINED_INDEX ) {
            warning("code bug: bad call to insert_new_token_to_go\n");
        }
        $level = $levels_to_go[$max_index_to_go];

        # FIXME: it seems to be necessary to use the next, rather than
        # previous, value of this variable when creating a new blank (align.t)
        #my $slevel         = $nesting_depth_to_go[$max_index_to_go];
        $nesting_blocks        = $nesting_blocks_to_go[$max_index_to_go];
        $ci_level              = $ci_levels_to_go[$max_index_to_go];
        $container_environment = $container_environment_to_go[$max_index_to_go];
        $in_continued_quote    = 0;
        $block_type            = "";
        $type_sequence         = "";
        store_token_to_go();
        restore_current_token();
        return;
    }

    sub print_line_of_tokens {

        my $line_of_tokens = shift;

        # This routine is called once per input line to process all of
        # the tokens on that line.  This is the first stage of
        # beautification.
        #
        # Full-line comments and blank lines may be processed immediately.
        #
        # For normal lines of code, the tokens are stored one-by-one,
        # via calls to 'sub store_token_to_go', until a known line break
        # point is reached.  Then, the batch of collected tokens is
        # passed along to 'sub output_line_to_go' for further
        # processing.  This routine decides if there should be
        # whitespace between each pair of non-white tokens, so later
        # routines only need to decide on any additional line breaks.
        # Any whitespace is initially a single space character.  Later,
        # the vertical aligner may expand that to be multiple space
        # characters if necessary for alignment.

        # extract input line number for error messages
        $input_line_number = $line_of_tokens->{_line_number};

        $rtoken_type            = $line_of_tokens->{_rtoken_type};
        $rtokens                = $line_of_tokens->{_rtokens};
        $rlevels                = $line_of_tokens->{_rlevels};
        $rslevels               = $line_of_tokens->{_rslevels};
        $rblock_type            = $line_of_tokens->{_rblock_type};
        $rcontainer_type        = $line_of_tokens->{_rcontainer_type};
        $rcontainer_environment = $line_of_tokens->{_rcontainer_environment};
        $rtype_sequence         = $line_of_tokens->{_rtype_sequence};
        $input_line             = $line_of_tokens->{_line_text};
        $rnesting_tokens        = $line_of_tokens->{_rnesting_tokens};
        $rci_levels             = $line_of_tokens->{_rci_levels};
        $rnesting_blocks        = $line_of_tokens->{_rnesting_blocks};

        $in_continued_quote = $starting_in_quote =
          $line_of_tokens->{_starting_in_quote};
        $in_quote        = $line_of_tokens->{_ending_in_quote};
        $ending_in_quote = $in_quote;
        $guessed_indentation_level =
          $line_of_tokens->{_guessed_indentation_level};

        my $j;
        my $j_next;
        my $jmax;
        my $next_nonblank_token;
        my $next_nonblank_token_type;
        my $rwhite_space_flag;

        $jmax                    = @$rtokens - 1;
        $block_type              = "";
        $container_type          = "";
        $container_environment   = "";
        $type_sequence           = "";
        $no_internal_newlines    = 1 - $rOpts_add_newlines;
        $is_static_block_comment = 0;

        # Handle a continued quote..
        if ($in_continued_quote) {

            # A line which is entirely a quote or pattern must go out
            # verbatim.  Note: the \n is contained in $input_line.
            if ( $jmax <= 0 ) {
                if ( ( $input_line =~ "\t" ) ) {
                    note_embedded_tab();
                }
                write_unindented_line("$input_line");
                $last_line_had_side_comment = 0;
                return;
            }
        }

        # Write line verbatim if we are in a formatting skip section
        if ($in_format_skipping_section) {
            write_unindented_line("$input_line");
            $last_line_had_side_comment = 0;

            # Note: extra space appended to comment simplifies pattern matching
            if (   $jmax == 0
                && $$rtoken_type[0] eq '#'
                && ( $$rtokens[0] . " " ) =~ /$format_skipping_pattern_end/o )
            {
                $in_format_skipping_section = 0;
                write_logfile_entry("Exiting formatting skip section\n");
                $file_writer_object->reset_consecutive_blank_lines();
            }
            return;
        }

        # See if we are entering a formatting skip section
        if (   $rOpts_format_skipping
            && $jmax == 0
            && $$rtoken_type[0] eq '#'
            && ( $$rtokens[0] . " " ) =~ /$format_skipping_pattern_begin/o )
        {
            flush();
            $in_format_skipping_section = 1;
            write_logfile_entry("Entering formatting skip section\n");
            write_unindented_line("$input_line");
            $last_line_had_side_comment = 0;
            return;
        }

        # delete trailing blank tokens
        if ( $jmax > 0 && $$rtoken_type[$jmax] eq 'b' ) { $jmax-- }

        # Handle a blank line..
        if ( $jmax < 0 ) {

            # If keep-old-blank-lines is zero, we delete all
            # old blank lines and let the blank line rules generate any
            # needed blanks.
            if ($rOpts_keep_old_blank_lines) {
                flush();
                $file_writer_object->write_blank_code_line(
                    $rOpts_keep_old_blank_lines == 2 );
                $last_line_leading_type = 'b';
            }
            $last_line_had_side_comment = 0;
            return;
        }

        # see if this is a static block comment (starts with ## by default)
        my $is_static_block_comment_without_leading_space = 0;
        if (   $jmax == 0
            && $$rtoken_type[0] eq '#'
            && $rOpts->{'static-block-comments'}
            && $input_line =~ /$static_block_comment_pattern/o )
        {
            $is_static_block_comment = 1;
            $is_static_block_comment_without_leading_space =
              substr( $input_line, 0, 1 ) eq '#';
        }

        # Check for comments which are line directives
        # Treat exactly as static block comments without leading space
        # reference: perlsyn, near end, section Plain Old Comments (Not!)
        # example: '# line 42 "new_filename.plx"'
        if (
               $jmax == 0
            && $$rtoken_type[0] eq '#'
            && $input_line =~ /^\#   \s*
                               line \s+ (\d+)   \s*
                               (?:\s("?)([^"]+)\2)? \s*
                               $/x
          )
        {
            $is_static_block_comment                       = 1;
            $is_static_block_comment_without_leading_space = 1;
        }

        # create a hanging side comment if appropriate
        my $is_hanging_side_comment;
        if (
               $jmax == 0
            && $$rtoken_type[0] eq '#'      # only token is a comment
            && $last_line_had_side_comment  # last line had side comment
            && $input_line =~ /^\s/         # there is some leading space
            && !$is_static_block_comment    # do not make static comment hanging
            && $rOpts->{'hanging-side-comments'}    # user is allowing
                                                    # hanging side comments
                                                    # like this
          )
        {

            # We will insert an empty qw string at the start of the token list
            # to force this comment to be a side comment. The vertical aligner
            # should then line it up with the previous side comment.
            $is_hanging_side_comment = 1;
            unshift @$rtoken_type,            'q';
            unshift @$rtokens,                '';
            unshift @$rlevels,                $$rlevels[0];
            unshift @$rslevels,               $$rslevels[0];
            unshift @$rblock_type,            '';
            unshift @$rcontainer_type,        '';
            unshift @$rcontainer_environment, '';
            unshift @$rtype_sequence,         '';
            unshift @$rnesting_tokens,        $$rnesting_tokens[0];
            unshift @$rci_levels,             $$rci_levels[0];
            unshift @$rnesting_blocks,        $$rnesting_blocks[0];
            $jmax = 1;
        }

        # remember if this line has a side comment
        $last_line_had_side_comment =
          ( $jmax > 0 && $$rtoken_type[$jmax] eq '#' );

        # Handle a block (full-line) comment..
        if ( ( $jmax == 0 ) && ( $$rtoken_type[0] eq '#' ) ) {

            if ( $rOpts->{'delete-block-comments'} ) { return }

            if ( $rOpts->{'tee-block-comments'} ) {
                $file_writer_object->tee_on();
            }

            destroy_one_line_block();
            output_line_to_go();

            # output a blank line before block comments
            if (
                # unless we follow a blank or comment line
                $last_line_leading_type !~ /^[#b]$/

                # only if allowed
                && $rOpts->{'blanks-before-comments'}

                # not if this is an empty comment line
                && $$rtokens[0] ne '#'

                # not after a short line ending in an opening token
                # because we already have space above this comment.
                # Note that the first comment in this if block, after
                # the 'if (', does not get a blank line because of this.
                && !$last_output_short_opening_token

                # never before static block comments
                && !$is_static_block_comment
              )
            {
                flush();    # switching to new output stream
                $file_writer_object->write_blank_code_line();
                $last_line_leading_type = 'b';
            }

            # TRIM COMMENTS -- This could be turned off as a option
            $$rtokens[0] =~ s/\s*$//;    # trim right end

            if (
                $rOpts->{'indent-block-comments'}
                && (  !$rOpts->{'indent-spaced-block-comments'}
                    || $input_line =~ /^\s+/ )
                && !$is_static_block_comment_without_leading_space
              )
            {
                extract_token(0);
                store_token_to_go();
                output_line_to_go();
            }
            else {
                flush();    # switching to new output stream
                $file_writer_object->write_code_line( $$rtokens[0] . "\n" );
                $last_line_leading_type = '#';
            }
            if ( $rOpts->{'tee-block-comments'} ) {
                $file_writer_object->tee_off();
            }
            return;
        }

        # compare input/output indentation except for continuation lines
        # (because they have an unknown amount of initial blank space)
        # and lines which are quotes (because they may have been outdented)
        # Note: this test is placed here because we know the continuation flag
        # at this point, which allows us to avoid non-meaningful checks.
        my $structural_indentation_level = $$rlevels[0];
        compare_indentation_levels( $guessed_indentation_level,
            $structural_indentation_level )
          unless ( $is_hanging_side_comment
            || $$rci_levels[0] > 0
            || $guessed_indentation_level == 0 && $$rtoken_type[0] eq 'Q' );

        #   Patch needed for MakeMaker.  Do not break a statement
        #   in which $VERSION may be calculated.  See MakeMaker.pm;
        #   this is based on the coding in it.
        #   The first line of a file that matches this will be eval'd:
        #       /([\$*])(([\w\:\']*)\bVERSION)\b.*\=/
        #   Examples:
        #     *VERSION = \'1.01';
        #     ( $VERSION ) = '$Revision: 1.74 $ ' =~ /\$Revision:\s+([^\s]+)/;
        #   We will pass such a line straight through without breaking
        #   it unless -npvl is used

        my $is_VERSION_statement = 0;

        if (
              !$saw_VERSION_in_this_file
            && $input_line =~ /VERSION/    # quick check to reject most lines
            && $input_line =~ /([\$*])(([\w\:\']*)\bVERSION)\b.*\=/
          )
        {
            $saw_VERSION_in_this_file = 1;
            $is_VERSION_statement     = 1;
            write_logfile_entry("passing VERSION line; -npvl deactivates\n");
            $no_internal_newlines = 1;
        }

        # take care of indentation-only
        # NOTE: In previous versions we sent all qw lines out immediately here.
        # No longer doing this: also write a line which is entirely a 'qw' list
        # to allow stacking of opening and closing tokens.  Note that interior
        # qw lines will still go out at the end of this routine.
        if ( $rOpts->{'indent-only'} ) {
            flush();
            trim($input_line);

            extract_token(0);
            $token                 = $input_line;
            $type                  = 'q';
            $block_type            = "";
            $container_type        = "";
            $container_environment = "";
            $type_sequence         = "";
            store_token_to_go();
            output_line_to_go();
            return;
        }

        push( @$rtokens,     ' ', ' ' );   # making $j+2 valid simplifies coding
        push( @$rtoken_type, 'b', 'b' );
        ($rwhite_space_flag) =
          set_white_space_flag( $jmax, $rtokens, $rtoken_type, $rblock_type );

        # if the buffer hasn't been flushed, add a leading space if
        # necessary to keep essential whitespace. This is really only
        # necessary if we are squeezing out all ws.
        if ( $max_index_to_go >= 0 ) {

            $old_line_count_in_batch++;

            if (
                is_essential_whitespace(
                    $last_last_nonblank_token,
                    $last_last_nonblank_type,
                    $tokens_to_go[$max_index_to_go],
                    $types_to_go[$max_index_to_go],
                    $$rtokens[0],
                    $$rtoken_type[0]
                )
              )
            {
                my $slevel = $$rslevels[0];
                insert_new_token_to_go( ' ', 'b', $slevel,
                    $no_internal_newlines );
            }
        }

        # If we just saw the end of an elsif block, write nag message
        # if we do not see another elseif or an else.
        if ($looking_for_else) {

            unless ( $$rtokens[0] =~ /^(elsif|else)$/ ) {
                write_logfile_entry("(No else block)\n");
            }
            $looking_for_else = 0;
        }

        # This is a good place to kill incomplete one-line blocks
        if (   ( $semicolons_before_block_self_destruct == 0 )
            && ( $max_index_to_go >= 0 )
            && ( $types_to_go[$max_index_to_go] eq ';' )
            && ( $$rtokens[0] ne '}' ) )
        {
            destroy_one_line_block();
            output_line_to_go();
        }

        # loop to process the tokens one-by-one
        $type  = 'b';
        $token = "";

        foreach $j ( 0 .. $jmax ) {

            # pull out the local values for this token
            extract_token($j);

            if ( $type eq '#' ) {

                # trim trailing whitespace
                # (there is no option at present to prevent this)
                $token =~ s/\s*$//;

                if (
                    $rOpts->{'delete-side-comments'}

                    # delete closing side comments if necessary
                    || (   $rOpts->{'delete-closing-side-comments'}
                        && $token =~ /$closing_side_comment_prefix_pattern/o
                        && $last_nonblank_block_type =~
                        /$closing_side_comment_list_pattern/o )
                  )
                {
                    if ( $types_to_go[$max_index_to_go] eq 'b' ) {
                        unstore_token_to_go();
                    }
                    last;
                }
            }

            # If we are continuing after seeing a right curly brace, flush
            # buffer unless we see what we are looking for, as in
            #   } else ...
            if ( $rbrace_follower && $type ne 'b' ) {

                unless ( $rbrace_follower->{$token} ) {
                    output_line_to_go();
                }
                $rbrace_follower = undef;
            }

            $j_next = ( $$rtoken_type[ $j + 1 ] eq 'b' ) ? $j + 2 : $j + 1;
            $next_nonblank_token      = $$rtokens[$j_next];
            $next_nonblank_token_type = $$rtoken_type[$j_next];

            #--------------------------------------------------------
            # Start of section to patch token text
            #--------------------------------------------------------

            # Modify certain tokens here for whitespace
            # The following is not yet done, but could be:
            #   sub (x x x)
            if ( $type =~ /^[wit]$/ ) {

                # Examples:
                # change '$  var'  to '$var' etc
                #        '-> new'  to '->new'
                if ( $token =~ /^([\$\&\%\*\@]|\-\>)\s/ ) {
                    $token =~ s/\s*//g;
                }

                # Split identifiers with leading arrows, inserting blanks if
                # necessary.  It is easier and safer here than in the
                # tokenizer.  For example '->new' becomes two tokens, '->' and
                # 'new' with a possible blank between.
                #
                # Note: there is a related patch in sub set_white_space_flag
                if ( $token =~ /^\-\>(.*)$/ && $1 ) {
                    my $token_save = $1;
                    my $type_save  = $type;

                    # store a blank to left of arrow if necessary
                    if (   $max_index_to_go >= 0
                        && $types_to_go[$max_index_to_go] ne 'b'
                        && $want_left_space{'->'} == WS_YES )
                    {
                        insert_new_token_to_go( ' ', 'b', $slevel,
                            $no_internal_newlines );
                    }

                    # then store the arrow
                    $token = '->';
                    $type  = $token;
                    store_token_to_go();

                    # then reset the current token to be the remainder,
                    # and reset the whitespace flag according to the arrow
                    $$rwhite_space_flag[$j] = $want_right_space{'->'};
                    $token                  = $token_save;
                    $type                   = $type_save;
                }

                if ( $token =~ /^sub/ ) { $token =~ s/\s+/ /g }

                # trim identifiers of trailing blanks which can occur
                # under some unusual circumstances, such as if the
                # identifier 'witch' has trailing blanks on input here:
                #
                # sub
                # witch
                # ()   # prototype may be on new line ...
                # ...
                if ( $type eq 'i' ) { $token =~ s/\s+$//g }
            }

            # change 'LABEL   :'   to 'LABEL:'
            elsif ( $type eq 'J' ) { $token =~ s/\s+//g }

            # patch to add space to something like "x10"
            # This avoids having to split this token in the pre-tokenizer
            elsif ( $type eq 'n' ) {
                if ( $token =~ /^x\d+/ ) { $token =~ s/x/x / }
            }

            elsif ( $type eq 'Q' ) {
                note_embedded_tab() if ( $token =~ "\t" );

                # make note of something like '$var = s/xxx/yyy/;'
                # in case it should have been '$var =~ s/xxx/yyy/;'
                if (
                       $token =~ /^(s|tr|y|m|\/)/
                    && $last_nonblank_token =~ /^(=|==|!=)$/

                    # precededed by simple scalar
                    && $last_last_nonblank_type eq 'i'
                    && $last_last_nonblank_token =~ /^\$/

                    # followed by some kind of termination
                    # (but give complaint if we can's see far enough ahead)
                    && $next_nonblank_token =~ /^[; \)\}]$/

                    # scalar is not decleared
                    && !(
                           $types_to_go[0] eq 'k'
                        && $tokens_to_go[0] =~ /^(my|our|local)$/
                    )
                  )
                {
                    my $guess = substr( $last_nonblank_token, 0, 1 ) . '~';
                    complain(
"Note: be sure you want '$last_nonblank_token' instead of '$guess' here\n"
                    );
                }
            }

           # trim blanks from right of qw quotes
           # (To avoid trimming qw quotes use -ntqw; the tokenizer handles this)
            elsif ( $type eq 'q' ) {
                $token =~ s/\s*$//;
                note_embedded_tab() if ( $token =~ "\t" );
            }

            #--------------------------------------------------------
            # End of section to patch token text
            #--------------------------------------------------------

            # insert any needed whitespace
            if (   ( $type ne 'b' )
                && ( $max_index_to_go >= 0 )
                && ( $types_to_go[$max_index_to_go] ne 'b' )
                && $rOpts_add_whitespace )
            {
                my $ws = $$rwhite_space_flag[$j];

                if ( $ws == 1 ) {
                    insert_new_token_to_go( ' ', 'b', $slevel,
                        $no_internal_newlines );
                }
            }

            # Do not allow breaks which would promote a side comment to a
            # block comment.  In order to allow a break before an opening
            # or closing BLOCK, followed by a side comment, those sections
            # of code will handle this flag separately.
            my $side_comment_follows = ( $next_nonblank_token_type eq '#' );
            my $is_opening_BLOCK =
              (      $type eq '{'
                  && $token eq '{'
                  && $block_type
                  && $block_type ne 't' );
            my $is_closing_BLOCK =
              (      $type eq '}'
                  && $token eq '}'
                  && $block_type
                  && $block_type ne 't' );

            if (   $side_comment_follows
                && !$is_opening_BLOCK
                && !$is_closing_BLOCK )
            {
                $no_internal_newlines = 1;
            }

            # We're only going to handle breaking for code BLOCKS at this
            # (top) level.  Other indentation breaks will be handled by
            # sub scan_list, which is better suited to dealing with them.
            if ($is_opening_BLOCK) {

                # Tentatively output this token.  This is required before
                # calling starting_one_line_block.  We may have to unstore
                # it, though, if we have to break before it.
                store_token_to_go($side_comment_follows);

                # Look ahead to see if we might form a one-line block
                my $too_long =
                  starting_one_line_block( $j, $jmax, $level, $slevel,
                    $ci_level, $rtokens, $rtoken_type, $rblock_type );
                clear_breakpoint_undo_stack();

                # to simplify the logic below, set a flag to indicate if
                # this opening brace is far from the keyword which introduces it
                my $keyword_on_same_line = 1;
                if (   ( $max_index_to_go >= 0 )
                    && ( $last_nonblank_type eq ')' ) )
                {
                    if (   $block_type =~ /^(if|else|elsif)$/
                        && ( $tokens_to_go[0] eq '}' )
                        && $rOpts_cuddled_else )
                    {
                        $keyword_on_same_line = 1;
                    }
                    elsif ( ( $slevel < $nesting_depth_to_go[0] ) || $too_long )
                    {
                        $keyword_on_same_line = 0;
                    }
                }

                # decide if user requested break before '{'
                my $want_break =

                  # use -bl flag if not a sub block of any type
                  $block_type !~ /^sub/
                  ? $rOpts->{'opening-brace-on-new-line'}

                  # use -sbl flag for a named sub block
                  : $block_type !~ /^sub\W*$/
                  ? $rOpts->{'opening-sub-brace-on-new-line'}

                  # use -asbl flag for an anonymous sub block
                  : $rOpts->{'opening-anonymous-sub-brace-on-new-line'};

                # Break before an opening '{' ...
                if (

                    # if requested
                    $want_break

                    # and we were unable to start looking for a block,
                    && $index_start_one_line_block == UNDEFINED_INDEX

                    # or if it will not be on same line as its keyword, so that
                    # it will be outdented (eval.t, overload.t), and the user
                    # has not insisted on keeping it on the right
                    || (   !$keyword_on_same_line
                        && !$rOpts->{'opening-brace-always-on-right'} )

                  )
                {

                    # but only if allowed
                    unless ($no_internal_newlines) {

                        # since we already stored this token, we must unstore it
                        unstore_token_to_go();

                        # then output the line
                        output_line_to_go();

                        # and now store this token at the start of a new line
                        store_token_to_go($side_comment_follows);
                    }
                }

                # Now update for side comment
                if ($side_comment_follows) { $no_internal_newlines = 1 }

                # now output this line
                unless ($no_internal_newlines) {
                    output_line_to_go();
                }
            }

            elsif ($is_closing_BLOCK) {

                # If there is a pending one-line block ..
                if ( $index_start_one_line_block != UNDEFINED_INDEX ) {

                    # we have to terminate it if..
                    if (

                    # it is too long (final length may be different from
                    # initial estimate). note: must allow 1 space for this token
                        excess_line_length( $index_start_one_line_block,
                            $max_index_to_go ) >= 0

                        # or if it has too many semicolons
                        || (   $semicolons_before_block_self_destruct == 0
                            && $last_nonblank_type ne ';' )
                      )
                    {
                        destroy_one_line_block();
                    }
                }

                # put a break before this closing curly brace if appropriate
                unless ( $no_internal_newlines
                    || $index_start_one_line_block != UNDEFINED_INDEX )
                {

                    # add missing semicolon if ...
                    # there are some tokens
                    if (
                        ( $max_index_to_go > 0 )

                        # and we don't have one
                        && ( $last_nonblank_type ne ';' )

                        # patch until some block type issues are fixed:
                        # Do not add semi-colon for block types '{',
                        # '}', and ';' because we cannot be sure yet
                        # that this is a block and not an anonomyous
                        # hash (blktype.t, blktype1.t)
                        && ( $block_type !~ /^[\{\};]$/ )

                        # patch: and do not add semi-colons for recently
                        # added block types (see tmp/semicolon.t)
                        && ( $block_type !~
                            /^(switch|case|given|when|default)$/ )

                        # it seems best not to add semicolons in these
                        # special block types: sort|map|grep
                        && ( !$is_sort_map_grep{$block_type} )

                        # and we are allowed to do so.
                        && $rOpts->{'add-semicolons'}
                      )
                    {

                        save_current_token();
                        $token  = ';';
                        $type   = ';';
                        $level  = $levels_to_go[$max_index_to_go];
                        $slevel = $nesting_depth_to_go[$max_index_to_go];
                        $nesting_blocks =
                          $nesting_blocks_to_go[$max_index_to_go];
                        $ci_level       = $ci_levels_to_go[$max_index_to_go];
                        $block_type     = "";
                        $container_type = "";
                        $container_environment = "";
                        $type_sequence         = "";

                        # Note - we remove any blank AFTER extracting its
                        # parameters such as level, etc, above
                        if ( $types_to_go[$max_index_to_go] eq 'b' ) {
                            unstore_token_to_go();
                        }
                        store_token_to_go();

                        note_added_semicolon();
                        restore_current_token();
                    }

                    # then write out everything before this closing curly brace
                    output_line_to_go();

                }

                # Now update for side comment
                if ($side_comment_follows) { $no_internal_newlines = 1 }

                # store the closing curly brace
                store_token_to_go();

                # ok, we just stored a closing curly brace.  Often, but
                # not always, we want to end the line immediately.
                # So now we have to check for special cases.

                # if this '}' successfully ends a one-line block..
                my $is_one_line_block = 0;
                my $keep_going        = 0;
                if ( $index_start_one_line_block != UNDEFINED_INDEX ) {

                    # Remember the type of token just before the
                    # opening brace.  It would be more general to use
                    # a stack, but this will work for one-line blocks.
                    $is_one_line_block =
                      $types_to_go[$index_start_one_line_block];

                    # we have to actually make it by removing tentative
                    # breaks that were set within it
                    undo_forced_breakpoint_stack(0);
                    set_nobreaks( $index_start_one_line_block,
                        $max_index_to_go - 1 );

                    # then re-initialize for the next one-line block
                    destroy_one_line_block();

                    # then decide if we want to break after the '}' ..
                    # We will keep going to allow certain brace followers as in:
                    #   do { $ifclosed = 1; last } unless $losing;
                    #
                    # But make a line break if the curly ends a
                    # significant block:
                    if (
                        $is_block_without_semicolon{$block_type}

                        # if needless semicolon follows we handle it later
                        && $next_nonblank_token ne ';'
                      )
                    {
                        output_line_to_go() unless ($no_internal_newlines);
                    }
                }

                # set string indicating what we need to look for brace follower
                # tokens
                if ( $block_type eq 'do' ) {
                    $rbrace_follower = \%is_do_follower;
                }
                elsif ( $block_type =~ /^(if|elsif|unless)$/ ) {
                    $rbrace_follower = \%is_if_brace_follower;
                }
                elsif ( $block_type eq 'else' ) {
                    $rbrace_follower = \%is_else_brace_follower;
                }

                # added eval for borris.t
                elsif ($is_sort_map_grep_eval{$block_type}
                    || $is_one_line_block eq 'G' )
                {
                    $rbrace_follower = undef;
                    $keep_going      = 1;
                }

                # anonymous sub
                elsif ( $block_type =~ /^sub\W*$/ ) {

                    if ($is_one_line_block) {
                        $rbrace_follower = \%is_anon_sub_1_brace_follower;
                    }
                    else {
                        $rbrace_follower = \%is_anon_sub_brace_follower;
                    }
                }

                # None of the above: specify what can follow a closing
                # brace of a block which is not an
                # if/elsif/else/do/sort/map/grep/eval
                # Testfiles:
                # 'Toolbar.pm', 'Menubar.pm', bless.t, '3rules.pl', 'break1.t
                else {
                    $rbrace_follower = \%is_other_brace_follower;
                }

                # See if an elsif block is followed by another elsif or else;
                # complain if not.
                if ( $block_type eq 'elsif' ) {

                    if ( $next_nonblank_token_type eq 'b' ) {    # end of line?
                        $looking_for_else = 1;    # ok, check on next line
                    }
                    else {

                        unless ( $next_nonblank_token =~ /^(elsif|else)$/ ) {
                            write_logfile_entry("No else block :(\n");
                        }
                    }
                }

                # keep going after certain block types (map,sort,grep,eval)
                # added eval for borris.t
                if ($keep_going) {

                    # keep going
                }

                # if no more tokens, postpone decision until re-entring
                elsif ( ( $next_nonblank_token_type eq 'b' )
                    && $rOpts_add_newlines )
                {
                    unless ($rbrace_follower) {
                        output_line_to_go() unless ($no_internal_newlines);
                    }
                }

                elsif ($rbrace_follower) {

                    unless ( $rbrace_follower->{$next_nonblank_token} ) {
                        output_line_to_go() unless ($no_internal_newlines);
                    }
                    $rbrace_follower = undef;
                }

                else {
                    output_line_to_go() unless ($no_internal_newlines);
                }

            }    # end treatment of closing block token

            # handle semicolon
            elsif ( $type eq ';' ) {

                # kill one-line blocks with too many semicolons
                $semicolons_before_block_self_destruct--;
                if (
                    ( $semicolons_before_block_self_destruct < 0 )
                    || (   $semicolons_before_block_self_destruct == 0
                        && $next_nonblank_token_type !~ /^[b\}]$/ )
                  )
                {
                    destroy_one_line_block();
                }

                # Remove unnecessary semicolons, but not after bare
                # blocks, where it could be unsafe if the brace is
                # mistokenized.
                if (
                    (
                        $last_nonblank_token eq '}'
                        && (
                            $is_block_without_semicolon{
                                $last_nonblank_block_type}
                            || $last_nonblank_block_type =~ /^sub\s+\w/
                            || $last_nonblank_block_type =~ /^\w+:$/ )
                    )
                    || $last_nonblank_type eq ';'
                  )
                {

                    if (
                        $rOpts->{'delete-semicolons'}

                        # don't delete ; before a # because it would promote it
                        # to a block comment
                        && ( $next_nonblank_token_type ne '#' )
                      )
                    {
                        note_deleted_semicolon();
                        output_line_to_go()
                          unless ( $no_internal_newlines
                            || $index_start_one_line_block != UNDEFINED_INDEX );
                        next;
                    }
                    else {
                        write_logfile_entry("Extra ';'\n");
                    }
                }
                store_token_to_go();

                output_line_to_go()
                  unless ( $no_internal_newlines
                    || ( $rOpts_keep_interior_semicolons && $j < $jmax )
                    || ( $next_nonblank_token eq '}' ) );

            }

            # handle here_doc target string
            elsif ( $type eq 'h' ) {
                $no_internal_newlines =
                  1;    # no newlines after seeing here-target
                destroy_one_line_block();
                store_token_to_go();
            }

            # handle all other token types
            else {

                # if this is a blank...
                if ( $type eq 'b' ) {

                    # make it just one character
                    $token = ' ' if $rOpts_add_whitespace;

                    # delete it if unwanted by whitespace rules
                    # or we are deleting all whitespace
                    my $ws = $$rwhite_space_flag[ $j + 1 ];
                    if ( ( defined($ws) && $ws == -1 )
                        || $rOpts_delete_old_whitespace )
                    {

                        # unless it might make a syntax error
                        next
                          unless is_essential_whitespace(
                            $last_last_nonblank_token,
                            $last_last_nonblank_type,
                            $tokens_to_go[$max_index_to_go],
                            $types_to_go[$max_index_to_go],
                            $$rtokens[ $j + 1 ],
                            $$rtoken_type[ $j + 1 ]
                          );
                    }
                }
                store_token_to_go();
            }

            # remember two previous nonblank OUTPUT tokens
            if ( $type ne '#' && $type ne 'b' ) {
                $last_last_nonblank_token = $last_nonblank_token;
                $last_last_nonblank_type  = $last_nonblank_type;
                $last_nonblank_token      = $token;
                $last_nonblank_type       = $type;
                $last_nonblank_block_type = $block_type;
            }

            # unset the continued-quote flag since it only applies to the
            # first token, and we want to resume normal formatting if
            # there are additional tokens on the line
            $in_continued_quote = 0;

        }    # end of loop over all tokens in this 'line_of_tokens'

        # we have to flush ..
        if (

            # if there is a side comment
            ( ( $type eq '#' ) && !$rOpts->{'delete-side-comments'} )

            # if this line ends in a quote
            # NOTE: This is critically important for insuring that quoted lines
            # do not get processed by things like -sot and -sct
            || $in_quote

            # if this is a VERSION statement
            || $is_VERSION_statement

            # to keep a label at the end of a line
            || $type eq 'J'

            # if we are instructed to keep all old line breaks
            || !$rOpts->{'delete-old-newlines'}
          )
        {
            destroy_one_line_block();
            output_line_to_go();
        }

        # mark old line breakpoints in current output stream
        if ( $max_index_to_go >= 0 && !$rOpts_ignore_old_breakpoints ) {
            $old_breakpoint_to_go[$max_index_to_go] = 1;
        }
    } ## end sub print_line_of_tokens
} ## end block print_line_of_tokens

# sub output_line_to_go sends one logical line of tokens on down the
# pipeline to the VerticalAligner package, breaking the line into continuation
# lines as necessary.  The line of tokens is ready to go in the "to_go"
# arrays.
sub output_line_to_go {

    # debug stuff; this routine can be called from many points
    FORMATTER_DEBUG_FLAG_OUTPUT && do {
        my ( $a, $b, $c ) = caller;
        write_diagnostics(
"OUTPUT: output_line_to_go called: $a $c $last_nonblank_type $last_nonblank_token, one_line=$index_start_one_line_block, tokens to write=$max_index_to_go\n"
        );
        my $output_str = join "", @tokens_to_go[ 0 .. $max_index_to_go ];
        write_diagnostics("$output_str\n");
    };

    # just set a tentative breakpoint if we might be in a one-line block
    if ( $index_start_one_line_block != UNDEFINED_INDEX ) {
        set_forced_breakpoint($max_index_to_go);
        return;
    }

    my $cscw_block_comment;
    $cscw_block_comment = add_closing_side_comment()
      if ( $rOpts->{'closing-side-comments'} && $max_index_to_go >= 0 );

    my $comma_arrow_count_contained = match_opening_and_closing_tokens();

    # tell the -lp option we are outputting a batch so it can close
    # any unfinished items in its stack
    finish_lp_batch();

    # If this line ends in a code block brace, set breaks at any
    # previous closing code block braces to breakup a chain of code
    # blocks on one line.  This is very rare but can happen for
    # user-defined subs.  For example we might be looking at this:
    #  BOOL { $server_data{uptime} > 0; } NUM { $server_data{load}; } STR {
    my $saw_good_break = 0;    # flag to force breaks even if short line
    if (

        # looking for opening or closing block brace
        $block_type_to_go[$max_index_to_go]

        # but not one of these which are never duplicated on a line:
        # until|while|for|if|elsif|else
        && !$is_block_without_semicolon{ $block_type_to_go[$max_index_to_go] }
      )
    {
        my $lev = $nesting_depth_to_go[$max_index_to_go];

        # Walk backwards from the end and
        # set break at any closing block braces at the same level.
        # But quit if we are not in a chain of blocks.
        for ( my $i = $max_index_to_go - 1 ; $i >= 0 ; $i-- ) {
            last if ( $levels_to_go[$i] < $lev );    # stop at a lower level
            next if ( $levels_to_go[$i] > $lev );    # skip past higher level

            if ( $block_type_to_go[$i] ) {
                if ( $tokens_to_go[$i] eq '}' ) {
                    set_forced_breakpoint($i);
                    $saw_good_break = 1;
                }
            }

            # quit if we see anything besides words, function, blanks
            # at this level
            elsif ( $types_to_go[$i] !~ /^[\(\)Gwib]$/ ) { last }
        }
    }

    my $imin = 0;
    my $imax = $max_index_to_go;

    # trim any blank tokens
    if ( $max_index_to_go >= 0 ) {
        if ( $types_to_go[$imin] eq 'b' ) { $imin++ }
        if ( $types_to_go[$imax] eq 'b' ) { $imax-- }
    }

    # anything left to write?
    if ( $imin <= $imax ) {

        # add a blank line before certain key types but not after a comment
        if ( $last_line_leading_type !~ /^[#]/ ) {
            my $want_blank    = 0;
            my $leading_token = $tokens_to_go[$imin];
            my $leading_type  = $types_to_go[$imin];

            # blank lines before subs except declarations and one-liners
            # MCONVERSION LOCATION - for sub tokenization change
            if ( $leading_token =~ /^(sub\s)/ && $leading_type eq 'i' ) {
                $want_blank = $rOpts->{'blank-lines-before-subs'}
                  if (
                    terminal_type( \@types_to_go, \@block_type_to_go, $imin,
                        $imax ) !~ /^[\;\}]$/
                  );
            }

            # break before all package declarations
            # MCONVERSION LOCATION - for tokenizaton change
            elsif ($leading_token =~ /^(package\s)/
                && $leading_type eq 'i' )
            {
                $want_blank = $rOpts->{'blank-lines-before-packages'};
            }

            # break before certain key blocks except one-liners
            if ( $leading_token =~ /^(BEGIN|END)$/ && $leading_type eq 'k' ) {
                $want_blank = $rOpts->{'blank-lines-before-subs'}
                  if (
                    terminal_type( \@types_to_go, \@block_type_to_go, $imin,
                        $imax ) ne '}'
                  );
            }

            # Break before certain block types if we haven't had a
            # break at this level for a while.  This is the
            # difficult decision..
            elsif ($leading_type eq 'k'
                && $last_line_leading_type ne 'b'
                && $leading_token =~ /^(unless|if|while|until|for|foreach)$/ )
            {
                my $lc = $nonblank_lines_at_depth[$last_line_leading_level];
                if ( !defined($lc) ) { $lc = 0 }

                $want_blank =
                     $rOpts->{'blanks-before-blocks'}
                  && $lc >= $rOpts->{'long-block-line-count'}
                  && $file_writer_object->get_consecutive_nonblank_lines() >=
                  $rOpts->{'long-block-line-count'}
                  && (
                    terminal_type( \@types_to_go, \@block_type_to_go, $imin,
                        $imax ) ne '}'
                  );
            }

            if ($want_blank) {

                # future: send blank line down normal path to VerticalAligner
                Perl::Tidy::VerticalAligner::flush();
                $file_writer_object->require_blank_code_lines($want_blank);
            }
        }

        # update blank line variables and count number of consecutive
        # non-blank, non-comment lines at this level
        $last_last_line_leading_level = $last_line_leading_level;
        $last_line_leading_level      = $levels_to_go[$imin];
        if ( $last_line_leading_level < 0 ) { $last_line_leading_level = 0 }
        $last_line_leading_type = $types_to_go[$imin];
        if (   $last_line_leading_level == $last_last_line_leading_level
            && $last_line_leading_type ne 'b'
            && $last_line_leading_type ne '#'
            && defined( $nonblank_lines_at_depth[$last_line_leading_level] ) )
        {
            $nonblank_lines_at_depth[$last_line_leading_level]++;
        }
        else {
            $nonblank_lines_at_depth[$last_line_leading_level] = 1;
        }

        FORMATTER_DEBUG_FLAG_FLUSH && do {
            my ( $package, $file, $line ) = caller;
            print STDOUT
"FLUSH: flushing from $package $file $line, types= $types_to_go[$imin] to $types_to_go[$imax]\n";
        };

        # add a couple of extra terminal blank tokens
        pad_array_to_go();

        # set all forced breakpoints for good list formatting
        my $is_long_line = excess_line_length( $imin, $max_index_to_go ) > 0;

        if (
               $is_long_line
            || $old_line_count_in_batch > 1

            # must always call scan_list() with unbalanced batches because it
            # is maintaining some stacks
            || is_unbalanced_batch()

            # call scan_list if we might want to break at commas
            || (
                $comma_count_in_batch
                && (   $rOpts_maximum_fields_per_table > 0
                    || $rOpts_comma_arrow_breakpoints == 0 )
            )

            # call scan_list if user may want to break open some one-line
            # hash references
            || (   $comma_arrow_count_contained
                && $rOpts_comma_arrow_breakpoints != 3 )
          )
        {
            ## This caused problems in one version of perl for unknown reasons:
            ## $saw_good_break ||= scan_list();
            my $sgb = scan_list();
            $saw_good_break ||= $sgb;
        }

        # let $ri_first and $ri_last be references to lists of
        # first and last tokens of line fragments to output..
        my ( $ri_first, $ri_last );

        # write a single line if..
        if (

            # we aren't allowed to add any newlines
            !$rOpts_add_newlines

            # or, we don't already have an interior breakpoint
            # and we didn't see a good breakpoint
            || (
                   !$forced_breakpoint_count
                && !$saw_good_break

                # and this line is 'short'
                && !$is_long_line
            )
          )
        {
            @$ri_first = ($imin);
            @$ri_last  = ($imax);
        }

        # otherwise use multiple lines
        else {

            ( $ri_first, $ri_last, my $colon_count ) =
              set_continuation_breaks($saw_good_break);

            break_all_chain_tokens( $ri_first, $ri_last );

            break_equals( $ri_first, $ri_last );

            # now we do a correction step to clean this up a bit
            # (The only time we would not do this is for debugging)
            if ( $rOpts->{'recombine'} ) {
                ( $ri_first, $ri_last ) =
                  recombine_breakpoints( $ri_first, $ri_last );
            }

            insert_final_breaks( $ri_first, $ri_last ) if $colon_count;
        }

        # do corrector step if -lp option is used
        my $do_not_pad = 0;
        if ($rOpts_line_up_parentheses) {
            $do_not_pad = correct_lp_indentation( $ri_first, $ri_last );
        }
        send_lines_to_vertical_aligner( $ri_first, $ri_last, $do_not_pad );
    }
    prepare_for_new_input_lines();

    # output any new -cscw block comment
    if ($cscw_block_comment) {
        flush();
        $file_writer_object->write_code_line( $cscw_block_comment . "\n" );
    }
}

sub note_added_semicolon {
    $last_added_semicolon_at = $input_line_number;
    if ( $added_semicolon_count == 0 ) {
        $first_added_semicolon_at = $last_added_semicolon_at;
    }
    $added_semicolon_count++;
    write_logfile_entry("Added ';' here\n");
}

sub note_deleted_semicolon {
    $last_deleted_semicolon_at = $input_line_number;
    if ( $deleted_semicolon_count == 0 ) {
        $first_deleted_semicolon_at = $last_deleted_semicolon_at;
    }
    $deleted_semicolon_count++;
    write_logfile_entry("Deleted unnecessary ';'\n");    # i hope ;)
}

sub note_embedded_tab {
    $embedded_tab_count++;
    $last_embedded_tab_at = $input_line_number;
    if ( !$first_embedded_tab_at ) {
        $first_embedded_tab_at = $last_embedded_tab_at;
    }

    if ( $embedded_tab_count <= MAX_NAG_MESSAGES ) {
        write_logfile_entry("Embedded tabs in quote or pattern\n");
    }
}

sub starting_one_line_block {

    # after seeing an opening curly brace, look for the closing brace
    # and see if the entire block will fit on a line.  This routine is
    # not always right because it uses the old whitespace, so a check
    # is made later (at the closing brace) to make sure we really
    # have a one-line block.  We have to do this preliminary check,
    # though, because otherwise we would always break at a semicolon
    # within a one-line block if the block contains multiple statements.

    my ( $j, $jmax, $level, $slevel, $ci_level, $rtokens, $rtoken_type,
        $rblock_type )
      = @_;

    # kill any current block - we can only go 1 deep
    destroy_one_line_block();

    # return value:
    #  1=distance from start of block to opening brace exceeds line length
    #  0=otherwise

    my $i_start = 0;

    # shouldn't happen: there must have been a prior call to
    # store_token_to_go to put the opening brace in the output stream
    if ( $max_index_to_go < 0 ) {
        warning("program bug: store_token_to_go called incorrectly\n");
        report_definite_bug();
    }
    else {

        # cannot use one-line blocks with cuddled else else/elsif lines
        if ( ( $tokens_to_go[0] eq '}' ) && $rOpts_cuddled_else ) {
            return 0;
        }
    }

    my $block_type = $$rblock_type[$j];

    # find the starting keyword for this block (such as 'if', 'else', ...)

    if ( $block_type =~ /^[\{\}\;\:]$/ || $block_type =~ /^package/ ) {
        $i_start = $max_index_to_go;
    }

    elsif ( $last_last_nonblank_token_to_go eq ')' ) {

        # For something like "if (xxx) {", the keyword "if" will be
        # just after the most recent break. This will be 0 unless
        # we have just killed a one-line block and are starting another.
        # (doif.t)
        # Note: cannot use inext_index_to_go[] here because that array
        # is still being constructed.
        $i_start = $index_max_forced_break + 1;
        if ( $types_to_go[$i_start] eq 'b' ) {
            $i_start++;
        }

        unless ( $tokens_to_go[$i_start] eq $block_type ) {
            return 0;
        }
    }

    # the previous nonblank token should start these block types
    elsif (( $last_last_nonblank_token_to_go eq $block_type )
        || ( $block_type =~ /^sub/ ) )
    {
        $i_start = $last_last_nonblank_index_to_go;
    }

    # patch for SWITCH/CASE to retain one-line case/when blocks
    elsif ( $block_type eq 'case' || $block_type eq 'when' ) {

        # Note: cannot use inext_index_to_go[] here because that array
        # is still being constructed.
        $i_start = $index_max_forced_break + 1;
        if ( $types_to_go[$i_start] eq 'b' ) {
            $i_start++;
        }
        unless ( $tokens_to_go[$i_start] eq $block_type ) {
            return 0;
        }
    }

    else {
        return 1;
    }

    my $pos = total_line_length( $i_start, $max_index_to_go ) - 1;

    my $i;

    # see if length is too long to even start
    if ( $pos > maximum_line_length($i_start) ) {
        return 1;
    }

    for ( $i = $j + 1 ; $i <= $jmax ; $i++ ) {

        # old whitespace could be arbitrarily large, so don't use it
        if   ( $$rtoken_type[$i] eq 'b' ) { $pos += 1 }
        else                              { $pos += rtoken_length($i) }

        # Return false result if we exceed the maximum line length,
        if ( $pos > maximum_line_length($i_start) ) {
            return 0;
        }

        # or encounter another opening brace before finding the closing brace.
        elsif ($$rtokens[$i] eq '{'
            && $$rtoken_type[$i] eq '{'
            && $$rblock_type[$i] )
        {
            return 0;
        }

        # if we find our closing brace..
        elsif ($$rtokens[$i] eq '}'
            && $$rtoken_type[$i] eq '}'
            && $$rblock_type[$i] )
        {

            # be sure any trailing comment also fits on the line
            my $i_nonblank =
              ( $$rtoken_type[ $i + 1 ] eq 'b' ) ? $i + 2 : $i + 1;

            # Patch for one-line sort/map/grep/eval blocks with side comments:
            # We will ignore the side comment length for sort/map/grep/eval
            # because this can lead to statements which change every time
            # perltidy is run.  Here is an example from Denis Moskowitz which
            # oscillates between these two states without this patch:

## --------
## grep { $_->foo ne 'bar' } # asdfa asdf asdf asdf asdf asdf asdf asdf asdf asdf asdf
##  @baz;
##
## grep {
##     $_->foo ne 'bar'
##   }    # asdfa asdf asdf asdf asdf asdf asdf asdf asdf asdf asdf
##   @baz;
## --------

            # When the first line is input it gets broken apart by the main
            # line break logic in sub print_line_of_tokens.
            # When the second line is input it gets recombined by
            # print_line_of_tokens and passed to the output routines.  The
            # output routines (set_continuation_breaks) do not break it apart
            # because the bond strengths are set to the highest possible value
            # for grep/map/eval/sort blocks, so the first version gets output.
            # It would be possible to fix this by changing bond strengths,
            # but they are high to prevent errors in older versions of perl.

            if ( $$rtoken_type[$i_nonblank] eq '#'
                && !$is_sort_map_grep{$block_type} )
            {

                $pos += rtoken_length($i_nonblank);

                if ( $i_nonblank > $i + 1 ) {

                    # source whitespace could be anything, assume
                    # at least one space before the hash on output
                    if ( $$rtoken_type[ $i + 1 ] eq 'b' ) { $pos += 1 }
                    else { $pos += rtoken_length( $i + 1 ) }
                }

                if ( $pos >= maximum_line_length($i_start) ) {
                    return 0;
                }
            }

            # ok, it's a one-line block
            create_one_line_block( $i_start, 20 );
            return 0;
        }

        # just keep going for other characters
        else {
        }
    }

    # Allow certain types of new one-line blocks to form by joining
    # input lines.  These can be safely done, but for other block types,
    # we keep old one-line blocks but do not form new ones. It is not
    # always a good idea to make as many one-line blocks as possible,
    # so other types are not done.  The user can always use -mangle.
    if ( $is_sort_map_grep_eval{$block_type} ) {
        create_one_line_block( $i_start, 1 );
    }

    return 0;
}

sub unstore_token_to_go {

    # remove most recent token from output stream
    if ( $max_index_to_go > 0 ) {
        $max_index_to_go--;
    }
    else {
        $max_index_to_go = UNDEFINED_INDEX;
    }

}

sub want_blank_line {
    flush();
    $file_writer_object->want_blank_line();
}

sub write_unindented_line {
    flush();
    $file_writer_object->write_line( $_[0] );
}

sub undo_ci {

    # Undo continuation indentation in certain sequences
    # For example, we can undo continuation indation in sort/map/grep chains
    #    my $dat1 = pack( "n*",
    #        map { $_, $lookup->{$_} }
    #          sort { $a <=> $b }
    #          grep { $lookup->{$_} ne $default } keys %$lookup );
    # To align the map/sort/grep keywords like this:
    #    my $dat1 = pack( "n*",
    #        map { $_, $lookup->{$_} }
    #        sort { $a <=> $b }
    #        grep { $lookup->{$_} ne $default } keys %$lookup );
    my ( $ri_first, $ri_last ) = @_;
    my ( $line_1, $line_2, $lev_last );
    my $this_line_is_semicolon_terminated;
    my $max_line = @$ri_first - 1;

    # looking at each line of this batch..
    # We are looking at leading tokens and looking for a sequence
    # all at the same level and higher level than enclosing lines.
    foreach my $line ( 0 .. $max_line ) {

        my $ibeg = $$ri_first[$line];
        my $lev  = $levels_to_go[$ibeg];
        if ( $line > 0 ) {

            # if we have started a chain..
            if ($line_1) {

                # see if it continues..
                if ( $lev == $lev_last ) {
                    if (   $types_to_go[$ibeg] eq 'k'
                        && $is_sort_map_grep{ $tokens_to_go[$ibeg] } )
                    {

                        # chain continues...
                        # check for chain ending at end of a a statement
                        if ( $line == $max_line ) {

                            # see of this line ends a statement
                            my $iend = $$ri_last[$line];
                            $this_line_is_semicolon_terminated =
                              $types_to_go[$iend] eq ';'

                              # with possible side comment
                              || ( $types_to_go[$iend] eq '#'
                                && $iend - $ibeg >= 2
                                && $types_to_go[ $iend - 2 ] eq ';'
                                && $types_to_go[ $iend - 1 ] eq 'b' );
                        }
                        $line_2 = $line if ($this_line_is_semicolon_terminated);
                    }
                    else {

                        # kill chain
                        $line_1 = undef;
                    }
                }
                elsif ( $lev < $lev_last ) {

                    # chain ends with previous line
                    $line_2 = $line - 1;
                }
                elsif ( $lev > $lev_last ) {

                    # kill chain
                    $line_1 = undef;
                }

                # undo the continuation indentation if a chain ends
                if ( defined($line_2) && defined($line_1) ) {
                    my $continuation_line_count = $line_2 - $line_1 + 1;
                    @ci_levels_to_go[ @$ri_first[ $line_1 .. $line_2 ] ] =
                      (0) x ($continuation_line_count);
                    @leading_spaces_to_go[ @$ri_first[ $line_1 .. $line_2 ] ] =
                      @reduced_spaces_to_go[ @$ri_first[ $line_1 .. $line_2 ] ];
                    $line_1 = undef;
                }
            }

            # not in a chain yet..
            else {

                # look for start of a new sort/map/grep chain
                if ( $lev > $lev_last ) {
                    if (   $types_to_go[$ibeg] eq 'k'
                        && $is_sort_map_grep{ $tokens_to_go[$ibeg] } )
                    {
                        $line_1 = $line;
                    }
                }
            }
        }
        $lev_last = $lev;
    }
}

sub undo_lp_ci {

    # If there is a single, long parameter within parens, like this:
    #
    #  $self->command( "/msg "
    #        . $infoline->chan
    #        . " You said $1, but did you know that it's square was "
    #        . $1 * $1 . " ?" );
    #
    # we can remove the continuation indentation of the 2nd and higher lines
    # to achieve this effect, which is more pleasing:
    #
    #  $self->command("/msg "
    #                 . $infoline->chan
    #                 . " You said $1, but did you know that it's square was "
    #                 . $1 * $1 . " ?");

    my ( $line_open, $i_start, $closing_index, $ri_first, $ri_last ) = @_;
    my $max_line = @$ri_first - 1;

    # must be multiple lines
    return unless $max_line > $line_open;

    my $lev_start     = $levels_to_go[$i_start];
    my $ci_start_plus = 1 + $ci_levels_to_go[$i_start];

    # see if all additional lines in this container have continuation
    # indentation
    my $n;
    my $line_1 = 1 + $line_open;
    for ( $n = $line_1 ; $n <= $max_line ; ++$n ) {
        my $ibeg = $$ri_first[$n];
        my $iend = $$ri_last[$n];
        if ( $ibeg eq $closing_index ) { $n--; last }
        return if ( $lev_start != $levels_to_go[$ibeg] );
        return if ( $ci_start_plus != $ci_levels_to_go[$ibeg] );
        last   if ( $closing_index <= $iend );
    }

    # we can reduce the indentation of all continuation lines
    my $continuation_line_count = $n - $line_open;
    @ci_levels_to_go[ @$ri_first[ $line_1 .. $n ] ] =
      (0) x ($continuation_line_count);
    @leading_spaces_to_go[ @$ri_first[ $line_1 .. $n ] ] =
      @reduced_spaces_to_go[ @$ri_first[ $line_1 .. $n ] ];
}

sub pad_token {

    # insert $pad_spaces before token number $ipad
    my ( $ipad, $pad_spaces ) = @_;
    if ( $pad_spaces > 0 ) {
        $tokens_to_go[$ipad] = ' ' x $pad_spaces . $tokens_to_go[$ipad];
    }
    elsif ( $pad_spaces == -1 && $tokens_to_go[$ipad] eq ' ' ) {
        $tokens_to_go[$ipad] = "";
    }
    else {

        # shouldn't happen
        return;
    }

    $token_lengths_to_go[$ipad] += $pad_spaces;
    for ( my $i = $ipad ; $i <= $max_index_to_go ; $i++ ) {
        $summed_lengths_to_go[ $i + 1 ] += $pad_spaces;
    }
}

{
    my %is_math_op;

    BEGIN {

        @_ = qw( + - * / );
        @is_math_op{@_} = (1) x scalar(@_);
    }

    sub set_logical_padding {

        # Look at a batch of lines and see if extra padding can improve the
        # alignment when there are certain leading operators. Here is an
        # example, in which some extra space is introduced before
        # '( $year' to make it line up with the subsequent lines:
        #
        #       if (   ( $Year < 1601 )
        #           || ( $Year > 2899 )
        #           || ( $EndYear < 1601 )
        #           || ( $EndYear > 2899 ) )
        #       {
        #           &Error_OutOfRange;
        #       }
        #
        my ( $ri_first, $ri_last ) = @_;
        my $max_line = @$ri_first - 1;

        my ( $ibeg, $ibeg_next, $ibegm, $iend, $iendm, $ipad, $line,
            $pad_spaces,
            $tok_next, $type_next, $has_leading_op_next, $has_leading_op );

        # looking at each line of this batch..
        foreach $line ( 0 .. $max_line - 1 ) {

            # see if the next line begins with a logical operator
            $ibeg      = $$ri_first[$line];
            $iend      = $$ri_last[$line];
            $ibeg_next = $$ri_first[ $line + 1 ];
            $tok_next  = $tokens_to_go[$ibeg_next];
            $type_next = $types_to_go[$ibeg_next];

            $has_leading_op_next = ( $tok_next =~ /^\w/ )
              ? $is_chain_operator{$tok_next}      # + - * / : ? && ||
              : $is_chain_operator{$type_next};    # and, or

            next unless ($has_leading_op_next);

            # next line must not be at lesser depth
            next
              if ( $nesting_depth_to_go[$ibeg] >
                $nesting_depth_to_go[$ibeg_next] );

            # identify the token in this line to be padded on the left
            $ipad = undef;

            # handle lines at same depth...
            if ( $nesting_depth_to_go[$ibeg] ==
                $nesting_depth_to_go[$ibeg_next] )
            {

                # if this is not first line of the batch ...
                if ( $line > 0 ) {

                    # and we have leading operator..
                    next if $has_leading_op;

                    # Introduce padding if..
                    # 1. the previous line is at lesser depth, or
                    # 2. the previous line ends in an assignment
                    # 3. the previous line ends in a 'return'
                    # 4. the previous line ends in a comma
                    # Example 1: previous line at lesser depth
                    #       if (   ( $Year < 1601 )      # <- we are here but
                    #           || ( $Year > 2899 )      #  list has not yet
                    #           || ( $EndYear < 1601 )   # collapsed vertically
                    #           || ( $EndYear > 2899 ) )
                    #       {
                    #
                    # Example 2: previous line ending in assignment:
                    #    $leapyear =
                    #        $year % 4   ? 0     # <- We are here
                    #      : $year % 100 ? 1
                    #      : $year % 400 ? 0
                    #      : 1;
                    #
                    # Example 3: previous line ending in comma:
                    #    push @expr,
                    #        /test/   ? undef
                    #      : eval($_) ? 1
                    #      : eval($_) ? 1
                    #      :            0;

                   # be sure levels agree (do not indent after an indented 'if')
                    next
                      if ( $levels_to_go[$ibeg] ne $levels_to_go[$ibeg_next] );

                    # allow padding on first line after a comma but only if:
                    # (1) this is line 2 and
                    # (2) there are at more than three lines and
                    # (3) lines 3 and 4 have the same leading operator
                    # These rules try to prevent padding within a long
                    # comma-separated list.
                    my $ok_comma;
                    if (   $types_to_go[$iendm] eq ','
                        && $line == 1
                        && $max_line > 2 )
                    {
                        my $ibeg_next_next = $$ri_first[ $line + 2 ];
                        my $tok_next_next  = $tokens_to_go[$ibeg_next_next];
                        $ok_comma = $tok_next_next eq $tok_next;
                    }

                    next
                      unless (
                           $is_assignment{ $types_to_go[$iendm] }
                        || $ok_comma
                        || ( $nesting_depth_to_go[$ibegm] <
                            $nesting_depth_to_go[$ibeg] )
                        || (   $types_to_go[$iendm] eq 'k'
                            && $tokens_to_go[$iendm] eq 'return' )
                      );

                    # we will add padding before the first token
                    $ipad = $ibeg;
                }

                # for first line of the batch..
                else {

                    # WARNING: Never indent if first line is starting in a
                    # continued quote, which would change the quote.
                    next if $starting_in_quote;

                    # if this is text after closing '}'
                    # then look for an interior token to pad
                    if ( $types_to_go[$ibeg] eq '}' ) {

                    }

                    # otherwise, we might pad if it looks really good
                    else {

                        # we might pad token $ibeg, so be sure that it
                        # is at the same depth as the next line.
                        next
                          if ( $nesting_depth_to_go[$ibeg] !=
                            $nesting_depth_to_go[$ibeg_next] );

                        # We can pad on line 1 of a statement if at least 3
                        # lines will be aligned. Otherwise, it
                        # can look very confusing.

                 # We have to be careful not to pad if there are too few
                 # lines.  The current rule is:
                 # (1) in general we require at least 3 consecutive lines
                 # with the same leading chain operator token,
                 # (2) but an exception is that we only require two lines
                 # with leading colons if there are no more lines.  For example,
                 # the first $i in the following snippet would get padding
                 # by the second rule:
                 #
                 #   $i == 1 ? ( "First", "Color" )
                 # : $i == 2 ? ( "Then",  "Rarity" )
                 # :           ( "Then",  "Name" );

                        if ( $max_line > 1 ) {
                            my $leading_token = $tokens_to_go[$ibeg_next];
                            my $tokens_differ;

                            # never indent line 1 of a '.' series because
                            # previous line is most likely at same level.
                            # TODO: we should also look at the leasing_spaces
                            # of the last output line and skip if it is same
                            # as this line.
                            next if ( $leading_token eq '.' );

                            my $count = 1;
                            foreach my $l ( 2 .. 3 ) {
                                last if ( $line + $l > $max_line );
                                my $ibeg_next_next = $$ri_first[ $line + $l ];
                                if ( $tokens_to_go[$ibeg_next_next] ne
                                    $leading_token )
                                {
                                    $tokens_differ = 1;
                                    last;
                                }
                                $count++;
                            }
                            next if ($tokens_differ);
                            next if ( $count < 3 && $leading_token ne ':' );
                            $ipad = $ibeg;
                        }
                        else {
                            next;
                        }
                    }
                }
            }

            # find interior token to pad if necessary
            if ( !defined($ipad) ) {

                for ( my $i = $ibeg ; ( $i < $iend ) && !$ipad ; $i++ ) {

                    # find any unclosed container
                    next
                      unless ( $type_sequence_to_go[$i]
                        && $mate_index_to_go[$i] > $iend );

                    # find next nonblank token to pad
                    $ipad = $inext_to_go[$i];
                    last if ( $ipad > $iend );
                }
                last unless $ipad;
            }

            # We cannot pad a leading token at the lowest level because
            # it could cause a bug in which the starting indentation
            # level is guessed incorrectly each time the code is run
            # though perltidy, thus causing the code to march off to
            # the right.  For example, the following snippet would have
            # this problem:

##     ov_method mycan( $package, '(""' ),       $package
##  or ov_method mycan( $package, '(0+' ),       $package
##  or ov_method mycan( $package, '(bool' ),     $package
##  or ov_method mycan( $package, '(nomethod' ), $package;

            # If this snippet is within a block this won't happen
            # unless the user just processes the snippet alone within
            # an editor.  In that case either the user will see and
            # fix the problem or it will be corrected next time the
            # entire file is processed with perltidy.
            next if ( $ipad == 0 && $levels_to_go[$ipad] == 0 );

## THIS PATCH REMOVES THE FOLLOWING POOR PADDING (math.t) with -pbp, BUT
## IT DID MORE HARM THAN GOOD
##            ceil(
##                      $font->{'loca'}->{'glyphs'}[$x]->read->{'xMin'} * 1000
##                    / $upem
##            ),
##?            # do not put leading padding for just 2 lines of math
##?            if (   $ipad == $ibeg
##?                && $line > 0
##?                && $levels_to_go[$ipad] > $levels_to_go[ $ipad - 1 ]
##?                && $is_math_op{$type_next}
##?                && $line + 2 <= $max_line )
##?            {
##?                my $ibeg_next_next = $$ri_first[ $line + 2 ];
##?                my $type_next_next = $types_to_go[$ibeg_next_next];
##?                next if !$is_math_op{$type_next_next};
##?            }

            # next line must not be at greater depth
            my $iend_next = $$ri_last[ $line + 1 ];
            next
              if ( $nesting_depth_to_go[ $iend_next + 1 ] >
                $nesting_depth_to_go[$ipad] );

            # lines must be somewhat similar to be padded..
            my $inext_next = $inext_to_go[$ibeg_next];
            my $type       = $types_to_go[$ipad];
            my $type_next  = $types_to_go[ $ipad + 1 ];

            # see if there are multiple continuation lines
            my $logical_continuation_lines = 1;
            if ( $line + 2 <= $max_line ) {
                my $leading_token  = $tokens_to_go[$ibeg_next];
                my $ibeg_next_next = $$ri_first[ $line + 2 ];
                if (   $tokens_to_go[$ibeg_next_next] eq $leading_token
                    && $nesting_depth_to_go[$ibeg_next] eq
                    $nesting_depth_to_go[$ibeg_next_next] )
                {
                    $logical_continuation_lines++;
                }
            }

            # see if leading types match
            my $types_match = $types_to_go[$inext_next] eq $type;
            my $matches_without_bang;

            # if first line has leading ! then compare the following token
            if ( !$types_match && $type eq '!' ) {
                $types_match = $matches_without_bang =
                  $types_to_go[$inext_next] eq $types_to_go[ $ipad + 1 ];
            }

            if (

                # either we have multiple continuation lines to follow
                # and we are not padding the first token
                ( $logical_continuation_lines > 1 && $ipad > 0 )

                # or..
                || (

                    # types must match
                    $types_match

                    # and keywords must match if keyword
                    && !(
                           $type eq 'k'
                        && $tokens_to_go[$ipad] ne $tokens_to_go[$inext_next]
                    )
                )
              )
            {

                #----------------------begin special checks--------------
                #
                # SPECIAL CHECK 1:
                # A check is needed before we can make the pad.
                # If we are in a list with some long items, we want each
                # item to stand out.  So in the following example, the
                # first line beginning with '$casefold->' would look good
                # padded to align with the next line, but then it
                # would be indented more than the last line, so we
                # won't do it.
                #
                #  ok(
                #      $casefold->{code}         eq '0041'
                #        && $casefold->{status}  eq 'C'
                #        && $casefold->{mapping} eq '0061',
                #      'casefold 0x41'
                #  );
                #
                # Note:
                # It would be faster, and almost as good, to use a comma
                # count, and not pad if comma_count > 1 and the previous
                # line did not end with a comma.
                #
                my $ok_to_pad = 1;

                my $ibg   = $$ri_first[ $line + 1 ];
                my $depth = $nesting_depth_to_go[ $ibg + 1 ];

                # just use simplified formula for leading spaces to avoid
                # needless sub calls
                my $lsp = $levels_to_go[$ibg] + $ci_levels_to_go[$ibg];

                # look at each line beyond the next ..
                my $l = $line + 1;
                foreach $l ( $line + 2 .. $max_line ) {
                    my $ibg = $$ri_first[$l];

                    # quit looking at the end of this container
                    last
                      if ( $nesting_depth_to_go[ $ibg + 1 ] < $depth )
                      || ( $nesting_depth_to_go[$ibg] < $depth );

                    # cannot do the pad if a later line would be
                    # outdented more
                    if ( $levels_to_go[$ibg] + $ci_levels_to_go[$ibg] < $lsp ) {
                        $ok_to_pad = 0;
                        last;
                    }
                }

                # don't pad if we end in a broken list
                if ( $l == $max_line ) {
                    my $i2 = $$ri_last[$l];
                    if ( $types_to_go[$i2] eq '#' ) {
                        my $i1 = $$ri_first[$l];
                        next
                          if (
                            terminal_type( \@types_to_go, \@block_type_to_go,
                                $i1, $i2 ) eq ','
                          );
                    }
                }

                # SPECIAL CHECK 2:
                # a minus may introduce a quoted variable, and we will
                # add the pad only if this line begins with a bare word,
                # such as for the word 'Button' here:
                #    [
                #         Button      => "Print letter \"~$_\"",
                #        -command     => [ sub { print "$_[0]\n" }, $_ ],
                #        -accelerator => "Meta+$_"
                #    ];
                #
                #  On the other hand, if 'Button' is quoted, it looks best
                #  not to pad:
                #    [
                #        'Button'     => "Print letter \"~$_\"",
                #        -command     => [ sub { print "$_[0]\n" }, $_ ],
                #        -accelerator => "Meta+$_"
                #    ];
                if ( $types_to_go[$ibeg_next] eq 'm' ) {
                    $ok_to_pad = 0 if $types_to_go[$ibeg] eq 'Q';
                }

                next unless $ok_to_pad;

                #----------------------end special check---------------

                my $length_1 = total_line_length( $ibeg,      $ipad - 1 );
                my $length_2 = total_line_length( $ibeg_next, $inext_next - 1 );
                $pad_spaces = $length_2 - $length_1;

                # If the first line has a leading ! and the second does
                # not, then remove one space to try to align the next
                # leading characters, which are often the same.  For example:
                #  if (  !$ts
                #      || $ts == $self->Holder
                #      || $self->Holder->Type eq "Arena" )
                #
                # This usually helps readability, but if there are subsequent
                # ! operators things will still get messed up.  For example:
                #
                #  if (  !exists $Net::DNS::typesbyname{$qtype}
                #      && exists $Net::DNS::classesbyname{$qtype}
                #      && !exists $Net::DNS::classesbyname{$qclass}
                #      && exists $Net::DNS::typesbyname{$qclass} )
                # We can't fix that.
                if ($matches_without_bang) { $pad_spaces-- }

                # make sure this won't change if -lp is used
                my $indentation_1 = $leading_spaces_to_go[$ibeg];
                if ( ref($indentation_1) ) {
                    if ( $indentation_1->get_RECOVERABLE_SPACES() == 0 ) {
                        my $indentation_2 = $leading_spaces_to_go[$ibeg_next];
                        unless ( $indentation_2->get_RECOVERABLE_SPACES() == 0 )
                        {
                            $pad_spaces = 0;
                        }
                    }
                }

                # we might be able to handle a pad of -1 by removing a blank
                # token
                if ( $pad_spaces < 0 ) {

                    if ( $pad_spaces == -1 ) {
                        if ( $ipad > $ibeg && $types_to_go[ $ipad - 1 ] eq 'b' )
                        {
                            pad_token( $ipad - 1, $pad_spaces );
                        }
                    }
                    $pad_spaces = 0;
                }

                # now apply any padding for alignment
                if ( $ipad >= 0 && $pad_spaces ) {

                    my $length_t = total_line_length( $ibeg, $iend );
                    if ( $pad_spaces + $length_t <= maximum_line_length($ibeg) )
                    {
                        pad_token( $ipad, $pad_spaces );
                    }
                }
            }
        }
        continue {
            $iendm          = $iend;
            $ibegm          = $ibeg;
            $has_leading_op = $has_leading_op_next;
        }    # end of loop over lines
        return;
    }
}

sub correct_lp_indentation {

    # When the -lp option is used, we need to make a last pass through
    # each line to correct the indentation positions in case they differ
    # from the predictions.  This is necessary because perltidy uses a
    # predictor/corrector method for aligning with opening parens.  The
    # predictor is usually good, but sometimes stumbles.  The corrector
    # tries to patch things up once the actual opening paren locations
    # are known.
    my ( $ri_first, $ri_last ) = @_;
    my $do_not_pad = 0;

    #  Note on flag '$do_not_pad':
    #  We want to avoid a situation like this, where the aligner inserts
    #  whitespace before the '=' to align it with a previous '=', because
    #  otherwise the parens might become mis-aligned in a situation like
    #  this, where the '=' has become aligned with the previous line,
    #  pushing the opening '(' forward beyond where we want it.
    #
    #  $mkFloor::currentRoom = '';
    #  $mkFloor::c_entry     = $c->Entry(
    #                                 -width        => '10',
    #                                 -relief       => 'sunken',
    #                                 ...
    #                                 );
    #
    #  We leave it to the aligner to decide how to do this.

    # first remove continuation indentation if appropriate
    my $max_line = @$ri_first - 1;

    # looking at each line of this batch..
    my ( $ibeg, $iend );
    my $line;
    foreach $line ( 0 .. $max_line ) {
        $ibeg = $$ri_first[$line];
        $iend = $$ri_last[$line];

        # looking at each token in this output line..
        my $i;
        foreach $i ( $ibeg .. $iend ) {

            # How many space characters to place before this token
            # for special alignment.  Actual padding is done in the
            # continue block.

            # looking for next unvisited indentation item
            my $indentation = $leading_spaces_to_go[$i];
            if ( !$indentation->get_MARKED() ) {
                $indentation->set_MARKED(1);

                # looking for indentation item for which we are aligning
                # with parens, braces, and brackets
                next unless ( $indentation->get_ALIGN_PAREN() );

                # skip closed container on this line
                if ( $i > $ibeg ) {
                    my $im = max( $ibeg, $iprev_to_go[$i] );
                    if (   $type_sequence_to_go[$im]
                        && $mate_index_to_go[$im] <= $iend )
                    {
                        next;
                    }
                }

                if ( $line == 1 && $i == $ibeg ) {
                    $do_not_pad = 1;
                }

                # Ok, let's see what the error is and try to fix it
                my $actual_pos;
                my $predicted_pos = $indentation->get_SPACES();
                if ( $i > $ibeg ) {

                    # token is mid-line - use length to previous token
                    $actual_pos = total_line_length( $ibeg, $i - 1 );

                    # for mid-line token, we must check to see if all
                    # additional lines have continuation indentation,
                    # and remove it if so.  Otherwise, we do not get
                    # good alignment.
                    my $closing_index = $indentation->get_CLOSED();
                    if ( $closing_index > $iend ) {
                        my $ibeg_next = $$ri_first[ $line + 1 ];
                        if ( $ci_levels_to_go[$ibeg_next] > 0 ) {
                            undo_lp_ci( $line, $i, $closing_index, $ri_first,
                                $ri_last );
                        }
                    }
                }
                elsif ( $line > 0 ) {

                    # handle case where token starts a new line;
                    # use length of previous line
                    my $ibegm = $$ri_first[ $line - 1 ];
                    my $iendm = $$ri_last[ $line - 1 ];
                    $actual_pos = total_line_length( $ibegm, $iendm );

                    # follow -pt style
                    ++$actual_pos
                      if ( $types_to_go[ $iendm + 1 ] eq 'b' );
                }
                else {

                    # token is first character of first line of batch
                    $actual_pos = $predicted_pos;
                }

                my $move_right = $actual_pos - $predicted_pos;

                # done if no error to correct (gnu2.t)
                if ( $move_right == 0 ) {
                    $indentation->set_RECOVERABLE_SPACES($move_right);
                    next;
                }

                # if we have not seen closure for this indentation in
                # this batch, we can only pass on a request to the
                # vertical aligner
                my $closing_index = $indentation->get_CLOSED();

                if ( $closing_index < 0 ) {
                    $indentation->set_RECOVERABLE_SPACES($move_right);
                    next;
                }

                # If necessary, look ahead to see if there is really any
                # leading whitespace dependent on this whitespace, and
                # also find the longest line using this whitespace.
                # Since it is always safe to move left if there are no
                # dependents, we only need to do this if we may have
                # dependent nodes or need to move right.

                my $right_margin = 0;
                my $have_child   = $indentation->get_HAVE_CHILD();

                my %saw_indentation;
                my $line_count = 1;
                $saw_indentation{$indentation} = $indentation;

                if ( $have_child || $move_right > 0 ) {
                    $have_child = 0;
                    my $max_length = 0;
                    if ( $i == $ibeg ) {
                        $max_length = total_line_length( $ibeg, $iend );
                    }

                    # look ahead at the rest of the lines of this batch..
                    my $line_t;
                    foreach $line_t ( $line + 1 .. $max_line ) {
                        my $ibeg_t = $$ri_first[$line_t];
                        my $iend_t = $$ri_last[$line_t];
                        last if ( $closing_index <= $ibeg_t );

                        # remember all different indentation objects
                        my $indentation_t = $leading_spaces_to_go[$ibeg_t];
                        $saw_indentation{$indentation_t} = $indentation_t;
                        $line_count++;

                        # remember longest line in the group
                        my $length_t = total_line_length( $ibeg_t, $iend_t );
                        if ( $length_t > $max_length ) {
                            $max_length = $length_t;
                        }
                    }
                    $right_margin = maximum_line_length($ibeg) - $max_length;
                    if ( $right_margin < 0 ) { $right_margin = 0 }
                }

                my $first_line_comma_count =
                  grep { $_ eq ',' } @types_to_go[ $ibeg .. $iend ];
                my $comma_count = $indentation->get_COMMA_COUNT();
                my $arrow_count = $indentation->get_ARROW_COUNT();

                # This is a simple approximate test for vertical alignment:
                # if we broke just after an opening paren, brace, bracket,
                # and there are 2 or more commas in the first line,
                # and there are no '=>'s,
                # then we are probably vertically aligned.  We could set
                # an exact flag in sub scan_list, but this is good
                # enough.
                my $indentation_count = keys %saw_indentation;
                my $is_vertically_aligned =
                  (      $i == $ibeg
                      && $first_line_comma_count > 1
                      && $indentation_count == 1
                      && ( $arrow_count == 0 || $arrow_count == $line_count ) );

                # Make the move if possible ..
                if (

                    # we can always move left
                    $move_right < 0

                    # but we should only move right if we are sure it will
                    # not spoil vertical alignment
                    || ( $comma_count == 0 )
                    || ( $comma_count > 0 && !$is_vertically_aligned )
                  )
                {
                    my $move =
                      ( $move_right <= $right_margin )
                      ? $move_right
                      : $right_margin;

                    foreach ( keys %saw_indentation ) {
                        $saw_indentation{$_}
                          ->permanently_decrease_AVAILABLE_SPACES( -$move );
                    }
                }

                # Otherwise, record what we want and the vertical aligner
                # will try to recover it.
                else {
                    $indentation->set_RECOVERABLE_SPACES($move_right);
                }
            }
        }
    }
    return $do_not_pad;
}

# flush is called to output any tokens in the pipeline, so that
# an alternate source of lines can be written in the correct order

sub flush {
    destroy_one_line_block();
    output_line_to_go();
    Perl::Tidy::VerticalAligner::flush();
}

sub reset_block_text_accumulator {

    # save text after 'if' and 'elsif' to append after 'else'
    if ($accumulating_text_for_block) {

        if ( $accumulating_text_for_block =~ /^(if|elsif)$/ ) {
            push @{$rleading_block_if_elsif_text}, $leading_block_text;
        }
    }
    $accumulating_text_for_block        = "";
    $leading_block_text                 = "";
    $leading_block_text_level           = 0;
    $leading_block_text_length_exceeded = 0;
    $leading_block_text_line_number     = 0;
    $leading_block_text_line_length     = 0;
}

sub set_block_text_accumulator {
    my $i = shift;
    $accumulating_text_for_block = $tokens_to_go[$i];
    if ( $accumulating_text_for_block !~ /^els/ ) {
        $rleading_block_if_elsif_text = [];
    }
    $leading_block_text       = "";
    $leading_block_text_level = $levels_to_go[$i];
    $leading_block_text_line_number =
      $vertical_aligner_object->get_output_line_number();
    $leading_block_text_length_exceeded = 0;

    # this will contain the column number of the last character
    # of the closing side comment
    $leading_block_text_line_length =
      length($csc_last_label) +
      length($accumulating_text_for_block) +
      length( $rOpts->{'closing-side-comment-prefix'} ) +
      $leading_block_text_level * $rOpts_indent_columns + 3;
}

sub accumulate_block_text {
    my $i = shift;

    # accumulate leading text for -csc, ignoring any side comments
    if (   $accumulating_text_for_block
        && !$leading_block_text_length_exceeded
        && $types_to_go[$i] ne '#' )
    {

        my $added_length = $token_lengths_to_go[$i];
        $added_length += 1 if $i == 0;
        my $new_line_length = $leading_block_text_line_length + $added_length;

        # we can add this text if we don't exceed some limits..
        if (

            # we must not have already exceeded the text length limit
            length($leading_block_text) <
            $rOpts_closing_side_comment_maximum_text

            # and either:
            # the new total line length must be below the line length limit
            # or the new length must be below the text length limit
            # (ie, we may allow one token to exceed the text length limit)
            && (
                $new_line_length <
                maximum_line_length_for_level($leading_block_text_level)

                || length($leading_block_text) + $added_length <
                $rOpts_closing_side_comment_maximum_text
            )

            # UNLESS: we are adding a closing paren before the brace we seek.
            # This is an attempt to avoid situations where the ... to be
            # added are longer than the omitted right paren, as in:

            #   foreach my $item (@a_rather_long_variable_name_here) {
            #      &whatever;
            #   } ## end foreach my $item (@a_rather_long_variable_name_here...

            || (
                $tokens_to_go[$i] eq ')'
                && (
                    (
                           $i + 1 <= $max_index_to_go
                        && $block_type_to_go[ $i + 1 ] eq
                        $accumulating_text_for_block
                    )
                    || (   $i + 2 <= $max_index_to_go
                        && $block_type_to_go[ $i + 2 ] eq
                        $accumulating_text_for_block )
                )
            )
          )
        {

            # add an extra space at each newline
            if ( $i == 0 ) { $leading_block_text .= ' ' }

            # add the token text
            $leading_block_text .= $tokens_to_go[$i];
            $leading_block_text_line_length = $new_line_length;
        }

        # show that text was truncated if necessary
        elsif ( $types_to_go[$i] ne 'b' ) {
            $leading_block_text_length_exceeded = 1;
## Please see file perltidy.ERR
            $leading_block_text .= '...';
        }
    }
}

{
    my %is_if_elsif_else_unless_while_until_for_foreach;

    BEGIN {

        # These block types may have text between the keyword and opening
        # curly.  Note: 'else' does not, but must be included to allow trailing
        # if/elsif text to be appended.
        # patch for SWITCH/CASE: added 'case' and 'when'
        @_ = qw(if elsif else unless while until for foreach case when);
        @is_if_elsif_else_unless_while_until_for_foreach{@_} =
          (1) x scalar(@_);
    }

    sub accumulate_csc_text {

        # called once per output buffer when -csc is used. Accumulates
        # the text placed after certain closing block braces.
        # Defines and returns the following for this buffer:

        my $block_leading_text = "";    # the leading text of the last '}'
        my $rblock_leading_if_elsif_text;
        my $i_block_leading_text =
          -1;    # index of token owning block_leading_text
        my $block_line_count    = 100;    # how many lines the block spans
        my $terminal_type       = 'b';    # type of last nonblank token
        my $i_terminal          = 0;      # index of last nonblank token
        my $terminal_block_type = "";

        # update most recent statement label
        $csc_last_label = "" unless ($csc_last_label);
        if ( $types_to_go[0] eq 'J' ) { $csc_last_label = $tokens_to_go[0] }
        my $block_label = $csc_last_label;

        # Loop over all tokens of this batch
        for my $i ( 0 .. $max_index_to_go ) {
            my $type       = $types_to_go[$i];
            my $block_type = $block_type_to_go[$i];
            my $token      = $tokens_to_go[$i];

            # remember last nonblank token type
            if ( $type ne '#' && $type ne 'b' ) {
                $terminal_type       = $type;
                $terminal_block_type = $block_type;
                $i_terminal          = $i;
            }

            my $type_sequence = $type_sequence_to_go[$i];
            if ( $block_type && $type_sequence ) {

                if ( $token eq '}' ) {

                    # restore any leading text saved when we entered this block
                    if ( defined( $block_leading_text{$type_sequence} ) ) {
                        ( $block_leading_text, $rblock_leading_if_elsif_text )
                          = @{ $block_leading_text{$type_sequence} };
                        $i_block_leading_text = $i;
                        delete $block_leading_text{$type_sequence};
                        $rleading_block_if_elsif_text =
                          $rblock_leading_if_elsif_text;
                    }

                    if ( defined( $csc_block_label{$type_sequence} ) ) {
                        $block_label = $csc_block_label{$type_sequence};
                        delete $csc_block_label{$type_sequence};
                    }

                    # if we run into a '}' then we probably started accumulating
                    # at something like a trailing 'if' clause..no harm done.
                    if (   $accumulating_text_for_block
                        && $levels_to_go[$i] <= $leading_block_text_level )
                    {
                        my $lev = $levels_to_go[$i];
                        reset_block_text_accumulator();
                    }

                    if ( defined( $block_opening_line_number{$type_sequence} ) )
                    {
                        my $output_line_number =
                          $vertical_aligner_object->get_output_line_number();
                        $block_line_count =
                          $output_line_number -
                          $block_opening_line_number{$type_sequence} + 1;
                        delete $block_opening_line_number{$type_sequence};
                    }
                    else {

                        # Error: block opening line undefined for this line..
                        # This shouldn't be possible, but it is not a
                        # significant problem.
                    }
                }

                elsif ( $token eq '{' ) {

                    my $line_number =
                      $vertical_aligner_object->get_output_line_number();
                    $block_opening_line_number{$type_sequence} = $line_number;

                    # set a label for this block, except for
                    # a bare block which already has the label
                    # A label can only be used on the next {
                    if ( $block_type =~ /:$/ ) { $csc_last_label = "" }
                    $csc_block_label{$type_sequence} = $csc_last_label;
                    $csc_last_label = "";

                    if (   $accumulating_text_for_block
                        && $levels_to_go[$i] == $leading_block_text_level )
                    {

                        if ( $accumulating_text_for_block eq $block_type ) {

                            # save any leading text before we enter this block
                            $block_leading_text{$type_sequence} = [
                                $leading_block_text,
                                $rleading_block_if_elsif_text
                            ];
                            $block_opening_line_number{$type_sequence} =
                              $leading_block_text_line_number;
                            reset_block_text_accumulator();
                        }
                        else {

                            # shouldn't happen, but not a serious error.
                            # We were accumulating -csc text for block type
                            # $accumulating_text_for_block and unexpectedly
                            # encountered a '{' for block type $block_type.
                        }
                    }
                }
            }

            if (   $type eq 'k'
                && $csc_new_statement_ok
                && $is_if_elsif_else_unless_while_until_for_foreach{$token}
                && $token =~ /$closing_side_comment_list_pattern/o )
            {
                set_block_text_accumulator($i);
            }
            else {

                # note: ignoring type 'q' because of tricks being played
                # with 'q' for hanging side comments
                if ( $type ne 'b' && $type ne '#' && $type ne 'q' ) {
                    $csc_new_statement_ok =
                      ( $block_type || $type eq 'J' || $type eq ';' );
                }
                if (   $type eq ';'
                    && $accumulating_text_for_block
                    && $levels_to_go[$i] == $leading_block_text_level )
                {
                    reset_block_text_accumulator();
                }
                else {
                    accumulate_block_text($i);
                }
            }
        }

        # Treat an 'else' block specially by adding preceding 'if' and
        # 'elsif' text.  Otherwise, the 'end else' is not helpful,
        # especially for cuddled-else formatting.
        if ( $terminal_block_type =~ /^els/ && $rblock_leading_if_elsif_text ) {
            $block_leading_text =
              make_else_csc_text( $i_terminal, $terminal_block_type,
                $block_leading_text, $rblock_leading_if_elsif_text );
        }

        # if this line ends in a label then remember it for the next pass
        $csc_last_label = "";
        if ( $terminal_type eq 'J' ) {
            $csc_last_label = $tokens_to_go[$i_terminal];
        }

        return ( $terminal_type, $i_terminal, $i_block_leading_text,
            $block_leading_text, $block_line_count, $block_label );
    }
}

sub make_else_csc_text {

    # create additional -csc text for an 'else' and optionally 'elsif',
    # depending on the value of switch
    # $rOpts_closing_side_comment_else_flag:
    #
    #  = 0 add 'if' text to trailing else
    #  = 1 same as 0 plus:
    #      add 'if' to 'elsif's if can fit in line length
    #      add last 'elsif' to trailing else if can fit in one line
    #  = 2 same as 1 but do not check if exceed line length
    #
    # $rif_elsif_text = a reference to a list of all previous closing
    # side comments created for this if block
    #
    my ( $i_terminal, $block_type, $block_leading_text, $rif_elsif_text ) = @_;
    my $csc_text = $block_leading_text;

    if (   $block_type eq 'elsif'
        && $rOpts_closing_side_comment_else_flag == 0 )
    {
        return $csc_text;
    }

    my $count = @{$rif_elsif_text};
    return $csc_text unless ($count);

    my $if_text = '[ if' . $rif_elsif_text->[0];

    # always show the leading 'if' text on 'else'
    if ( $block_type eq 'else' ) {
        $csc_text .= $if_text;
    }

    # see if that's all
    if ( $rOpts_closing_side_comment_else_flag == 0 ) {
        return $csc_text;
    }

    my $last_elsif_text = "";
    if ( $count > 1 ) {
        $last_elsif_text = ' [elsif' . $rif_elsif_text->[ $count - 1 ];
        if ( $count > 2 ) { $last_elsif_text = ' [...' . $last_elsif_text; }
    }

    # tentatively append one more item
    my $saved_text = $csc_text;
    if ( $block_type eq 'else' ) {
        $csc_text .= $last_elsif_text;
    }
    else {
        $csc_text .= ' ' . $if_text;
    }

    # all done if no length checks requested
    if ( $rOpts_closing_side_comment_else_flag == 2 ) {
        return $csc_text;
    }

    # undo it if line length exceeded
    my $length =
      length($csc_text) +
      length($block_type) +
      length( $rOpts->{'closing-side-comment-prefix'} ) +
      $levels_to_go[$i_terminal] * $rOpts_indent_columns + 3;
    if ( $length > maximum_line_length_for_level($leading_block_text_level) ) {
        $csc_text = $saved_text;
    }
    return $csc_text;
}

{    # sub balance_csc_text

    my %matching_char;

    BEGIN {
        %matching_char = (
            '{' => '}',
            '(' => ')',
            '[' => ']',
            '}' => '{',
            ')' => '(',
            ']' => '[',
        );
    }

    sub balance_csc_text {

        # Append characters to balance a closing side comment so that editors
        # such as vim can correctly jump through code.
        # Simple Example:
        #  input  = ## end foreach my $foo ( sort { $b  ...
        #  output = ## end foreach my $foo ( sort { $b  ...})

        # NOTE: This routine does not currently filter out structures within
        # quoted text because the bounce algorithims in text editors do not
        # necessarily do this either (a version of vim was checked and
        # did not do this).

        # Some complex examples which will cause trouble for some editors:
        #  while ( $mask_string =~ /\{[^{]*?\}/g ) {
        #  if ( $mask_str =~ /\}\s*els[^\{\}]+\{$/ ) {
        #  if ( $1 eq '{' ) {
        # test file test1/braces.pl has many such examples.

        my ($csc) = @_;

        # loop to examine characters one-by-one, RIGHT to LEFT and
        # build a balancing ending, LEFT to RIGHT.
        for ( my $pos = length($csc) - 1 ; $pos >= 0 ; $pos-- ) {

            my $char = substr( $csc, $pos, 1 );

            # ignore everything except structural characters
            next unless ( $matching_char{$char} );

            # pop most recently appended character
            my $top = chop($csc);

            # push it back plus the mate to the newest character
            # unless they balance each other.
            $csc = $csc . $top . $matching_char{$char} unless $top eq $char;
        }

        # return the balanced string
        return $csc;
    }
}

sub add_closing_side_comment {

    # add closing side comments after closing block braces if -csc used
    my $cscw_block_comment;

    #---------------------------------------------------------------
    # Step 1: loop through all tokens of this line to accumulate
    # the text needed to create the closing side comments. Also see
    # how the line ends.
    #---------------------------------------------------------------

    my ( $terminal_type, $i_terminal, $i_block_leading_text,
        $block_leading_text, $block_line_count, $block_label )
      = accumulate_csc_text();

    #---------------------------------------------------------------
    # Step 2: make the closing side comment if this ends a block
    #---------------------------------------------------------------
    my $have_side_comment = $i_terminal != $max_index_to_go;

    # if this line might end in a block closure..
    if (
        $terminal_type eq '}'

        # ..and either
        && (

            # the block is long enough
            ( $block_line_count >= $rOpts->{'closing-side-comment-interval'} )

            # or there is an existing comment to check
            || (   $have_side_comment
                && $rOpts->{'closing-side-comment-warnings'} )
        )

        # .. and if this is one of the types of interest
        && $block_type_to_go[$i_terminal] =~
        /$closing_side_comment_list_pattern/o

        # .. but not an anonymous sub
        # These are not normally of interest, and their closing braces are
        # often followed by commas or semicolons anyway.  This also avoids
        # possible erratic output due to line numbering inconsistencies
        # in the cases where their closing braces terminate a line.
        && $block_type_to_go[$i_terminal] ne 'sub'

        # ..and the corresponding opening brace must is not in this batch
        # (because we do not need to tag one-line blocks, although this
        # should also be caught with a positive -csci value)
        && $mate_index_to_go[$i_terminal] < 0

        # ..and either
        && (

            # this is the last token (line doesn't have a side comment)
            !$have_side_comment

            # or the old side comment is a closing side comment
            || $tokens_to_go[$max_index_to_go] =~
            /$closing_side_comment_prefix_pattern/o
        )
      )
    {

        # then make the closing side comment text
        if ($block_label) { $block_label .= " " }
        my $token =
"$rOpts->{'closing-side-comment-prefix'} $block_label$block_type_to_go[$i_terminal]";

        # append any extra descriptive text collected above
        if ( $i_block_leading_text == $i_terminal ) {
            $token .= $block_leading_text;
        }

        $token = balance_csc_text($token)
          if $rOpts->{'closing-side-comments-balanced'};

        $token =~ s/\s*$//;    # trim any trailing whitespace

        # handle case of existing closing side comment
        if ($have_side_comment) {

            # warn if requested and tokens differ significantly
            if ( $rOpts->{'closing-side-comment-warnings'} ) {
                my $old_csc = $tokens_to_go[$max_index_to_go];
                my $new_csc = $token;
                $new_csc =~ s/\s+//g;            # trim all whitespace
                $old_csc =~ s/\s+//g;            # trim all whitespace
                $new_csc =~ s/[\]\)\}\s]*$//;    # trim trailing structures
                $old_csc =~ s/[\]\)\}\s]*$//;    # trim trailing structures
                $new_csc =~ s/(\.\.\.)$//;       # trim trailing '...'
                my $new_trailing_dots = $1;
                $old_csc =~ s/(\.\.\.)\s*$//;    # trim trailing '...'

                # Patch to handle multiple closing side comments at
                # else and elsif's.  These have become too complicated
                # to check, so if we see an indication of
                # '[ if' or '[ # elsif', then assume they were made
                # by perltidy.
                if ( $block_type_to_go[$i_terminal] eq 'else' ) {
                    if ( $old_csc =~ /\[\s*elsif/ ) { $old_csc = $new_csc }
                }
                elsif ( $block_type_to_go[$i_terminal] eq 'elsif' ) {
                    if ( $old_csc =~ /\[\s*if/ ) { $old_csc = $new_csc }
                }

                # if old comment is contained in new comment,
                # only compare the common part.
                if ( length($new_csc) > length($old_csc) ) {
                    $new_csc = substr( $new_csc, 0, length($old_csc) );
                }

                # if the new comment is shorter and has been limited,
                # only compare the common part.
                if ( length($new_csc) < length($old_csc)
                    && $new_trailing_dots )
                {
                    $old_csc = substr( $old_csc, 0, length($new_csc) );
                }

                # any remaining difference?
                if ( $new_csc ne $old_csc ) {

                    # just leave the old comment if we are below the threshold
                    # for creating side comments
                    if ( $block_line_count <
                        $rOpts->{'closing-side-comment-interval'} )
                    {
                        $token = undef;
                    }

                    # otherwise we'll make a note of it
                    else {

                        warning(
"perltidy -cscw replaced: $tokens_to_go[$max_index_to_go]\n"
                        );

                     # save the old side comment in a new trailing block comment
                        my ( $day, $month, $year ) = (localtime)[ 3, 4, 5 ];
                        $year  += 1900;
                        $month += 1;
                        $cscw_block_comment =
"## perltidy -cscw $year-$month-$day: $tokens_to_go[$max_index_to_go]";
                    }
                }
                else {

                    # No differences.. we can safely delete old comment if we
                    # are below the threshold
                    if ( $block_line_count <
                        $rOpts->{'closing-side-comment-interval'} )
                    {
                        $token = undef;
                        unstore_token_to_go()
                          if ( $types_to_go[$max_index_to_go] eq '#' );
                        unstore_token_to_go()
                          if ( $types_to_go[$max_index_to_go] eq 'b' );
                    }
                }
            }

            # switch to the new csc (unless we deleted it!)
            $tokens_to_go[$max_index_to_go] = $token if $token;
        }

        # handle case of NO existing closing side comment
        else {

            # insert the new side comment into the output token stream
            my $type          = '#';
            my $block_type    = '';
            my $type_sequence = '';
            my $container_environment =
              $container_environment_to_go[$max_index_to_go];
            my $level                = $levels_to_go[$max_index_to_go];
            my $slevel               = $nesting_depth_to_go[$max_index_to_go];
            my $no_internal_newlines = 0;

            my $nesting_blocks     = $nesting_blocks_to_go[$max_index_to_go];
            my $ci_level           = $ci_levels_to_go[$max_index_to_go];
            my $in_continued_quote = 0;

            # first insert a blank token
            insert_new_token_to_go( ' ', 'b', $slevel, $no_internal_newlines );

            # then the side comment
            insert_new_token_to_go( $token, $type, $slevel,
                $no_internal_newlines );
        }
    }
    return $cscw_block_comment;
}

sub previous_nonblank_token {
    my ($i)  = @_;
    my $name = "";
    my $im   = $i - 1;
    return "" if ( $im < 0 );
    if ( $types_to_go[$im] eq 'b' ) { $im--; }
    return "" if ( $im < 0 );
    $name = $tokens_to_go[$im];

    # prepend any sub name to an isolated -> to avoid unwanted alignments
    # [test case is test8/penco.pl]
    if ( $name eq '->' ) {
        $im--;
        if ( $im >= 0 && $types_to_go[$im] ne 'b' ) {
            $name = $tokens_to_go[$im] . $name;
        }
    }
    return $name;
}

sub send_lines_to_vertical_aligner {

    my ( $ri_first, $ri_last, $do_not_pad ) = @_;

    my $rindentation_list = [0];    # ref to indentations for each line

    # define the array @matching_token_to_go for the output tokens
    # which will be non-blank for each special token (such as =>)
    # for which alignment is required.
    set_vertical_alignment_markers( $ri_first, $ri_last );

    # flush if necessary to avoid unwanted alignment
    my $must_flush = 0;
    if ( @$ri_first > 1 ) {

        # flush before a long if statement
        if ( $types_to_go[0] eq 'k' && $tokens_to_go[0] =~ /^(if|unless)$/ ) {
            $must_flush = 1;
        }
    }
    if ($must_flush) {
        Perl::Tidy::VerticalAligner::flush();
    }

    undo_ci( $ri_first, $ri_last );

    set_logical_padding( $ri_first, $ri_last );

    # loop to prepare each line for shipment
    my $n_last_line = @$ri_first - 1;
    my $in_comma_list;
    for my $n ( 0 .. $n_last_line ) {
        my $ibeg = $$ri_first[$n];
        my $iend = $$ri_last[$n];

        my ( $rtokens, $rfields, $rpatterns ) =
          make_alignment_patterns( $ibeg, $iend );

        # Set flag to show how much level changes between this line
        # and the next line, if we have it.
        my $ljump = 0;
        if ( $n < $n_last_line ) {
            my $ibegp = $$ri_first[ $n + 1 ];
            $ljump = $levels_to_go[$ibegp] - $levels_to_go[$iend];
        }

        my ( $indentation, $lev, $level_end, $terminal_type,
            $is_semicolon_terminated, $is_outdented_line )
          = set_adjusted_indentation( $ibeg, $iend, $rfields, $rpatterns,
            $ri_first, $ri_last, $rindentation_list, $ljump );

        # we will allow outdenting of long lines..
        my $outdent_long_lines = (

            # which are long quotes, if allowed
            ( $types_to_go[$ibeg] eq 'Q' && $rOpts->{'outdent-long-quotes'} )

            # which are long block comments, if allowed
              || (
                   $types_to_go[$ibeg] eq '#'
                && $rOpts->{'outdent-long-comments'}

                # but not if this is a static block comment
                && !$is_static_block_comment
              )
        );

        my $level_jump =
          $nesting_depth_to_go[ $iend + 1 ] - $nesting_depth_to_go[$ibeg];

        my $rvertical_tightness_flags =
          set_vertical_tightness_flags( $n, $n_last_line, $ibeg, $iend,
            $ri_first, $ri_last );

        # flush an outdented line to avoid any unwanted vertical alignment
        Perl::Tidy::VerticalAligner::flush() if ($is_outdented_line);

        # Set a flag at the final ':' of a ternary chain to request
        # vertical alignment of the final term.  Here is a
        # slightly complex example:
        #
        # $self->{_text} = (
        #    !$section        ? ''
        #   : $type eq 'item' ? "the $section entry"
        #   :                   "the section on $section"
        # )
        # . (
        #   $page
        #   ? ( $section ? ' in ' : '' ) . "the $page$page_ext manpage"
        #   : ' elsewhere in this document'
        # );
        #
        my $is_terminal_ternary = 0;
        if (   $tokens_to_go[$ibeg] eq ':'
            || $n > 0 && $tokens_to_go[ $$ri_last[ $n - 1 ] ] eq ':' )
        {
            my $last_leading_type = ":";
            if ( $n > 0 ) {
                my $iprev = $$ri_first[ $n - 1 ];
                $last_leading_type = $types_to_go[$iprev];
            }
            if (   $terminal_type ne ';'
                && $n_last_line > $n
                && $level_end == $lev )
            {
                my $inext = $$ri_first[ $n + 1 ];
                $level_end     = $levels_to_go[$inext];
                $terminal_type = $types_to_go[$inext];
            }

            $is_terminal_ternary = $last_leading_type eq ':'
              && ( ( $terminal_type eq ';' && $level_end <= $lev )
                || ( $terminal_type ne ':' && $level_end < $lev ) )

              # the termainal term must not contain any ternary terms, as in
              # my $ECHO = (
              #       $Is_MSWin32 ? ".\\echo$$"
              #     : $Is_MacOS   ? ":echo$$"
              #     : ( $Is_NetWare ? "echo$$" : "./echo$$" )
              # );
              && !grep /^[\?\:]$/, @types_to_go[ $ibeg + 1 .. $iend ];
        }

        # send this new line down the pipe
        my $forced_breakpoint = $forced_breakpoint_to_go[$iend];
        Perl::Tidy::VerticalAligner::valign_input(
            $lev,
            $level_end,
            $indentation,
            $rfields,
            $rtokens,
            $rpatterns,
            $forced_breakpoint_to_go[$iend] || $in_comma_list,
            $outdent_long_lines,
            $is_terminal_ternary,
            $is_semicolon_terminated,
            $do_not_pad,
            $rvertical_tightness_flags,
            $level_jump,
        );
        $in_comma_list =
          $tokens_to_go[$iend] eq ',' && $forced_breakpoint_to_go[$iend];

        # flush an outdented line to avoid any unwanted vertical alignment
        Perl::Tidy::VerticalAligner::flush() if ($is_outdented_line);

        $do_not_pad = 0;

        # Set flag indicating if this line ends in an opening
        # token and is very short, so that a blank line is not
        # needed if the subsequent line is a comment.
        # Examples of what we are looking for:
        #   {
        #   && (
        #   BEGIN {
        #   default {
        #   sub {
        $last_output_short_opening_token

          # line ends in opening token
          = $types_to_go[$iend] =~ /^[\{\(\[L]$/

          # and either
          && (
            # line has either single opening token
            $iend == $ibeg

            # or is a single token followed by opening token.
            # Note that sub identifiers have blanks like 'sub doit'
            || ( $iend - $ibeg <= 2 && $tokens_to_go[$ibeg] !~ /\s+/ )
          )

          # and limit total to 10 character widths
          && token_sequence_length( $ibeg, $iend ) <= 10;

    }    # end of loop to output each line

    # remember indentation of lines containing opening containers for
    # later use by sub set_adjusted_indentation
    save_opening_indentation( $ri_first, $ri_last, $rindentation_list );
}

{    # begin make_alignment_patterns

    my %block_type_map;
    my %keyword_map;

    BEGIN {

        # map related block names into a common name to
        # allow alignment
        %block_type_map = (
            'unless'  => 'if',
            'else'    => 'if',
            'elsif'   => 'if',
            'when'    => 'if',
            'default' => 'if',
            'case'    => 'if',
            'sort'    => 'map',
            'grep'    => 'map',
        );

        # map certain keywords to the same 'if' class to align
        # long if/elsif sequences. [elsif.pl]
        %keyword_map = (
            'unless'  => 'if',
            'else'    => 'if',
            'elsif'   => 'if',
            'when'    => 'given',
            'default' => 'given',
            'case'    => 'switch',

            # treat an 'undef' similar to numbers and quotes
            'undef' => 'Q',
        );
    }

    sub make_alignment_patterns {

        # Here we do some important preliminary work for the
        # vertical aligner.  We create three arrays for one
        # output line. These arrays contain strings that can
        # be tested by the vertical aligner to see if
        # consecutive lines can be aligned vertically.
        #
        # The three arrays are indexed on the vertical
        # alignment fields and are:
        # @tokens - a list of any vertical alignment tokens for this line.
        #   These are tokens, such as '=' '&&' '#' etc which
        #   we want to might align vertically.  These are
        #   decorated with various information such as
        #   nesting depth to prevent unwanted vertical
        #   alignment matches.
        # @fields - the actual text of the line between the vertical alignment
        #   tokens.
        # @patterns - a modified list of token types, one for each alignment
        #   field.  These should normally each match before alignment is
        #   allowed, even when the alignment tokens match.
        my ( $ibeg, $iend ) = @_;
        my @tokens   = ();
        my @fields   = ();
        my @patterns = ();
        my $i_start  = $ibeg;
        my $i;

        my $depth                 = 0;
        my @container_name        = ("");
        my @multiple_comma_arrows = (undef);

        my $j = 0;    # field index

        $patterns[0] = "";
        for $i ( $ibeg .. $iend ) {

            # Keep track of containers balanced on this line only.
            # These are used below to prevent unwanted cross-line alignments.
            # Unbalanced containers already avoid aligning across
            # container boundaries.
            if ( $tokens_to_go[$i] eq '(' ) {

                # if container is balanced on this line...
                my $i_mate = $mate_index_to_go[$i];
                if ( $i_mate > $i && $i_mate <= $iend ) {
                    $depth++;
                    my $seqno = $type_sequence_to_go[$i];
                    my $count = comma_arrow_count($seqno);
                    $multiple_comma_arrows[$depth] = $count && $count > 1;

                    # Append the previous token name to make the container name
                    # more unique.  This name will also be given to any commas
                    # within this container, and it helps avoid undesirable
                    # alignments of different types of containers.
                    my $name = previous_nonblank_token($i);
                    $name =~ s/^->//;
                    $container_name[$depth] = "+" . $name;

                    # Make the container name even more unique if necessary.
                    # If we are not vertically aligning this opening paren,
                    # append a character count to avoid bad alignment because
                    # it usually looks bad to align commas within continers
                    # for which the opening parens do not align.  Here
                    # is an example very BAD alignment of commas (because
                    # the atan2 functions are not all aligned):
                    #    $XY =
                    #      $X * $RTYSQP1 * atan2( $X, $RTYSQP1 ) +
                    #      $Y * $RTXSQP1 * atan2( $Y, $RTXSQP1 ) -
                    #      $X * atan2( $X,            1 ) -
                    #      $Y * atan2( $Y,            1 );
                    #
                    # On the other hand, it is usually okay to align commas if
                    # opening parens align, such as:
                    #    glVertex3d( $cx + $s * $xs, $cy,            $z );
                    #    glVertex3d( $cx,            $cy + $s * $ys, $z );
                    #    glVertex3d( $cx - $s * $xs, $cy,            $z );
                    #    glVertex3d( $cx,            $cy - $s * $ys, $z );
                    #
                    # To distinguish between these situations, we will
                    # append the length of the line from the previous matching
                    # token, or beginning of line, to the function name.  This
                    # will allow the vertical aligner to reject undesirable
                    # matches.

                    # if we are not aligning on this paren...
                    if ( $matching_token_to_go[$i] eq '' ) {

                        # Sum length from previous alignment, or start of line.
                        my $len =
                          ( $i_start == $ibeg )
                          ? total_line_length( $i_start, $i - 1 )
                          : token_sequence_length( $i_start, $i - 1 );

                        # tack length onto the container name to make unique
                        $container_name[$depth] .= "-" . $len;
                    }
                }
            }
            elsif ( $tokens_to_go[$i] eq ')' ) {
                $depth-- if $depth > 0;
            }

            # if we find a new synchronization token, we are done with
            # a field
            if ( $i > $i_start && $matching_token_to_go[$i] ne '' ) {

                my $tok = my $raw_tok = $matching_token_to_go[$i];

                # make separators in different nesting depths unique
                # by appending the nesting depth digit.
                if ( $raw_tok ne '#' ) {
                    $tok .= "$nesting_depth_to_go[$i]";
                }

                # also decorate commas with any container name to avoid
                # unwanted cross-line alignments.
                if ( $raw_tok eq ',' || $raw_tok eq '=>' ) {
                    if ( $container_name[$depth] ) {
                        $tok .= $container_name[$depth];
                    }
                }

                # Patch to avoid aligning leading and trailing if, unless.
                # Mark trailing if, unless statements with container names.
                # This makes them different from leading if, unless which
                # are not so marked at present.  If we ever need to name
                # them too, we could use ci to distinguish them.
                # Example problem to avoid:
                #    return ( 2, "DBERROR" )
                #      if ( $retval == 2 );
                #    if   ( scalar @_ ) {
                #        my ( $a, $b, $c, $d, $e, $f ) = @_;
                #    }
                if ( $raw_tok eq '(' ) {
                    my $ci = $ci_levels_to_go[$ibeg];
                    if (   $container_name[$depth] =~ /^\+(if|unless)/
                        && $ci )
                    {
                        $tok .= $container_name[$depth];
                    }
                }

                # Decorate block braces with block types to avoid
                # unwanted alignments such as the following:
                # foreach ( @{$routput_array} ) { $fh->print($_) }
                # eval                          { $fh->close() };
                if ( $raw_tok eq '{' && $block_type_to_go[$i] ) {
                    my $block_type = $block_type_to_go[$i];

                    # map certain related block types to allow
                    # else blocks to align
                    $block_type = $block_type_map{$block_type}
                      if ( defined( $block_type_map{$block_type} ) );

                    # remove sub names to allow one-line sub braces to align
                    # regardless of name
                    if ( $block_type =~ /^sub / ) { $block_type = 'sub' }

                    # allow all control-type blocks to align
                    if ( $block_type =~ /^[A-Z]+$/ ) { $block_type = 'BEGIN' }

                    $tok .= $block_type;
                }

                # concatenate the text of the consecutive tokens to form
                # the field
                push( @fields,
                    join( '', @tokens_to_go[ $i_start .. $i - 1 ] ) );

                # store the alignment token for this field
                push( @tokens, $tok );

                # get ready for the next batch
                $i_start = $i;
                $j++;
                $patterns[$j] = "";
            }

            # continue accumulating tokens
            # handle non-keywords..
            if ( $types_to_go[$i] ne 'k' ) {
                my $type = $types_to_go[$i];

                # Mark most things before arrows as a quote to
                # get them to line up. Testfile: mixed.pl.
                if ( ( $i < $iend - 1 ) && ( $type =~ /^[wnC]$/ ) ) {
                    my $next_type = $types_to_go[ $i + 1 ];
                    my $i_next_nonblank =
                      ( ( $next_type eq 'b' ) ? $i + 2 : $i + 1 );

                    if ( $types_to_go[$i_next_nonblank] eq '=>' ) {
                        $type = 'Q';

                        # Patch to ignore leading minus before words,
                        # by changing pattern 'mQ' into just 'Q',
                        # so that we can align things like this:
                        #  Button   => "Print letter \"~$_\"",
                        #  -command => [ sub { print "$_[0]\n" }, $_ ],
                        if ( $patterns[$j] eq 'm' ) { $patterns[$j] = "" }
                    }
                }

                # patch to make numbers and quotes align
                if ( $type eq 'n' ) { $type = 'Q' }

                # patch to ignore any ! in patterns
                if ( $type eq '!' ) { $type = '' }

                $patterns[$j] .= $type;
            }

            # for keywords we have to use the actual text
            else {

                my $tok = $tokens_to_go[$i];

                # but map certain keywords to a common string to allow
                # alignment.
                $tok = $keyword_map{$tok}
                  if ( defined( $keyword_map{$tok} ) );
                $patterns[$j] .= $tok;
            }
        }

        # done with this line .. join text of tokens to make the last field
        push( @fields, join( '', @tokens_to_go[ $i_start .. $iend ] ) );
        return ( \@tokens, \@fields, \@patterns );
    }

}    # end make_alignment_patterns

{    # begin unmatched_indexes

    # closure to keep track of unbalanced containers.
    # arrays shared by the routines in this block:
    my @unmatched_opening_indexes_in_this_batch;
    my @unmatched_closing_indexes_in_this_batch;
    my %comma_arrow_count;

    sub is_unbalanced_batch {
        @unmatched_opening_indexes_in_this_batch +
          @unmatched_closing_indexes_in_this_batch;
    }

    sub comma_arrow_count {
        my $seqno = $_[0];
        return $comma_arrow_count{$seqno};
    }

    sub match_opening_and_closing_tokens {

        # Match up indexes of opening and closing braces, etc, in this batch.
        # This has to be done after all tokens are stored because unstoring
        # of tokens would otherwise cause trouble.

        @unmatched_opening_indexes_in_this_batch = ();
        @unmatched_closing_indexes_in_this_batch = ();
        %comma_arrow_count                       = ();
        my $comma_arrow_count_contained = 0;

        my ( $i, $i_mate, $token );
        foreach $i ( 0 .. $max_index_to_go ) {
            if ( $type_sequence_to_go[$i] ) {
                $token = $tokens_to_go[$i];
                if ( $token =~ /^[\(\[\{\?]$/ ) {
                    push @unmatched_opening_indexes_in_this_batch, $i;
                }
                elsif ( $token =~ /^[\)\]\}\:]$/ ) {

                    $i_mate = pop @unmatched_opening_indexes_in_this_batch;
                    if ( defined($i_mate) && $i_mate >= 0 ) {
                        if ( $type_sequence_to_go[$i_mate] ==
                            $type_sequence_to_go[$i] )
                        {
                            $mate_index_to_go[$i]      = $i_mate;
                            $mate_index_to_go[$i_mate] = $i;
                            my $seqno = $type_sequence_to_go[$i];
                            if ( $comma_arrow_count{$seqno} ) {
                                $comma_arrow_count_contained +=
                                  $comma_arrow_count{$seqno};
                            }
                        }
                        else {
                            push @unmatched_opening_indexes_in_this_batch,
                              $i_mate;
                            push @unmatched_closing_indexes_in_this_batch, $i;
                        }
                    }
                    else {
                        push @unmatched_closing_indexes_in_this_batch, $i;
                    }
                }
            }
            elsif ( $tokens_to_go[$i] eq '=>' ) {
                if (@unmatched_opening_indexes_in_this_batch) {
                    my $j     = $unmatched_opening_indexes_in_this_batch[-1];
                    my $seqno = $type_sequence_to_go[$j];
                    $comma_arrow_count{$seqno}++;
                }
            }
        }
        return $comma_arrow_count_contained;
    }

    sub save_opening_indentation {

        # This should be called after each batch of tokens is output. It
        # saves indentations of lines of all unmatched opening tokens.
        # These will be used by sub get_opening_indentation.

        my ( $ri_first, $ri_last, $rindentation_list ) = @_;

        # we no longer need indentations of any saved indentations which
        # are unmatched closing tokens in this batch, because we will
        # never encounter them again.  So we can delete them to keep
        # the hash size down.
        foreach (@unmatched_closing_indexes_in_this_batch) {
            my $seqno = $type_sequence_to_go[$_];
            delete $saved_opening_indentation{$seqno};
        }

        # we need to save indentations of any unmatched opening tokens
        # in this batch because we may need them in a subsequent batch.
        foreach (@unmatched_opening_indexes_in_this_batch) {
            my $seqno = $type_sequence_to_go[$_];
            $saved_opening_indentation{$seqno} = [
                lookup_opening_indentation(
                    $_, $ri_first, $ri_last, $rindentation_list
                )
            ];
        }
    }
}    # end unmatched_indexes

sub get_opening_indentation {

    # get the indentation of the line which output the opening token
    # corresponding to a given closing token in the current output batch.
    #
    # given:
    # $i_closing - index in this line of a closing token ')' '}' or ']'
    #
    # $ri_first - reference to list of the first index $i for each output
    #               line in this batch
    # $ri_last - reference to list of the last index $i for each output line
    #              in this batch
    # $rindentation_list - reference to a list containing the indentation
    #            used for each line.
    #
    # return:
    #   -the indentation of the line which contained the opening token
    #    which matches the token at index $i_opening
    #   -and its offset (number of columns) from the start of the line
    #
    my ( $i_closing, $ri_first, $ri_last, $rindentation_list ) = @_;

    # first, see if the opening token is in the current batch
    my $i_opening = $mate_index_to_go[$i_closing];
    my ( $indent, $offset, $is_leading, $exists );
    $exists = 1;
    if ( $i_opening >= 0 ) {

        # it is..look up the indentation
        ( $indent, $offset, $is_leading ) =
          lookup_opening_indentation( $i_opening, $ri_first, $ri_last,
            $rindentation_list );
    }

    # if not, it should have been stored in the hash by a previous batch
    else {
        my $seqno = $type_sequence_to_go[$i_closing];
        if ($seqno) {
            if ( $saved_opening_indentation{$seqno} ) {
                ( $indent, $offset, $is_leading ) =
                  @{ $saved_opening_indentation{$seqno} };
            }

            # some kind of serious error
            # (example is badfile.t)
            else {
                $indent     = 0;
                $offset     = 0;
                $is_leading = 0;
                $exists     = 0;
            }
        }

        # if no sequence number it must be an unbalanced container
        else {
            $indent     = 0;
            $offset     = 0;
            $is_leading = 0;
            $exists     = 0;
        }
    }
    return ( $indent, $offset, $is_leading, $exists );
}

sub lookup_opening_indentation {

    # get the indentation of the line in the current output batch
    # which output a selected opening token
    #
    # given:
    #   $i_opening - index of an opening token in the current output batch
    #                whose line indentation we need
    #   $ri_first - reference to list of the first index $i for each output
    #               line in this batch
    #   $ri_last - reference to list of the last index $i for each output line
    #              in this batch
    #   $rindentation_list - reference to a list containing the indentation
    #            used for each line.  (NOTE: the first slot in
    #            this list is the last returned line number, and this is
    #            followed by the list of indentations).
    #
    # return
    #   -the indentation of the line which contained token $i_opening
    #   -and its offset (number of columns) from the start of the line

    my ( $i_opening, $ri_start, $ri_last, $rindentation_list ) = @_;

    my $nline = $rindentation_list->[0];    # line number of previous lookup

    # reset line location if necessary
    $nline = 0 if ( $i_opening < $ri_start->[$nline] );

    # find the correct line
    unless ( $i_opening > $ri_last->[-1] ) {
        while ( $i_opening > $ri_last->[$nline] ) { $nline++; }
    }

    # error - token index is out of bounds - shouldn't happen
    else {
        warning(
"non-fatal program bug in lookup_opening_indentation - index out of range\n"
        );
        report_definite_bug();
        $nline = $#{$ri_last};
    }

    $rindentation_list->[0] =
      $nline;    # save line number to start looking next call
    my $ibeg       = $ri_start->[$nline];
    my $offset     = token_sequence_length( $ibeg, $i_opening ) - 1;
    my $is_leading = ( $ibeg == $i_opening );
    return ( $rindentation_list->[ $nline + 1 ], $offset, $is_leading );
}

{
    my %is_if_elsif_else_unless_while_until_for_foreach;

    BEGIN {

        # These block types may have text between the keyword and opening
        # curly.  Note: 'else' does not, but must be included to allow trailing
        # if/elsif text to be appended.
        # patch for SWITCH/CASE: added 'case' and 'when'
        @_ = qw(if elsif else unless while until for foreach case when);
        @is_if_elsif_else_unless_while_until_for_foreach{@_} =
          (1) x scalar(@_);
    }

    sub set_adjusted_indentation {

        # This routine has the final say regarding the actual indentation of
        # a line.  It starts with the basic indentation which has been
        # defined for the leading token, and then takes into account any
        # options that the user has set regarding special indenting and
        # outdenting.

        my ( $ibeg, $iend, $rfields, $rpatterns, $ri_first, $ri_last,
            $rindentation_list, $level_jump )
          = @_;

        # we need to know the last token of this line
        my ( $terminal_type, $i_terminal ) =
          terminal_type( \@types_to_go, \@block_type_to_go, $ibeg, $iend );

        my $is_outdented_line = 0;

        my $is_semicolon_terminated = $terminal_type eq ';'
          && $nesting_depth_to_go[$iend] < $nesting_depth_to_go[$ibeg];

        ##########################################################
        # Section 1: set a flag and a default indentation
        #
        # Most lines are indented according to the initial token.
        # But it is common to outdent to the level just after the
        # terminal token in certain cases...
        # adjust_indentation flag:
        #       0 - do not adjust
        #       1 - outdent
        #       2 - vertically align with opening token
        #       3 - indent
        ##########################################################
        my $adjust_indentation         = 0;
        my $default_adjust_indentation = $adjust_indentation;

        my (
            $opening_indentation, $opening_offset,
            $is_leading,          $opening_exists
        );

        # if we are at a closing token of some type..
        if ( $types_to_go[$ibeg] =~ /^[\)\}\]R]$/ ) {

            # get the indentation of the line containing the corresponding
            # opening token
            (
                $opening_indentation, $opening_offset,
                $is_leading,          $opening_exists
              )
              = get_opening_indentation( $ibeg, $ri_first, $ri_last,
                $rindentation_list );

            # First set the default behavior:
            if (

                # default behavior is to outdent closing lines
                # of the form:   ");  };  ];  )->xxx;"
                $is_semicolon_terminated

                # and 'cuddled parens' of the form:   ")->pack("
                || (
                       $terminal_type eq '('
                    && $types_to_go[$ibeg] eq ')'
                    && ( $nesting_depth_to_go[$iend] + 1 ==
                        $nesting_depth_to_go[$ibeg] )
                )

                # and when the next line is at a lower indentation level
                # PATCH: and only if the style allows undoing continuation
                # for all closing token types. We should really wait until
                # the indentation of the next line is known and then make
                # a decision, but that would require another pass.
                || ( $level_jump < 0 && !$some_closing_token_indentation )
              )
            {
                $adjust_indentation = 1;
            }

            # outdent something like '),'
            if (
                $terminal_type eq ','

                # allow just one character before the comma
                && $i_terminal == $ibeg + 1

                # requre LIST environment; otherwise, we may outdent too much --
                # this can happen in calls without parentheses (overload.t);
                && $container_environment_to_go[$i_terminal] eq 'LIST'
              )
            {
                $adjust_indentation = 1;
            }

            # undo continuation indentation of a terminal closing token if
            # it is the last token before a level decrease.  This will allow
            # a closing token to line up with its opening counterpart, and
            # avoids a indentation jump larger than 1 level.
            if (   $types_to_go[$i_terminal] =~ /^[\}\]\)R]$/
                && $i_terminal == $ibeg )
            {
                my $ci        = $ci_levels_to_go[$ibeg];
                my $lev       = $levels_to_go[$ibeg];
                my $next_type = $types_to_go[ $ibeg + 1 ];
                my $i_next_nonblank =
                  ( ( $next_type eq 'b' ) ? $ibeg + 2 : $ibeg + 1 );
                if (   $i_next_nonblank <= $max_index_to_go
                    && $levels_to_go[$i_next_nonblank] < $lev )
                {
                    $adjust_indentation = 1;
                }
            }

            # YVES patch 1 of 2:
            # Undo ci of line with leading closing eval brace,
            # but not beyond the indention of the line with
            # the opening brace.
            if (   $block_type_to_go[$ibeg] eq 'eval'
                && !$rOpts->{'line-up-parentheses'}
                && !$rOpts->{'indent-closing-brace'} )
            {
                (
                    $opening_indentation, $opening_offset,
                    $is_leading,          $opening_exists
                  )
                  = get_opening_indentation( $ibeg, $ri_first, $ri_last,
                    $rindentation_list );
                my $indentation = $leading_spaces_to_go[$ibeg];
                if ( defined($opening_indentation)
                    && $indentation > $opening_indentation )
                {
                    $adjust_indentation = 1;
                }
            }

            $default_adjust_indentation = $adjust_indentation;

            # Now modify default behavior according to user request:
            # handle option to indent non-blocks of the form );  };  ];
            # But don't do special indentation to something like ')->pack('
            if ( !$block_type_to_go[$ibeg] ) {
                my $cti = $closing_token_indentation{ $tokens_to_go[$ibeg] };
                if ( $cti == 1 ) {
                    if (   $i_terminal <= $ibeg + 1
                        || $is_semicolon_terminated )
                    {
                        $adjust_indentation = 2;
                    }
                    else {
                        $adjust_indentation = 0;
                    }
                }
                elsif ( $cti == 2 ) {
                    if ($is_semicolon_terminated) {
                        $adjust_indentation = 3;
                    }
                    else {
                        $adjust_indentation = 0;
                    }
                }
                elsif ( $cti == 3 ) {
                    $adjust_indentation = 3;
                }
            }

            # handle option to indent blocks
            else {
                if (
                    $rOpts->{'indent-closing-brace'}
                    && (
                        $i_terminal == $ibeg    #  isolated terminal '}'
                        || $is_semicolon_terminated
                    )
                  )                             #  } xxxx ;
                {
                    $adjust_indentation = 3;
                }
            }
        }

        # if at ');', '};', '>;', and '];' of a terminal qw quote
        elsif ($$rpatterns[0] =~ /^qb*;$/
            && $$rfields[0] =~ /^([\)\}\]\>]);$/ )
        {
            if ( $closing_token_indentation{$1} == 0 ) {
                $adjust_indentation = 1;
            }
            else {
                $adjust_indentation = 3;
            }
        }

        # if line begins with a ':', align it with any
        # previous line leading with corresponding ?
        elsif ( $types_to_go[$ibeg] eq ':' ) {
            (
                $opening_indentation, $opening_offset,
                $is_leading,          $opening_exists
              )
              = get_opening_indentation( $ibeg, $ri_first, $ri_last,
                $rindentation_list );
            if ($is_leading) { $adjust_indentation = 2; }
        }

        ##########################################################
        # Section 2: set indentation according to flag set above
        #
        # Select the indentation object to define leading
        # whitespace.  If we are outdenting something like '} } );'
        # then we want to use one level below the last token
        # ($i_terminal) in order to get it to fully outdent through
        # all levels.
        ##########################################################
        my $indentation;
        my $lev;
        my $level_end = $levels_to_go[$iend];

        if ( $adjust_indentation == 0 ) {
            $indentation = $leading_spaces_to_go[$ibeg];
            $lev         = $levels_to_go[$ibeg];
        }
        elsif ( $adjust_indentation == 1 ) {
            $indentation = $reduced_spaces_to_go[$i_terminal];
            $lev         = $levels_to_go[$i_terminal];
        }

        # handle indented closing token which aligns with opening token
        elsif ( $adjust_indentation == 2 ) {

            # handle option to align closing token with opening token
            $lev = $levels_to_go[$ibeg];

            # calculate spaces needed to align with opening token
            my $space_count =
              get_SPACES($opening_indentation) + $opening_offset;

            # Indent less than the previous line.
            #
            # Problem: For -lp we don't exactly know what it was if there
            # were recoverable spaces sent to the aligner.  A good solution
            # would be to force a flush of the vertical alignment buffer, so
            # that we would know.  For now, this rule is used for -lp:
            #
            # When the last line did not start with a closing token we will
            # be optimistic that the aligner will recover everything wanted.
            #
            # This rule will prevent us from breaking a hierarchy of closing
            # tokens, and in a worst case will leave a closing paren too far
            # indented, but this is better than frequently leaving it not
            # indented enough.
            my $last_spaces = get_SPACES($last_indentation_written);
            if ( $last_leading_token !~ /^[\}\]\)]$/ ) {
                $last_spaces +=
                  get_RECOVERABLE_SPACES($last_indentation_written);
            }

            # reset the indentation to the new space count if it works
            # only options are all or none: nothing in-between looks good
            $lev = $levels_to_go[$ibeg];
            if ( $space_count < $last_spaces ) {
                if ($rOpts_line_up_parentheses) {
                    my $lev = $levels_to_go[$ibeg];
                    $indentation =
                      new_lp_indentation_item( $space_count, $lev, 0, 0, 0 );
                }
                else {
                    $indentation = $space_count;
                }
            }

            # revert to default if it doesn't work
            else {
                $space_count = leading_spaces_to_go($ibeg);
                if ( $default_adjust_indentation == 0 ) {
                    $indentation = $leading_spaces_to_go[$ibeg];
                }
                elsif ( $default_adjust_indentation == 1 ) {
                    $indentation = $reduced_spaces_to_go[$i_terminal];
                    $lev         = $levels_to_go[$i_terminal];
                }
            }
        }

        # Full indentaion of closing tokens (-icb and -icp or -cti=2)
        else {

            # handle -icb (indented closing code block braces)
            # Updated method for indented block braces: indent one full level if
            # there is no continuation indentation.  This will occur for major
            # structures such as sub, if, else, but not for things like map
            # blocks.
            #
            # Note: only code blocks without continuation indentation are
            # handled here (if, else, unless, ..). In the following snippet,
            # the terminal brace of the sort block will have continuation
            # indentation as shown so it will not be handled by the coding
            # here.  We would have to undo the continuation indentation to do
            # this, but it probably looks ok as is.  This is a possible future
            # update for semicolon terminated lines.
            #
            #     if ($sortby eq 'date' or $sortby eq 'size') {
            #         @files = sort {
            #             $file_data{$a}{$sortby} <=> $file_data{$b}{$sortby}
            #                 or $a cmp $b
            #                 } @files;
            #         }
            #
            if (   $block_type_to_go[$ibeg]
                && $ci_levels_to_go[$i_terminal] == 0 )
            {
                my $spaces = get_SPACES( $leading_spaces_to_go[$i_terminal] );
                $indentation = $spaces + $rOpts_indent_columns;

                # NOTE: for -lp we could create a new indentation object, but
                # there is probably no need to do it
            }

            # handle -icp and any -icb block braces which fall through above
            # test such as the 'sort' block mentioned above.
            else {

                # There are currently two ways to handle -icp...
                # One way is to use the indentation of the previous line:
                # $indentation = $last_indentation_written;

                # The other way is to use the indentation that the previous line
                # would have had if it hadn't been adjusted:
                $indentation = $last_unadjusted_indentation;

                # Current method: use the minimum of the two. This avoids
                # inconsistent indentation.
                if ( get_SPACES($last_indentation_written) <
                    get_SPACES($indentation) )
                {
                    $indentation = $last_indentation_written;
                }
            }

            # use previous indentation but use own level
            # to cause list to be flushed properly
            $lev = $levels_to_go[$ibeg];
        }

        # remember indentation except for multi-line quotes, which get
        # no indentation
        unless ( $ibeg == 0 && $starting_in_quote ) {
            $last_indentation_written    = $indentation;
            $last_unadjusted_indentation = $leading_spaces_to_go[$ibeg];
            $last_leading_token          = $tokens_to_go[$ibeg];
        }

        # be sure lines with leading closing tokens are not outdented more
        # than the line which contained the corresponding opening token.

        #############################################################
        # updated per bug report in alex_bug.pl: we must not
        # mess with the indentation of closing logical braces so
        # we must treat something like '} else {' as if it were
        # an isolated brace my $is_isolated_block_brace = (
        # $iend == $ibeg ) && $block_type_to_go[$ibeg];
        #############################################################
        my $is_isolated_block_brace = $block_type_to_go[$ibeg]
          && ( $iend == $ibeg
            || $is_if_elsif_else_unless_while_until_for_foreach{
                $block_type_to_go[$ibeg]
            } );

        # only do this for a ':; which is aligned with its leading '?'
        my $is_unaligned_colon = $types_to_go[$ibeg] eq ':' && !$is_leading;
        if (   defined($opening_indentation)
            && !$is_isolated_block_brace
            && !$is_unaligned_colon )
        {
            if ( get_SPACES($opening_indentation) > get_SPACES($indentation) ) {
                $indentation = $opening_indentation;
            }
        }

        # remember the indentation of each line of this batch
        push @{$rindentation_list}, $indentation;

        # outdent lines with certain leading tokens...
        if (

            # must be first word of this batch
            $ibeg == 0

            # and ...
            && (

                # certain leading keywords if requested
                (
                       $rOpts->{'outdent-keywords'}
                    && $types_to_go[$ibeg] eq 'k'
                    && $outdent_keyword{ $tokens_to_go[$ibeg] }
                )

                # or labels if requested
                || ( $rOpts->{'outdent-labels'} && $types_to_go[$ibeg] eq 'J' )

                # or static block comments if requested
                || (   $types_to_go[$ibeg] eq '#'
                    && $rOpts->{'outdent-static-block-comments'}
                    && $is_static_block_comment )
            )
          )

        {
            my $space_count = leading_spaces_to_go($ibeg);
            if ( $space_count > 0 ) {
                $space_count -= $rOpts_continuation_indentation;
                $is_outdented_line = 1;
                if ( $space_count < 0 ) { $space_count = 0 }

                # do not promote a spaced static block comment to non-spaced;
                # this is not normally necessary but could be for some
                # unusual user inputs (such as -ci = -i)
                if ( $types_to_go[$ibeg] eq '#' && $space_count == 0 ) {
                    $space_count = 1;
                }

                if ($rOpts_line_up_parentheses) {
                    $indentation =
                      new_lp_indentation_item( $space_count, $lev, 0, 0, 0 );
                }
                else {
                    $indentation = $space_count;
                }
            }
        }

        return ( $indentation, $lev, $level_end, $terminal_type,
            $is_semicolon_terminated, $is_outdented_line );
    }
}

sub set_vertical_tightness_flags {

    my ( $n, $n_last_line, $ibeg, $iend, $ri_first, $ri_last ) = @_;

    # Define vertical tightness controls for the nth line of a batch.
    # We create an array of parameters which tell the vertical aligner
    # if we should combine this line with the next line to achieve the
    # desired vertical tightness.  The array of parameters contains:
    #
    #   [0] type: 1=opening non-block    2=closing non-block
    #             3=opening block brace  4=closing block brace
    #
    #   [1] flag: if opening: 1=no multiple steps, 2=multiple steps ok
    #             if closing: spaces of padding to use
    #   [2] sequence number of container
    #   [3] valid flag: do not append if this flag is false. Will be
    #       true if appropriate -vt flag is set.  Otherwise, Will be
    #       made true only for 2 line container in parens with -lp
    #
    # These flags are used by sub set_leading_whitespace in
    # the vertical aligner

    my $rvertical_tightness_flags = [ 0, 0, 0, 0, 0, 0 ];

    #--------------------------------------------------------------
    # Vertical Tightness Flags Section 1:
    # Handle Lines 1 .. n-1 but not the last line
    # For non-BLOCK tokens, we will need to examine the next line
    # too, so we won't consider the last line.
    #--------------------------------------------------------------
    if ( $n < $n_last_line ) {

        #--------------------------------------------------------------
        # Vertical Tightness Flags Section 1a:
        # Look for Type 1, last token of this line is a non-block opening token
        #--------------------------------------------------------------
        my $ibeg_next = $$ri_first[ $n + 1 ];
        my $token_end = $tokens_to_go[$iend];
        my $iend_next = $$ri_last[ $n + 1 ];
        if (
               $type_sequence_to_go[$iend]
            && !$block_type_to_go[$iend]
            && $is_opening_token{$token_end}
            && (
                $opening_vertical_tightness{$token_end} > 0

                # allow 2-line method call to be closed up
                || (   $rOpts_line_up_parentheses
                    && $token_end eq '('
                    && $iend > $ibeg
                    && $types_to_go[ $iend - 1 ] ne 'b' )
            )
          )
        {

            # avoid multiple jumps in nesting depth in one line if
            # requested
            my $ovt       = $opening_vertical_tightness{$token_end};
            my $iend_next = $$ri_last[ $n + 1 ];
            unless (
                $ovt < 2
                && ( $nesting_depth_to_go[ $iend_next + 1 ] !=
                    $nesting_depth_to_go[$ibeg_next] )
              )
            {

                # If -vt flag has not been set, mark this as invalid
                # and aligner will validate it if it sees the closing paren
                # within 2 lines.
                my $valid_flag = $ovt;
                @{$rvertical_tightness_flags} =
                  ( 1, $ovt, $type_sequence_to_go[$iend], $valid_flag );
            }
        }

        #--------------------------------------------------------------
        # Vertical Tightness Flags Section 1b:
        # Look for Type 2, first token of next line is a non-block closing
        # token .. and be sure this line does not have a side comment
        #--------------------------------------------------------------
        my $token_next = $tokens_to_go[$ibeg_next];
        if (   $type_sequence_to_go[$ibeg_next]
            && !$block_type_to_go[$ibeg_next]
            && $is_closing_token{$token_next}
            && $types_to_go[$iend] !~ '#' )    # for safety, shouldn't happen!
        {
            my $ovt = $opening_vertical_tightness{$token_next};
            my $cvt = $closing_vertical_tightness{$token_next};
            if (

                # never append a trailing line like   )->pack(
                # because it will throw off later alignment
                (
                    $nesting_depth_to_go[$ibeg_next] ==
                    $nesting_depth_to_go[ $iend_next + 1 ] + 1
                )
                && (
                    $cvt == 2
                    || (
                        $container_environment_to_go[$ibeg_next] ne 'LIST'
                        && (
                            $cvt == 1

                            # allow closing up 2-line method calls
                            || (   $rOpts_line_up_parentheses
                                && $token_next eq ')' )
                        )
                    )
                )
              )
            {

                # decide which trailing closing tokens to append..
                my $ok = 0;
                if ( $cvt == 2 || $iend_next == $ibeg_next ) { $ok = 1 }
                else {
                    my $str = join( '',
                        @types_to_go[ $ibeg_next + 1 .. $ibeg_next + 2 ] );

                    # append closing token if followed by comment or ';'
                    if ( $str =~ /^b?[#;]/ ) { $ok = 1 }
                }

                if ($ok) {
                    my $valid_flag = $cvt;
                    @{$rvertical_tightness_flags} = (
                        2,
                        $tightness{$token_next} == 2 ? 0 : 1,
                        $type_sequence_to_go[$ibeg_next], $valid_flag,
                    );
                }
            }
        }

        #--------------------------------------------------------------
        # Vertical Tightness Flags Section 1c:
        # Implement the Opening Token Right flag (Type 2)..
        # If requested, move an isolated trailing opening token to the end of
        # the previous line which ended in a comma.  We could do this
        # in sub recombine_breakpoints but that would cause problems
        # with -lp formatting.  The problem is that indentation will
        # quickly move far to the right in nested expressions.  By
        # doing it after indentation has been set, we avoid changes
        # to the indentation.  Actual movement of the token takes place
        # in sub valign_output_step_B.
        #--------------------------------------------------------------
        if (
            $opening_token_right{ $tokens_to_go[$ibeg_next] }

            # previous line is not opening
            # (use -sot to combine with it)
            && !$is_opening_token{$token_end}

            # previous line ended in one of these
            # (add other cases if necessary; '=>' and '.' are not necessary
            && !$block_type_to_go[$ibeg_next]

            # this is a line with just an opening token
            && (   $iend_next == $ibeg_next
                || $iend_next == $ibeg_next + 2
                && $types_to_go[$iend_next] eq '#' )

            # looks bad if we align vertically with the wrong container
            && $tokens_to_go[$ibeg] ne $tokens_to_go[$ibeg_next]
          )
        {
            my $valid_flag = 1;
            my $spaces = ( $types_to_go[ $ibeg_next - 1 ] eq 'b' ) ? 1 : 0;
            @{$rvertical_tightness_flags} =
              ( 2, $spaces, $type_sequence_to_go[$ibeg_next], $valid_flag, );
        }

        #--------------------------------------------------------------
        # Vertical Tightness Flags Section 1d:
        # Stacking of opening and closing tokens (Type 2)
        #--------------------------------------------------------------
        my $stackable;
        my $token_beg_next = $tokens_to_go[$ibeg_next];

        # patch to make something like 'qw(' behave like an opening paren
        # (aran.t)
        if ( $types_to_go[$ibeg_next] eq 'q' ) {
            if ( $token_beg_next =~ /^qw\s*([\[\(\{])$/ ) {
                $token_beg_next = $1;
            }
        }

        if (   $is_closing_token{$token_end}
            && $is_closing_token{$token_beg_next} )
        {
            $stackable = $stack_closing_token{$token_beg_next}
              unless ( $block_type_to_go[$ibeg_next] )
              ;    # shouldn't happen; just checking
        }
        elsif ($is_opening_token{$token_end}
            && $is_opening_token{$token_beg_next} )
        {
            $stackable = $stack_opening_token{$token_beg_next}
              unless ( $block_type_to_go[$ibeg_next] )
              ;    # shouldn't happen; just checking
        }

        if ($stackable) {

            my $is_semicolon_terminated;
            if ( $n + 1 == $n_last_line ) {
                my ( $terminal_type, $i_terminal ) = terminal_type(
                    \@types_to_go, \@block_type_to_go,
                    $ibeg_next,    $iend_next
                );
                $is_semicolon_terminated = $terminal_type eq ';'
                  && $nesting_depth_to_go[$iend_next] <
                  $nesting_depth_to_go[$ibeg_next];
            }

            # this must be a line with just an opening token
            # or end in a semicolon
            if (
                $is_semicolon_terminated
                || (   $iend_next == $ibeg_next
                    || $iend_next == $ibeg_next + 2
                    && $types_to_go[$iend_next] eq '#' )
              )
            {
                my $valid_flag = 1;
                my $spaces = ( $types_to_go[ $ibeg_next - 1 ] eq 'b' ) ? 1 : 0;
                @{$rvertical_tightness_flags} =
                  ( 2, $spaces, $type_sequence_to_go[$ibeg_next], $valid_flag,
                  );
            }
        }
    }

    #--------------------------------------------------------------
    # Vertical Tightness Flags Section 2:
    # Handle type 3, opening block braces on last line of the batch
    # Check for a last line with isolated opening BLOCK curly
    #--------------------------------------------------------------
    elsif ($rOpts_block_brace_vertical_tightness
        && $ibeg eq $iend
        && $types_to_go[$iend] eq '{'
        && $block_type_to_go[$iend] =~
        /$block_brace_vertical_tightness_pattern/o )
    {
        @{$rvertical_tightness_flags} =
          ( 3, $rOpts_block_brace_vertical_tightness, 0, 1 );
    }

    #--------------------------------------------------------------
    # Vertical Tightness Flags Section 3:
    # Handle type 4, a closing block brace on the last line of the batch Check
    # for a last line with isolated closing BLOCK curly
    #--------------------------------------------------------------
    elsif ($rOpts_stack_closing_block_brace
        && $ibeg eq $iend
        && $block_type_to_go[$iend]
        && $types_to_go[$iend] eq '}' )
    {
        my $spaces = $rOpts_block_brace_tightness == 2 ? 0 : 1;
        @{$rvertical_tightness_flags} =
          ( 4, $spaces, $type_sequence_to_go[$iend], 1 );
    }

    # pack in the sequence numbers of the ends of this line
    $rvertical_tightness_flags->[4] = get_seqno($ibeg);
    $rvertical_tightness_flags->[5] = get_seqno($iend);
    return $rvertical_tightness_flags;
}

sub get_seqno {

    # get opening and closing sequence numbers of a token for the vertical
    # aligner.  Assign qw quotes a value to allow qw opening and closing tokens
    # to be treated somewhat like opening and closing tokens for stacking
    # tokens by the vertical aligner.
    my ($ii) = @_;
    my $seqno = $type_sequence_to_go[$ii];
    if ( $types_to_go[$ii] eq 'q' ) {
        my $SEQ_QW = -1;
        if ( $ii > 0 ) {
            $seqno = $SEQ_QW if ( $tokens_to_go[$ii] =~ /^qw\s*[\(\{\[]/ );
        }
        else {
            if ( !$ending_in_quote ) {
                $seqno = $SEQ_QW if ( $tokens_to_go[$ii] =~ /[\)\}\]]$/ );
            }
        }
    }
    return ($seqno);
}

{
    my %is_vertical_alignment_type;
    my %is_vertical_alignment_keyword;
    my %is_terminal_alignment_type;

    BEGIN {

        # Removed =~ from list to improve chances of alignment
        @_ = qw#
          = **= += *= &= <<= &&= -= /= |= >>= ||= //= .= %= ^= x=
          { ? : => && || // ~~ !~~
          #;
        @is_vertical_alignment_type{@_} = (1) x scalar(@_);

        # only align these at end of line
        @_ = qw(&& ||);
        @is_terminal_alignment_type{@_} = (1) x scalar(@_);

        # eq and ne were removed from this list to improve alignment chances
        @_ = qw(if unless and or err for foreach while until);
        @is_vertical_alignment_keyword{@_} = (1) x scalar(@_);
    }

    sub set_vertical_alignment_markers {

        # This routine takes the first step toward vertical alignment of the
        # lines of output text.  It looks for certain tokens which can serve as
        # vertical alignment markers (such as an '=').
        #
        # Method: We look at each token $i in this output batch and set
        # $matching_token_to_go[$i] equal to those tokens at which we would
        # accept vertical alignment.

        # nothing to do if we aren't allowed to change whitespace
        if ( !$rOpts_add_whitespace ) {
            for my $i ( 0 .. $max_index_to_go ) {
                $matching_token_to_go[$i] = '';
            }
            return;
        }

        my ( $ri_first, $ri_last ) = @_;

        # remember the index of last nonblank token before any sidecomment
        my $i_terminal = $max_index_to_go;
        if ( $types_to_go[$i_terminal] eq '#' ) {
            if ( $i_terminal > 0 && $types_to_go[ --$i_terminal ] eq 'b' ) {
                if ( $i_terminal > 0 ) { --$i_terminal }
            }
        }

        # look at each line of this batch..
        my $last_vertical_alignment_before_index;
        my $vert_last_nonblank_type;
        my $vert_last_nonblank_token;
        my $vert_last_nonblank_block_type;
        my $max_line = @$ri_first - 1;
        my ( $i, $type, $token, $block_type, $alignment_type );
        my ( $ibeg, $iend, $line );

        foreach $line ( 0 .. $max_line ) {
            $ibeg                                 = $$ri_first[$line];
            $iend                                 = $$ri_last[$line];
            $last_vertical_alignment_before_index = -1;
            $vert_last_nonblank_type              = '';
            $vert_last_nonblank_token             = '';
            $vert_last_nonblank_block_type        = '';

            # look at each token in this output line..
            foreach $i ( $ibeg .. $iend ) {
                $alignment_type = '';
                $type           = $types_to_go[$i];
                $block_type     = $block_type_to_go[$i];
                $token          = $tokens_to_go[$i];

                # check for flag indicating that we should not align
                # this token
                if ( $matching_token_to_go[$i] ) {
                    $matching_token_to_go[$i] = '';
                    next;
                }

                #--------------------------------------------------------
                # First see if we want to align BEFORE this token
                #--------------------------------------------------------

                # The first possible token that we can align before
                # is index 2 because: 1) it doesn't normally make sense to
                # align before the first token and 2) the second
                # token must be a blank if we are to align before
                # the third
                if ( $i < $ibeg + 2 ) { }

                # must follow a blank token
                elsif ( $types_to_go[ $i - 1 ] ne 'b' ) { }

                # align a side comment --
                elsif ( $type eq '#' ) {

                    unless (

                        # it is a static side comment
                        (
                               $rOpts->{'static-side-comments'}
                            && $token =~ /$static_side_comment_pattern/o
                        )

                        # or a closing side comment
                        || (   $vert_last_nonblank_block_type
                            && $token =~
                            /$closing_side_comment_prefix_pattern/o )
                      )
                    {
                        $alignment_type = $type;
                    }    ## Example of a static side comment
                }

                # otherwise, do not align two in a row to create a
                # blank field
                elsif ( $last_vertical_alignment_before_index == $i - 2 ) { }

                # align before one of these keywords
                # (within a line, since $i>1)
                elsif ( $type eq 'k' ) {

                    #  /^(if|unless|and|or|eq|ne)$/
                    if ( $is_vertical_alignment_keyword{$token} ) {
                        $alignment_type = $token;
                    }
                }

                # align before one of these types..
                # Note: add '.' after new vertical aligner is operational
                elsif ( $is_vertical_alignment_type{$type} ) {
                    $alignment_type = $token;

                    # Do not align a terminal token.  Although it might
                    # occasionally look ok to do this, this has been found to be
                    # a good general rule.  The main problems are:
                    # (1) that the terminal token (such as an = or :) might get
                    # moved far to the right where it is hard to see because
                    # nothing follows it, and
                    # (2) doing so may prevent other good alignments.
                    # Current exceptions are && and ||
                    if ( $i == $iend || $i >= $i_terminal ) {
                        $alignment_type = ""
                          unless ( $is_terminal_alignment_type{$type} );
                    }

                    # Do not align leading ': (' or '. ('.  This would prevent
                    # alignment in something like the following:
                    #   $extra_space .=
                    #       ( $input_line_number < 10 )  ? "  "
                    #     : ( $input_line_number < 100 ) ? " "
                    #     :                                "";
                    # or
                    #  $code =
                    #      ( $case_matters ? $accessor : " lc($accessor) " )
                    #    . ( $yesno        ? " eq "       : " ne " )
                    if (   $i == $ibeg + 2
                        && $types_to_go[$ibeg] =~ /^[\.\:]$/
                        && $types_to_go[ $i - 1 ] eq 'b' )
                    {
                        $alignment_type = "";
                    }

                    # For a paren after keyword, only align something like this:
                    #    if    ( $a ) { &a }
                    #    elsif ( $b ) { &b }
                    if ( $token eq '(' && $vert_last_nonblank_type eq 'k' ) {
                        $alignment_type = ""
                          unless $vert_last_nonblank_token =~
                          /^(if|unless|elsif)$/;
                    }

                    # be sure the alignment tokens are unique
                    # This didn't work well: reason not determined
                    # if ($token ne $type) {$alignment_type .= $type}
                }

                # NOTE: This is deactivated because it causes the previous
                # if/elsif alignment to fail
                #elsif ( $type eq '}' && $token eq '}' && $block_type_to_go[$i])
                #{ $alignment_type = $type; }

                if ($alignment_type) {
                    $last_vertical_alignment_before_index = $i;
                }

                #--------------------------------------------------------
                # Next see if we want to align AFTER the previous nonblank
                #--------------------------------------------------------

                # We want to line up ',' and interior ';' tokens, with the added
                # space AFTER these tokens.  (Note: interior ';' is included
                # because it may occur in short blocks).
                if (

                    # we haven't already set it
                    !$alignment_type

                    # and its not the first token of the line
                    && ( $i > $ibeg )

                    # and it follows a blank
                    && $types_to_go[ $i - 1 ] eq 'b'

                    # and previous token IS one of these:
                    && ( $vert_last_nonblank_type =~ /^[\,\;]$/ )

                    # and it's NOT one of these
                    && ( $type !~ /^[b\#\)\]\}]$/ )

                    # then go ahead and align
                  )

                {
                    $alignment_type = $vert_last_nonblank_type;
                }

                #--------------------------------------------------------
                # then store the value
                #--------------------------------------------------------
                $matching_token_to_go[$i] = $alignment_type;
                if ( $type ne 'b' ) {
                    $vert_last_nonblank_type       = $type;
                    $vert_last_nonblank_token      = $token;
                    $vert_last_nonblank_block_type = $block_type;
                }
            }
        }
    }
}

sub terminal_type {

    #    returns type of last token on this line (terminal token), as follows:
    #    returns # for a full-line comment
    #    returns ' ' for a blank line
    #    otherwise returns final token type

    my ( $rtype, $rblock_type, $ibeg, $iend ) = @_;

    # check for full-line comment..
    if ( $$rtype[$ibeg] eq '#' ) {
        return wantarray ? ( $$rtype[$ibeg], $ibeg ) : $$rtype[$ibeg];
    }
    else {

        # start at end and walk bakwards..
        for ( my $i = $iend ; $i >= $ibeg ; $i-- ) {

            # skip past any side comment and blanks
            next if ( $$rtype[$i] eq 'b' );
            next if ( $$rtype[$i] eq '#' );

            # found it..make sure it is a BLOCK termination,
            # but hide a terminal } after sort/grep/map because it is not
            # necessarily the end of the line.  (terminal.t)
            my $terminal_type = $$rtype[$i];
            if (
                $terminal_type eq '}'
                && ( !$$rblock_type[$i]
                    || ( $is_sort_map_grep_eval_do{ $$rblock_type[$i] } ) )
              )
            {
                $terminal_type = 'b';
            }
            return wantarray ? ( $terminal_type, $i ) : $terminal_type;
        }

        # empty line
        return wantarray ? ( ' ', $ibeg ) : ' ';
    }
}

{    # set_bond_strengths

    my %is_good_keyword_breakpoint;
    my %is_lt_gt_le_ge;

    my %binary_bond_strength;
    my %nobreak_lhs;
    my %nobreak_rhs;

    my @bias_tokens;
    my $delta_bias;

    sub bias_table_key {
        my ( $type, $token ) = @_;
        my $bias_table_key = $type;
        if ( $type eq 'k' ) {
            $bias_table_key = $token;
            if ( $token eq 'err' ) { $bias_table_key = 'or' }
        }
        return $bias_table_key;
    }

    sub set_bond_strengths {

        BEGIN {

            @_ = qw(if unless while until for foreach);
            @is_good_keyword_breakpoint{@_} = (1) x scalar(@_);

            @_ = qw(lt gt le ge);
            @is_lt_gt_le_ge{@_} = (1) x scalar(@_);
            #
            # The decision about where to break a line depends upon a "bond
            # strength" between tokens.  The LOWER the bond strength, the MORE
            # likely a break.  A bond strength may be any value but to simplify
            # things there are several pre-defined strength levels:

            #    NO_BREAK    => 10000;
            #    VERY_STRONG => 100;
            #    STRONG      => 2.1;
            #    NOMINAL     => 1.1;
            #    WEAK        => 0.8;
            #    VERY_WEAK   => 0.55;

            # The strength values are based on trial-and-error, and need to be
            # tweaked occasionally to get desired results.  Some comments:
            #
            #   1. Only relative strengths are important.  small differences
            #      in strengths can make big formatting differences.
            #   2. Each indentation level adds one unit of bond strength.
            #   3. A value of NO_BREAK makes an unbreakable bond
            #   4. A value of VERY_WEAK is the strength of a ','
            #   5. Values below NOMINAL are considered ok break points.
            #   6. Values above NOMINAL are considered poor break points.
            #
            # The bond strengths should roughly follow precenence order where
            # possible.  If you make changes, please check the results very
            # carefully on a variety of scripts.  Testing with the -extrude
            # options is particularly helpful in exercising all of the rules.

            # Wherever possible, bond strengths are defined in the following
            # tables.  There are two main stages to setting bond strengths and
            # two types of tables:
            #
            # The first stage involves looking at each token individually and
            # defining left and right bond strengths, according to if we want
            # to break to the left or right side, and how good a break point it
            # is.  For example tokens like =, ||, && make good break points and
            # will have low strengths, but one might want to break on either
            # side to put them at the end of one line or beginning of the next.
            #
            # The second stage involves looking at certain pairs of tokens and
            # defining a bond strength for that particular pair.  This second
            # stage has priority.

            #---------------------------------------------------------------
            # Bond Strength BEGIN Section 1.
            # Set left and right bond strengths of individual tokens.
            #---------------------------------------------------------------

            # NOTE: NO_BREAK's set in this section first are HINTS which will
            # probably not be honored. Essential NO_BREAKS's should be set in
            # BEGIN Section 2 or hardwired in the NO_BREAK coding near the end
            # of this subroutine.

            # Note that we are setting defaults in this section.  The user
            # cannot change bond strengths but can cause the left and right
            # bond strengths of any token type to be swapped through the use of
            # the -wba and -wbb flags. In this way the user can determine if a
            # breakpoint token should appear at the end of one line or the
            # beginning of the next line.

            # The hash keys in this section are token types, plus the text of
            # certain keywords like 'or', 'and'.

            # no break around possible filehandle
            $left_bond_strength{'Z'}  = NO_BREAK;
            $right_bond_strength{'Z'} = NO_BREAK;

            # never put a bare word on a new line:
            # example print (STDERR, "bla"); will fail with break after (
            $left_bond_strength{'w'} = NO_BREAK;

            # blanks always have infinite strength to force breaks after
            # real tokens
            $right_bond_strength{'b'} = NO_BREAK;

            # try not to break on exponentation
            @_                       = qw" ** .. ... <=> ";
            @left_bond_strength{@_}  = (STRONG) x scalar(@_);
            @right_bond_strength{@_} = (STRONG) x scalar(@_);

            # The comma-arrow has very low precedence but not a good break point
            $left_bond_strength{'=>'}  = NO_BREAK;
            $right_bond_strength{'=>'} = NOMINAL;

            # ok to break after label
            $left_bond_strength{'J'}  = NO_BREAK;
            $right_bond_strength{'J'} = NOMINAL;
            $left_bond_strength{'j'}  = STRONG;
            $right_bond_strength{'j'} = STRONG;
            $left_bond_strength{'A'}  = STRONG;
            $right_bond_strength{'A'} = STRONG;

            $left_bond_strength{'->'}  = STRONG;
            $right_bond_strength{'->'} = VERY_STRONG;

            $left_bond_strength{'CORE::'}  = NOMINAL;
            $right_bond_strength{'CORE::'} = NO_BREAK;

            # breaking AFTER modulus operator is ok:
            @_ = qw" % ";
            @left_bond_strength{@_} = (STRONG) x scalar(@_);
            @right_bond_strength{@_} =
              ( 0.1 * NOMINAL + 0.9 * STRONG ) x scalar(@_);

            # Break AFTER math operators * and /
            @_                       = qw" * / x  ";
            @left_bond_strength{@_}  = (STRONG) x scalar(@_);
            @right_bond_strength{@_} = (NOMINAL) x scalar(@_);

            # Break AFTER weakest math operators + and -
            # Make them weaker than * but a bit stronger than '.'
            @_ = qw" + - ";
            @left_bond_strength{@_} = (STRONG) x scalar(@_);
            @right_bond_strength{@_} =
              ( 0.91 * NOMINAL + 0.09 * WEAK ) x scalar(@_);

            # breaking BEFORE these is just ok:
            @_                       = qw" >> << ";
            @right_bond_strength{@_} = (STRONG) x scalar(@_);
            @left_bond_strength{@_}  = (NOMINAL) x scalar(@_);

            # breaking before the string concatenation operator seems best
            # because it can be hard to see at the end of a line
            $right_bond_strength{'.'} = STRONG;
            $left_bond_strength{'.'}  = 0.9 * NOMINAL + 0.1 * WEAK;

            @_                       = qw"} ] ) R";
            @left_bond_strength{@_}  = (STRONG) x scalar(@_);
            @right_bond_strength{@_} = (NOMINAL) x scalar(@_);

            # make these a little weaker than nominal so that they get
            # favored for end-of-line characters
            @_ = qw"!= == =~ !~ ~~ !~~";
            @left_bond_strength{@_} = (STRONG) x scalar(@_);
            @right_bond_strength{@_} =
              ( 0.9 * NOMINAL + 0.1 * WEAK ) x scalar(@_);

            # break AFTER these
            @_ = qw" < >  | & >= <=";
            @left_bond_strength{@_} = (VERY_STRONG) x scalar(@_);
            @right_bond_strength{@_} =
              ( 0.8 * NOMINAL + 0.2 * WEAK ) x scalar(@_);

            # breaking either before or after a quote is ok
            # but bias for breaking before a quote
            $left_bond_strength{'Q'}  = NOMINAL;
            $right_bond_strength{'Q'} = NOMINAL + 0.02;
            $left_bond_strength{'q'}  = NOMINAL;
            $right_bond_strength{'q'} = NOMINAL;

            # starting a line with a keyword is usually ok
            $left_bond_strength{'k'} = NOMINAL;

            # we usually want to bond a keyword strongly to what immediately
            # follows, rather than leaving it stranded at the end of a line
            $right_bond_strength{'k'} = STRONG;

            $left_bond_strength{'G'}  = NOMINAL;
            $right_bond_strength{'G'} = STRONG;

            # assignment operators
            @_ = qw(
              = **= += *= &= <<= &&=
              -= /= |= >>= ||= //=
              .= %= ^=
              x=
            );

            # Default is is to break AFTER various assignment operators
            @left_bond_strength{@_} = (STRONG) x scalar(@_);
            @right_bond_strength{@_} =
              ( 0.4 * WEAK + 0.6 * VERY_WEAK ) x scalar(@_);

            # Default is to break BEFORE '&&' and '||' and '//'
            # set strength of '||' to same as '=' so that chains like
            # $a = $b || $c || $d   will break before the first '||'
            $right_bond_strength{'||'} = NOMINAL;
            $left_bond_strength{'||'}  = $right_bond_strength{'='};

            # same thing for '//'
            $right_bond_strength{'//'} = NOMINAL;
            $left_bond_strength{'//'}  = $right_bond_strength{'='};

            # set strength of && a little higher than ||
            $right_bond_strength{'&&'} = NOMINAL;
            $left_bond_strength{'&&'}  = $left_bond_strength{'||'} + 0.1;

            $left_bond_strength{';'}  = VERY_STRONG;
            $right_bond_strength{';'} = VERY_WEAK;
            $left_bond_strength{'f'}  = VERY_STRONG;

            # make right strength of for ';' a little less than '='
            # to make for contents break after the ';' to avoid this:
            #   for ( $j = $number_of_fields - 1 ; $j < $item_count ; $j +=
            #     $number_of_fields )
            # and make it weaker than ',' and 'and' too
            $right_bond_strength{'f'} = VERY_WEAK - 0.03;

            # The strengths of ?/: should be somewhere between
            # an '=' and a quote (NOMINAL),
            # make strength of ':' slightly less than '?' to help
            # break long chains of ? : after the colons
            $left_bond_strength{':'}  = 0.4 * WEAK + 0.6 * NOMINAL;
            $right_bond_strength{':'} = NO_BREAK;
            $left_bond_strength{'?'}  = $left_bond_strength{':'} + 0.01;
            $right_bond_strength{'?'} = NO_BREAK;

            $left_bond_strength{','}  = VERY_STRONG;
            $right_bond_strength{','} = VERY_WEAK;

            # remaining digraphs and trigraphs not defined above
            @_                       = qw( :: <> ++ --);
            @left_bond_strength{@_}  = (WEAK) x scalar(@_);
            @right_bond_strength{@_} = (STRONG) x scalar(@_);

            # Set bond strengths of certain keywords
            # make 'or', 'err', 'and' slightly weaker than a ','
            $left_bond_strength{'and'}  = VERY_WEAK - 0.01;
            $left_bond_strength{'or'}   = VERY_WEAK - 0.02;
            $left_bond_strength{'err'}  = VERY_WEAK - 0.02;
            $left_bond_strength{'xor'}  = NOMINAL;
            $right_bond_strength{'and'} = NOMINAL;
            $right_bond_strength{'or'}  = NOMINAL;
            $right_bond_strength{'err'} = NOMINAL;
            $right_bond_strength{'xor'} = STRONG;

            #---------------------------------------------------------------
            # Bond Strength BEGIN Section 2.
            # Set binary rules for bond strengths between certain token types.
            #---------------------------------------------------------------

            #  We have a little problem making tables which apply to the
            #  container tokens.  Here is a list of container tokens and
            #  their types:
            #
            #   type    tokens // meaning
            #      {    {, [, ( // indent
            #      }    }, ], ) // outdent
            #      [    [ // left non-structural [ (enclosing an array index)
            #      ]    ] // right non-structural square bracket
            #      (    ( // left non-structural paren
            #      )    ) // right non-structural paren
            #      L    { // left non-structural curly brace (enclosing a key)
            #      R    } // right non-structural curly brace
            #
            #  Some rules apply to token types and some to just the token
            #  itself.  We solve the problem by combining type and token into a
            #  new hash key for the container types.
            #
            #  If a rule applies to a token 'type' then we need to make rules
            #  for each of these 'type.token' combinations:
            #  Type    Type.Token
            #  {       {{, {[, {(
            #  [       [[
            #  (       ((
            #  L       L{
            #  }       }}, }], })
            #  ]       ]]
            #  )       ))
            #  R       R}
            #
            #  If a rule applies to a token then we need to make rules for
            #  these 'type.token' combinations:
            #  Token   Type.Token
            #  {       {{, L{
            #  [       {[, [[
            #  (       {(, ((
            #  }       }}, R}
            #  ]       }], ]]
            #  )       }), ))

            # allow long lines before final { in an if statement, as in:
            #    if (..........
            #      ..........)
            #    {
            #
            # Otherwise, the line before the { tends to be too short.

            $binary_bond_strength{'))'}{'{{'} = VERY_WEAK + 0.03;
            $binary_bond_strength{'(('}{'{{'} = NOMINAL;

            # break on something like '} (', but keep this stronger than a ','
            # example is in 'howe.pl'
            $binary_bond_strength{'R}'}{'(('} = 0.8 * VERY_WEAK + 0.2 * WEAK;
            $binary_bond_strength{'}}'}{'(('} = 0.8 * VERY_WEAK + 0.2 * WEAK;

            # keep matrix and hash indices together
            # but make them a little below STRONG to allow breaking open
            # something like {'some-word'}{'some-very-long-word'} at the }{
            # (bracebrk.t)
            $binary_bond_strength{']]'}{'[['} = 0.9 * STRONG + 0.1 * NOMINAL;
            $binary_bond_strength{']]'}{'L{'} = 0.9 * STRONG + 0.1 * NOMINAL;
            $binary_bond_strength{'R}'}{'[['} = 0.9 * STRONG + 0.1 * NOMINAL;
            $binary_bond_strength{'R}'}{'L{'} = 0.9 * STRONG + 0.1 * NOMINAL;

            # increase strength to the point where a break in the following
            # will be after the opening paren rather than at the arrow:
            #    $a->$b($c);
            $binary_bond_strength{'i'}{'->'} = 1.45 * STRONG;

            $binary_bond_strength{'))'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;
            $binary_bond_strength{']]'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;
            $binary_bond_strength{'})'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;
            $binary_bond_strength{'}]'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;
            $binary_bond_strength{'}}'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;
            $binary_bond_strength{'R}'}{'->'} = 0.1 * STRONG + 0.9 * NOMINAL;

            $binary_bond_strength{'))'}{'[['} = 0.2 * STRONG + 0.8 * NOMINAL;
            $binary_bond_strength{'})'}{'[['} = 0.2 * STRONG + 0.8 * NOMINAL;
            $binary_bond_strength{'))'}{'{['} = 0.2 * STRONG + 0.8 * NOMINAL;
            $binary_bond_strength{'})'}{'{['} = 0.2 * STRONG + 0.8 * NOMINAL;

            #---------------------------------------------------------------
            # Binary NO_BREAK rules
            #---------------------------------------------------------------

            # use strict requires that bare word and => not be separated
            $binary_bond_strength{'C'}{'=>'} = NO_BREAK;
            $binary_bond_strength{'U'}{'=>'} = NO_BREAK;

            # Never break between a bareword and a following paren because
            # perl may give an error.  For example, if a break is placed
            # between 'to_filehandle' and its '(' the following line will
            # give a syntax error [Carp.pm]: my( $no) =fileno(
            # to_filehandle( $in)) ;
            $binary_bond_strength{'C'}{'(('} = NO_BREAK;
            $binary_bond_strength{'C'}{'{('} = NO_BREAK;
            $binary_bond_strength{'U'}{'(('} = NO_BREAK;
            $binary_bond_strength{'U'}{'{('} = NO_BREAK;

            # use strict requires that bare word within braces not start new
            # line
            $binary_bond_strength{'L{'}{'w'} = NO_BREAK;

            $binary_bond_strength{'w'}{'R}'} = NO_BREAK;

            # use strict requires that bare word and => not be separated
            $binary_bond_strength{'w'}{'=>'} = NO_BREAK;

            # use strict does not allow separating type info from trailing { }
            # testfile is readmail.pl
            $binary_bond_strength{'t'}{'L{'} = NO_BREAK;
            $binary_bond_strength{'i'}{'L{'} = NO_BREAK;

            # As a defensive measure, do not break between a '(' and a
            # filehandle.  In some cases, this can cause an error.  For
            # example, the following program works:
            #    my $msg="hi!\n";
            #    print
            #    ( STDOUT
            #    $msg
            #    );
            #
            # But this program fails:
            #    my $msg="hi!\n";
            #    print
            #    (
            #    STDOUT
            #    $msg
            #    );
            #
            # This is normally only a problem with the 'extrude' option
            $binary_bond_strength{'(('}{'Y'} = NO_BREAK;
            $binary_bond_strength{'{('}{'Y'} = NO_BREAK;

            # never break between sub name and opening paren
            $binary_bond_strength{'w'}{'(('} = NO_BREAK;
            $binary_bond_strength{'w'}{'{('} = NO_BREAK;

            # keep '}' together with ';'
            $binary_bond_strength{'}}'}{';'} = NO_BREAK;

            # Breaking before a ++ can cause perl to guess wrong. For
            # example the following line will cause a syntax error
            # with -extrude if we break between '$i' and '++' [fixstyle2]
            #   print( ( $i++ & 1 ) ? $_ : ( $change{$_} || $_ ) );
            $nobreak_lhs{'++'} = NO_BREAK;

            # Do not break before a possible file handle
            $nobreak_lhs{'Z'} = NO_BREAK;

            # use strict hates bare words on any new line.  For
            # example, a break before the underscore here provokes the
            # wrath of use strict:
            # if ( -r $fn && ( -s _ || $AllowZeroFilesize)) {
            $nobreak_rhs{'F'}      = NO_BREAK;
            $nobreak_rhs{'CORE::'} = NO_BREAK;

            #---------------------------------------------------------------
            # Bond Strength BEGIN Section 3.
            # Define tables and values for applying a small bias to the above
            # values.
            #---------------------------------------------------------------
            # Adding a small 'bias' to strengths is a simple way to make a line
            # break at the first of a sequence of identical terms.  For
            # example, to force long string of conditional operators to break
            # with each line ending in a ':', we can add a small number to the
            # bond strength of each ':' (colon.t)
            @bias_tokens = qw( : && || f and or . );    # tokens which get bias
            $delta_bias = 0.0001;    # a very small strength level

        } ## end BEGIN

        # patch-its always ok to break at end of line
        $nobreak_to_go[$max_index_to_go] = 0;

        # we start a new set of bias values for each line
        my %bias;
        @bias{@bias_tokens} = (0) x scalar(@bias_tokens);
        my $code_bias = -.01;        # bias for closing block braces

        my $type  = 'b';
        my $token = ' ';
        my $last_type;
        my $last_nonblank_type  = $type;
        my $last_nonblank_token = $token;
        my $list_str            = $left_bond_strength{'?'};

        my ( $block_type, $i_next, $i_next_nonblank, $next_nonblank_token,
            $next_nonblank_type, $next_token, $next_type, $total_nesting_depth,
        );

        # main loop to compute bond strengths between each pair of tokens
        for ( my $i = 0 ; $i <= $max_index_to_go ; $i++ ) {
            $last_type = $type;
            if ( $type ne 'b' ) {
                $last_nonblank_type  = $type;
                $last_nonblank_token = $token;
            }
            $type = $types_to_go[$i];

            # strength on both sides of a blank is the same
            if ( $type eq 'b' && $last_type ne 'b' ) {
                $bond_strength_to_go[$i] = $bond_strength_to_go[ $i - 1 ];
                next;
            }

            $token               = $tokens_to_go[$i];
            $block_type          = $block_type_to_go[$i];
            $i_next              = $i + 1;
            $next_type           = $types_to_go[$i_next];
            $next_token          = $tokens_to_go[$i_next];
            $total_nesting_depth = $nesting_depth_to_go[$i_next];
            $i_next_nonblank     = ( ( $next_type eq 'b' ) ? $i + 2 : $i + 1 );
            $next_nonblank_type  = $types_to_go[$i_next_nonblank];
            $next_nonblank_token = $tokens_to_go[$i_next_nonblank];

            # We are computing the strength of the bond between the current
            # token and the NEXT token.

            #---------------------------------------------------------------
            # Bond Strength Section 1:
            # First Approximation.
            # Use minimum of individual left and right tabulated bond
            # strengths.
            #---------------------------------------------------------------
            my $bsr = $right_bond_strength{$type};
            my $bsl = $left_bond_strength{$next_nonblank_type};

            # define right bond strengths of certain keywords
            if ( $type eq 'k' && defined( $right_bond_strength{$token} ) ) {
                $bsr = $right_bond_strength{$token};
            }
            elsif ( $token eq 'ne' or $token eq 'eq' ) {
                $bsr = NOMINAL;
            }

            # set terminal bond strength to the nominal value
            # this will cause good preceding breaks to be retained
            if ( $i_next_nonblank > $max_index_to_go ) {
                $bsl = NOMINAL;
            }

            # define right bond strengths of certain keywords
            if ( $next_nonblank_type eq 'k'
                && defined( $left_bond_strength{$next_nonblank_token} ) )
            {
                $bsl = $left_bond_strength{$next_nonblank_token};
            }
            elsif ($next_nonblank_token eq 'ne'
                or $next_nonblank_token eq 'eq' )
            {
                $bsl = NOMINAL;
            }
            elsif ( $is_lt_gt_le_ge{$next_nonblank_token} ) {
                $bsl = 0.9 * NOMINAL + 0.1 * STRONG;
            }

            # Use the minimum of the left and right strengths.  Note: it might
            # seem that we would want to keep a NO_BREAK if either token has
            # this value.  This didn't work, for example because in an arrow
            # list, it prevents the comma from separating from the following
            # bare word (which is probably quoted by its arrow).  So necessary
            # NO_BREAK's have to be handled as special cases in the final
            # section.
            if ( !defined($bsr) ) { $bsr = VERY_STRONG }
            if ( !defined($bsl) ) { $bsl = VERY_STRONG }
            my $bond_str = ( $bsr < $bsl ) ? $bsr : $bsl;
            my $bond_str_1 = $bond_str;

            #---------------------------------------------------------------
            # Bond Strength Section 2:
            # Apply hardwired rules..
            #---------------------------------------------------------------

            # Patch to put terminal or clauses on a new line: Weaken the bond
            # at an || followed by die or similar keyword to make the terminal
            # or clause fall on a new line, like this:
            #
            #   my $class = shift
            #     || die "Cannot add broadcast:  No class identifier found";
            #
            # Otherwise the break will be at the previous '=' since the || and
            # = have the same starting strength and the or is biased, like
            # this:
            #
            # my $class =
            #   shift || die "Cannot add broadcast:  No class identifier found";
            #
            # In any case if the user places a break at either the = or the ||
            # it should remain there.
            if ( $type eq '||' || $type eq 'k' && $token eq 'or' ) {
                if ( $next_nonblank_token =~ /^(die|confess|croak|warn)$/ ) {
                    if ( $want_break_before{$token} && $i > 0 ) {
                        $bond_strength_to_go[ $i - 1 ] -= $delta_bias;
                    }
                    else {
                        $bond_str -= $delta_bias;
                    }
                }
            }

            # good to break after end of code blocks
            if ( $type eq '}' && $block_type && $next_nonblank_type ne ';' ) {

                $bond_str = 0.5 * WEAK + 0.5 * VERY_WEAK + $code_bias;
                $code_bias += $delta_bias;
            }

            if ( $type eq 'k' ) {

                # allow certain control keywords to stand out
                if (   $next_nonblank_type eq 'k'
                    && $is_last_next_redo_return{$token} )
                {
                    $bond_str = 0.45 * WEAK + 0.55 * VERY_WEAK;
                }

                # Don't break after keyword my.  This is a quick fix for a
                # rare problem with perl. An example is this line from file
                # Container.pm:

                # foreach my $question( Debian::DebConf::ConfigDb::gettree(
                # $this->{'question'} ) )

                if ( $token eq 'my' ) {
                    $bond_str = NO_BREAK;
                }

            }

            # good to break before 'if', 'unless', etc
            if ( $is_if_brace_follower{$next_nonblank_token} ) {
                $bond_str = VERY_WEAK;
            }

            if ( $next_nonblank_type eq 'k' && $type ne 'CORE::' ) {

                # FIXME: needs more testing
                if ( $is_keyword_returning_list{$next_nonblank_token} ) {
                    $bond_str = $list_str if ( $bond_str > $list_str );
                }

                # keywords like 'unless', 'if', etc, within statements
                # make good breaks
                if ( $is_good_keyword_breakpoint{$next_nonblank_token} ) {
                    $bond_str = VERY_WEAK / 1.05;
                }
            }

            # try not to break before a comma-arrow
            elsif ( $next_nonblank_type eq '=>' ) {
                if ( $bond_str < STRONG ) { $bond_str = STRONG }
            }

            #---------------------------------------------------------------
            # Additional hardwired NOBREAK rules
            #---------------------------------------------------------------

            # map1.t -- correct for a quirk in perl
            if (   $token eq '('
                && $next_nonblank_type eq 'i'
                && $last_nonblank_type eq 'k'
                && $is_sort_map_grep{$last_nonblank_token} )

              #     /^(sort|map|grep)$/ )
            {
                $bond_str = NO_BREAK;
            }

            # extrude.t: do not break before paren at:
            #    -l pid_filename(
            if ( $last_nonblank_type eq 'F' && $next_nonblank_token eq '(' ) {
                $bond_str = NO_BREAK;
            }

            # in older version of perl, use strict can cause problems with
            # breaks before bare words following opening parens.  For example,
            # this will fail under older versions if a break is made between
            # '(' and 'MAIL': use strict; open( MAIL, "a long filename or
            # command"); close MAIL;
            if ( $type eq '{' ) {

                if ( $token eq '(' && $next_nonblank_type eq 'w' ) {

                    # but it's fine to break if the word is followed by a '=>'
                    # or if it is obviously a sub call
                    my $i_next_next_nonblank = $i_next_nonblank + 1;
                    my $next_next_type = $types_to_go[$i_next_next_nonblank];
                    if (   $next_next_type eq 'b'
                        && $i_next_nonblank < $max_index_to_go )
                    {
                        $i_next_next_nonblank++;
                        $next_next_type = $types_to_go[$i_next_next_nonblank];
                    }

                    # We'll check for an old breakpoint and keep a leading
                    # bareword if it was that way in the input file.
                    # Presumably it was ok that way.  For example, the
                    # following would remain unchanged:
                    #
                    # @months = (
                    #   January,   February, March,    April,
                    #   May,       June,     July,     August,
                    #   September, October,  November, December,
                    # );
                    #
                    # This should be sufficient:
                    if (
                        !$old_breakpoint_to_go[$i]
                        && (   $next_next_type eq ','
                            || $next_next_type eq '}' )
                      )
                    {
                        $bond_str = NO_BREAK;
                    }
                }
            }

            # Do not break between a possible filehandle and a ? or / and do
            # not introduce a break after it if there is no blank
            # (extrude.t)
            elsif ( $type eq 'Z' ) {

                # dont break..
                if (

                    # if there is no blank and we do not want one. Examples:
                    #    print $x++    # do not break after $x
                    #    print HTML"HELLO"   # break ok after HTML
                    (
                           $next_type ne 'b'
                        && defined( $want_left_space{$next_type} )
                        && $want_left_space{$next_type} == WS_NO
                    )

                    # or we might be followed by the start of a quote
                    || $next_nonblank_type =~ /^[\/\?]$/
                  )
                {
                    $bond_str = NO_BREAK;
                }
            }

            # Breaking before a ? before a quote can cause trouble if
            # they are not separated by a blank.
            # Example: a syntax error occurs if you break before the ? here
            #  my$logic=join$all?' && ':' || ',@regexps;
            # From: Professional_Perl_Programming_Code/multifind.pl
            if ( $next_nonblank_type eq '?' ) {
                $bond_str = NO_BREAK
                  if ( $types_to_go[ $i_next_nonblank + 1 ] eq 'Q' );
            }

            # Breaking before a . followed by a number
            # can cause trouble if there is no intervening space
            # Example: a syntax error occurs if you break before the .2 here
            #  $str .= pack($endian.2, ensurrogate($ord));
            # From: perl58/Unicode.pm
            elsif ( $next_nonblank_type eq '.' ) {
                $bond_str = NO_BREAK
                  if ( $types_to_go[ $i_next_nonblank + 1 ] eq 'n' );
            }

            # patch to put cuddled elses back together when on multiple
            # lines, as in: } \n else \n { \n
            if ($rOpts_cuddled_else) {

                if (   ( $token eq 'else' ) && ( $next_nonblank_type eq '{' )
                    || ( $type eq '}' ) && ( $next_nonblank_token eq 'else' ) )
                {
                    $bond_str = NO_BREAK;
                }
            }
            my $bond_str_2 = $bond_str;

            #---------------------------------------------------------------
            # End of hardwired rules
            #---------------------------------------------------------------

            #---------------------------------------------------------------
            # Bond Strength Section 3:
            # Apply table rules. These have priority over the above
            # hardwired rules.
            #---------------------------------------------------------------

            my $tabulated_bond_str;
            my $ltype = $type;
            my $rtype = $next_nonblank_type;
            if ( $token =~ /^[\(\[\{\)\]\}]/ ) { $ltype = $type . $token }
            if ( $next_nonblank_token =~ /^[\(\[\{\)\]\}]/ ) {
                $rtype = $next_nonblank_type . $next_nonblank_token;
            }

            if ( $binary_bond_strength{$ltype}{$rtype} ) {
                $bond_str           = $binary_bond_strength{$ltype}{$rtype};
                $tabulated_bond_str = $bond_str;
            }

            if ( $nobreak_rhs{$ltype} || $nobreak_lhs{$rtype} ) {
                $bond_str           = NO_BREAK;
                $tabulated_bond_str = $bond_str;
            }
            my $bond_str_3 = $bond_str;

            # If the hardwired rules conflict with the tabulated bond
            # strength then there is an inconsistency that should be fixed
            FORMATTER_DEBUG_FLAG_BOND_TABLES
              && $tabulated_bond_str
              && $bond_str_1
              && $bond_str_1 != $bond_str_2
              && $bond_str_2 != $tabulated_bond_str
              && do {
                print STDERR
"BOND_TABLES: ltype=$ltype rtype=$rtype $bond_str_1->$bond_str_2->$bond_str_3\n";
              };

           #-----------------------------------------------------------------
           # Bond Strength Section 4:
           # Modify strengths of certain tokens which often occur in sequence
           # by adding a small bias to each one in turn so that the breaks
           # occur from left to right.
           #
           # Note that we only changing strengths by small amounts here,
           # and usually increasing, so we should not be altering any NO_BREAKs.
           # Other routines which check for NO_BREAKs will use a tolerance
           # of one to avoid any problem.
           #-----------------------------------------------------------------

            # The bias tables use special keys
            my $left_key = bias_table_key( $type, $token );
            my $right_key =
              bias_table_key( $next_nonblank_type, $next_nonblank_token );

            # add any bias set by sub scan_list at old comma break points.
            if ( $type eq ',' ) { $bond_str += $bond_strength_to_go[$i] }

            # bias left token
            elsif ( defined( $bias{$left_key} ) ) {
                if ( !$want_break_before{$left_key} ) {
                    $bias{$left_key} += $delta_bias;
                    $bond_str += $bias{$left_key};
                }
            }

            # bias right token
            if ( defined( $bias{$right_key} ) ) {
                if ( $want_break_before{$right_key} ) {

                    # for leading '.' align all but 'short' quotes; the idea
                    # is to not place something like "\n" on a single line.
                    if ( $right_key eq '.' ) {
                        unless (
                            $last_nonblank_type eq '.'
                            && (
                                length($token) <=
                                $rOpts_short_concatenation_item_length )
                            && ( $token !~ /^[\)\]\}]$/ )
                          )
                        {
                            $bias{$right_key} += $delta_bias;
                        }
                    }
                    else {
                        $bias{$right_key} += $delta_bias;
                    }
                    $bond_str += $bias{$right_key};
                }
            }
            my $bond_str_4 = $bond_str;

            #---------------------------------------------------------------
            # Bond Strength Section 5:
            # Fifth Approximation.
            # Take nesting depth into account by adding the nesting depth
            # to the bond strength.
            #---------------------------------------------------------------
            my $strength;

            if ( defined($bond_str) && !$nobreak_to_go[$i] ) {
                if ( $total_nesting_depth > 0 ) {
                    $strength = $bond_str + $total_nesting_depth;
                }
                else {
                    $strength = $bond_str;
                }
            }
            else {
                $strength = NO_BREAK;
            }

            # always break after side comment
            if ( $type eq '#' ) { $strength = 0 }

            $bond_strength_to_go[$i] = $strength;

            FORMATTER_DEBUG_FLAG_BOND && do {
                my $str = substr( $token, 0, 15 );
                $str .= ' ' x ( 16 - length($str) );
                print STDOUT
"BOND:  i=$i $str $type $next_nonblank_type depth=$total_nesting_depth strength=$bond_str_1 -> $bond_str_2 -> $bond_str_3 -> $bond_str_4 $bond_str -> $strength \n";
            };
        } ## end main loop
    } ## end sub set_bond_strengths
}

sub pad_array_to_go {

    # to simplify coding in scan_list and set_bond_strengths, it helps
    # to create some extra blank tokens at the end of the arrays
    $tokens_to_go[ $max_index_to_go + 1 ] = '';
    $tokens_to_go[ $max_index_to_go + 2 ] = '';
    $types_to_go[ $max_index_to_go + 1 ]  = 'b';
    $types_to_go[ $max_index_to_go + 2 ]  = 'b';
    $nesting_depth_to_go[ $max_index_to_go + 1 ] =
      $nesting_depth_to_go[$max_index_to_go];

    #    /^[R\}\)\]]$/
    if ( $is_closing_type{ $types_to_go[$max_index_to_go] } ) {
        if ( $nesting_depth_to_go[$max_index_to_go] <= 0 ) {

            # shouldn't happen:
            unless ( get_saw_brace_error() ) {
                warning(
"Program bug in scan_list: hit nesting error which should have been caught\n"
                );
                report_definite_bug();
            }
        }
        else {
            $nesting_depth_to_go[ $max_index_to_go + 1 ] -= 1;
        }
    }

    #       /^[L\{\(\[]$/
    elsif ( $is_opening_type{ $types_to_go[$max_index_to_go] } ) {
        $nesting_depth_to_go[ $max_index_to_go + 1 ] += 1;
    }
}

{    # begin scan_list

    my (
        $block_type,               $current_depth,
        $depth,                    $i,
        $i_last_nonblank_token,    $last_colon_sequence_number,
        $last_nonblank_token,      $last_nonblank_type,
        $last_nonblank_block_type, $last_old_breakpoint_count,
        $minimum_depth,            $next_nonblank_block_type,
        $next_nonblank_token,      $next_nonblank_type,
        $old_breakpoint_count,     $starting_breakpoint_count,
        $starting_depth,           $token,
        $type,                     $type_sequence,
    );

    my (
        @breakpoint_stack,              @breakpoint_undo_stack,
        @comma_index,                   @container_type,
        @identifier_count_stack,        @index_before_arrow,
        @interrupted_list,              @item_count_stack,
        @last_comma_index,              @last_dot_index,
        @last_nonblank_type,            @old_breakpoint_count_stack,
        @opening_structure_index_stack, @rfor_semicolon_list,
        @has_old_logical_breakpoints,   @rand_or_list,
        @i_equals,
    );

    # routine to define essential variables when we go 'up' to
    # a new depth
    sub check_for_new_minimum_depth {
        my $depth = shift;
        if ( $depth < $minimum_depth ) {

            $minimum_depth = $depth;

            # these arrays need not retain values between calls
            $breakpoint_stack[$depth]              = $starting_breakpoint_count;
            $container_type[$depth]                = "";
            $identifier_count_stack[$depth]        = 0;
            $index_before_arrow[$depth]            = -1;
            $interrupted_list[$depth]              = 1;
            $item_count_stack[$depth]              = 0;
            $last_nonblank_type[$depth]            = "";
            $opening_structure_index_stack[$depth] = -1;

            $breakpoint_undo_stack[$depth]       = undef;
            $comma_index[$depth]                 = undef;
            $last_comma_index[$depth]            = undef;
            $last_dot_index[$depth]              = undef;
            $old_breakpoint_count_stack[$depth]  = undef;
            $has_old_logical_breakpoints[$depth] = 0;
            $rand_or_list[$depth]                = [];
            $rfor_semicolon_list[$depth]         = [];
            $i_equals[$depth]                    = -1;

            # these arrays must retain values between calls
            if ( !defined( $has_broken_sublist[$depth] ) ) {
                $dont_align[$depth]         = 0;
                $has_broken_sublist[$depth] = 0;
                $want_comma_break[$depth]   = 0;
            }
        }
    }

    # routine to decide which commas to break at within a container;
    # returns:
    #   $bp_count = number of comma breakpoints set
    #   $do_not_break_apart = a flag indicating if container need not
    #     be broken open
    sub set_comma_breakpoints {

        my $dd                 = shift;
        my $bp_count           = 0;
        my $do_not_break_apart = 0;

        # anything to do?
        if ( $item_count_stack[$dd] ) {

            # handle commas not in containers...
            if ( $dont_align[$dd] ) {
                do_uncontained_comma_breaks($dd);
            }

            # handle commas within containers...
            else {
                my $fbc = $forced_breakpoint_count;

                # always open comma lists not preceded by keywords,
                # barewords, identifiers (that is, anything that doesn't
                # look like a function call)
                my $must_break_open = $last_nonblank_type[$dd] !~ /^[kwiU]$/;

                set_comma_breakpoints_do(
                    $dd,
                    $opening_structure_index_stack[$dd],
                    $i,
                    $item_count_stack[$dd],
                    $identifier_count_stack[$dd],
                    $comma_index[$dd],
                    $next_nonblank_type,
                    $container_type[$dd],
                    $interrupted_list[$dd],
                    \$do_not_break_apart,
                    $must_break_open,
                );
                $bp_count = $forced_breakpoint_count - $fbc;
                $do_not_break_apart = 0 if $must_break_open;
            }
        }
        return ( $bp_count, $do_not_break_apart );
    }

    sub do_uncontained_comma_breaks {

        # Handle commas not in containers...
        # This is a catch-all routine for commas that we
        # don't know what to do with because the don't fall
        # within containers.  We will bias the bond strength
        # to break at commas which ended lines in the input
        # file.  This usually works better than just trying
        # to put as many items on a line as possible.  A
        # downside is that if the input file is garbage it
        # won't work very well. However, the user can always
        # prevent following the old breakpoints with the
        # -iob flag.
        my $dd                    = shift;
        my $bias                  = -.01;
        my $old_comma_break_count = 0;
        foreach my $ii ( @{ $comma_index[$dd] } ) {
            if ( $old_breakpoint_to_go[$ii] ) {
                $old_comma_break_count++;
                $bond_strength_to_go[$ii] = $bias;

                # reduce bias magnitude to force breaks in order
                $bias *= 0.99;
            }
        }

        # Also put a break before the first comma if
        # (1) there was a break there in the input, and
        # (2) there was exactly one old break before the first comma break
        # (3) OLD: there are multiple old comma breaks
        # (3) NEW: there are one or more old comma breaks (see return example)
        #
        # For example, we will follow the user and break after
        # 'print' in this snippet:
        #    print
        #      "conformability (Not the same dimension)\n",
        #      "\t", $have, " is ", text_unit($hu), "\n",
        #      "\t", $want, " is ", text_unit($wu), "\n",
        #      ;
        #
        # Another example, just one comma, where we will break after
        # the return:
        #  return
        #    $x * cos($a) - $y * sin($a),
        #    $x * sin($a) + $y * cos($a);

        # Breaking a print statement:
        # print SAVEOUT
        #   ( $? & 127 ) ? " (SIG#" . ( $? & 127 ) . ")" : "",
        #   ( $? & 128 ) ? " -- core dumped" : "", "\n";
        #
        #  But we will not force a break after the opening paren here
        #  (causes a blinker):
        #        $heap->{stream}->set_output_filter(
        #            poe::filter::reference->new('myotherfreezer') ),
        #          ;
        #
        my $i_first_comma = $comma_index[$dd]->[0];
        if ( $old_breakpoint_to_go[$i_first_comma] ) {
            my $level_comma = $levels_to_go[$i_first_comma];
            my $ibreak      = -1;
            my $obp_count   = 0;
            for ( my $ii = $i_first_comma - 1 ; $ii >= 0 ; $ii -= 1 ) {
                if ( $old_breakpoint_to_go[$ii] ) {
                    $obp_count++;
                    last if ( $obp_count > 1 );
                    $ibreak = $ii
                      if ( $levels_to_go[$ii] == $level_comma );
                }
            }

            # Changed rule from multiple old commas to just one here:
            if ( $ibreak >= 0 && $obp_count == 1 && $old_comma_break_count > 0 )
            {
                # Do not to break before an opening token because
                # it can lead to "blinkers".
                my $ibreakm = $ibreak;
                $ibreakm-- if ( $types_to_go[$ibreakm] eq 'b' );
                if ( $ibreakm >= 0 && $types_to_go[$ibreakm] !~ /^[\(\{\[L]$/ )
                {
                    set_forced_breakpoint($ibreak);
                }
            }
        }
    }

    my %is_logical_container;

    BEGIN {
        @_ = qw# if elsif unless while and or err not && | || ? : ! #;
        @is_logical_container{@_} = (1) x scalar(@_);
    }

    sub set_for_semicolon_breakpoints {
        my $dd = shift;
        foreach ( @{ $rfor_semicolon_list[$dd] } ) {
            set_forced_breakpoint($_);
        }
    }

    sub set_logical_breakpoints {
        my $dd = shift;
        if (
               $item_count_stack[$dd] == 0
            && $is_logical_container{ $container_type[$dd] }

            || $has_old_logical_breakpoints[$dd]
          )
        {

            # Look for breaks in this order:
            # 0   1    2   3
            # or  and  ||  &&
            foreach my $i ( 0 .. 3 ) {
                if ( $rand_or_list[$dd][$i] ) {
                    foreach ( @{ $rand_or_list[$dd][$i] } ) {
                        set_forced_breakpoint($_);
                    }

                    # break at any 'if' and 'unless' too
                    foreach ( @{ $rand_or_list[$dd][4] } ) {
                        set_forced_breakpoint($_);
                    }
                    $rand_or_list[$dd] = [];
                    last;
                }
            }
        }
    }

    sub is_unbreakable_container {

        # never break a container of one of these types
        # because bad things can happen (map1.t)
        my $dd = shift;
        $is_sort_map_grep{ $container_type[$dd] };
    }

    sub scan_list {

        # This routine is responsible for setting line breaks for all lists,
        # so that hierarchical structure can be displayed and so that list
        # items can be vertically aligned.  The output of this routine is
        # stored in the array @forced_breakpoint_to_go, which is used to set
        # final breakpoints.

        $starting_depth = $nesting_depth_to_go[0];

        $block_type                 = ' ';
        $current_depth              = $starting_depth;
        $i                          = -1;
        $last_colon_sequence_number = -1;
        $last_nonblank_token        = ';';
        $last_nonblank_type         = ';';
        $last_nonblank_block_type   = ' ';
        $last_old_breakpoint_count  = 0;
        $minimum_depth = $current_depth + 1;    # forces update in check below
        $old_breakpoint_count      = 0;
        $starting_breakpoint_count = $forced_breakpoint_count;
        $token                     = ';';
        $type                      = ';';
        $type_sequence             = '';

        my $total_depth_variation = 0;
        my $i_old_assignment_break;
        my $depth_last = $starting_depth;

        check_for_new_minimum_depth($current_depth);

        my $is_long_line = excess_line_length( 0, $max_index_to_go ) > 0;
        my $want_previous_breakpoint = -1;

        my $saw_good_breakpoint;
        my $i_line_end   = -1;
        my $i_line_start = -1;

        # loop over all tokens in this batch
        while ( ++$i <= $max_index_to_go ) {
            if ( $type ne 'b' ) {
                $i_last_nonblank_token    = $i - 1;
                $last_nonblank_type       = $type;
                $last_nonblank_token      = $token;
                $last_nonblank_block_type = $block_type;
            } ## end if ( $type ne 'b' )
            $type          = $types_to_go[$i];
            $block_type    = $block_type_to_go[$i];
            $token         = $tokens_to_go[$i];
            $type_sequence = $type_sequence_to_go[$i];
            my $next_type       = $types_to_go[ $i + 1 ];
            my $next_token      = $tokens_to_go[ $i + 1 ];
            my $i_next_nonblank = ( ( $next_type eq 'b' ) ? $i + 2 : $i + 1 );
            $next_nonblank_type       = $types_to_go[$i_next_nonblank];
            $next_nonblank_token      = $tokens_to_go[$i_next_nonblank];
            $next_nonblank_block_type = $block_type_to_go[$i_next_nonblank];

            # set break if flag was set
            if ( $want_previous_breakpoint >= 0 ) {
                set_forced_breakpoint($want_previous_breakpoint);
                $want_previous_breakpoint = -1;
            }

            $last_old_breakpoint_count = $old_breakpoint_count;
            if ( $old_breakpoint_to_go[$i] ) {
                $i_line_end   = $i;
                $i_line_start = $i_next_nonblank;

                $old_breakpoint_count++;

                # Break before certain keywords if user broke there and
                # this is a 'safe' break point. The idea is to retain
                # any preferred breaks for sequential list operations,
                # like a schwartzian transform.
                if ($rOpts_break_at_old_keyword_breakpoints) {
                    if (
                           $next_nonblank_type eq 'k'
                        && $is_keyword_returning_list{$next_nonblank_token}
                        && (   $type =~ /^[=\)\]\}Riw]$/
                            || $type eq 'k'
                            && $is_keyword_returning_list{$token} )
                      )
                    {

                        # we actually have to set this break next time through
                        # the loop because if we are at a closing token (such
                        # as '}') which forms a one-line block, this break might
                        # get undone.
                        $want_previous_breakpoint = $i;
                    } ## end if ( $next_nonblank_type...)
                } ## end if ($rOpts_break_at_old_keyword_breakpoints)

                # Break before attributes if user broke there
                if ($rOpts_break_at_old_attribute_breakpoints) {
                    if ( $next_nonblank_type eq 'A' ) {
                        $want_previous_breakpoint = $i;
                    }
                }

                # remember an = break as possible good break point
                if ( $is_assignment{$type} ) {
                    $i_old_assignment_break = $i;
                }
                elsif ( $is_assignment{$next_nonblank_type} ) {
                    $i_old_assignment_break = $i_next_nonblank;
                }
            } ## end if ( $old_breakpoint_to_go...)
            next if ( $type eq 'b' );
            $depth = $nesting_depth_to_go[ $i + 1 ];

            $total_depth_variation += abs( $depth - $depth_last );
            $depth_last = $depth;

            # safety check - be sure we always break after a comment
            # Shouldn't happen .. an error here probably means that the
            # nobreak flag did not get turned off correctly during
            # formatting.
            if ( $type eq '#' ) {
                if ( $i != $max_index_to_go ) {
                    warning(
"Non-fatal program bug: backup logic needed to break after a comment\n"
                    );
                    report_definite_bug();
                    $nobreak_to_go[$i] = 0;
                    set_forced_breakpoint($i);
                } ## end if ( $i != $max_index_to_go)
            } ## end if ( $type eq '#' )

            # Force breakpoints at certain tokens in long lines.
            # Note that such breakpoints will be undone later if these tokens
            # are fully contained within parens on a line.
            if (

                # break before a keyword within a line
                $type eq 'k'
                && $i > 0

                # if one of these keywords:
                && $token =~ /^(if|unless|while|until|for)$/

                # but do not break at something like '1 while'
                && ( $last_nonblank_type ne 'n' || $i > 2 )

                # and let keywords follow a closing 'do' brace
                && $last_nonblank_block_type ne 'do'

                && (
                    $is_long_line

                    # or container is broken (by side-comment, etc)
                    || (   $next_nonblank_token eq '('
                        && $mate_index_to_go[$i_next_nonblank] < $i )
                )
              )
            {
                set_forced_breakpoint( $i - 1 );
            } ## end if ( $type eq 'k' && $i...)

            # remember locations of '||'  and '&&' for possible breaks if we
            # decide this is a long logical expression.
            if ( $type eq '||' ) {
                push @{ $rand_or_list[$depth][2] }, $i;
                ++$has_old_logical_breakpoints[$depth]
                  if ( ( $i == $i_line_start || $i == $i_line_end )
                    && $rOpts_break_at_old_logical_breakpoints );
            } ## end if ( $type eq '||' )
            elsif ( $type eq '&&' ) {
                push @{ $rand_or_list[$depth][3] }, $i;
                ++$has_old_logical_breakpoints[$depth]
                  if ( ( $i == $i_line_start || $i == $i_line_end )
                    && $rOpts_break_at_old_logical_breakpoints );
            } ## end elsif ( $type eq '&&' )
            elsif ( $type eq 'f' ) {
                push @{ $rfor_semicolon_list[$depth] }, $i;
            }
            elsif ( $type eq 'k' ) {
                if ( $token eq 'and' ) {
                    push @{ $rand_or_list[$depth][1] }, $i;
                    ++$has_old_logical_breakpoints[$depth]
                      if ( ( $i == $i_line_start || $i == $i_line_end )
                        && $rOpts_break_at_old_logical_breakpoints );
                } ## end if ( $token eq 'and' )

                # break immediately at 'or's which are probably not in a logical
                # block -- but we will break in logical breaks below so that
                # they do not add to the forced_breakpoint_count
                elsif ( $token eq 'or' ) {
                    push @{ $rand_or_list[$depth][0] }, $i;
                    ++$has_old_logical_breakpoints[$depth]
                      if ( ( $i == $i_line_start || $i == $i_line_end )
                        && $rOpts_break_at_old_logical_breakpoints );
                    if ( $is_logical_container{ $container_type[$depth] } ) {
                    }
                    else {
                        if ($is_long_line) { set_forced_breakpoint($i) }
                        elsif ( ( $i == $i_line_start || $i == $i_line_end )
                            && $rOpts_break_at_old_logical_breakpoints )
                        {
                            $saw_good_breakpoint = 1;
                        }
                    } ## end else [ if ( $is_logical_container...)]
                } ## end elsif ( $token eq 'or' )
                elsif ( $token eq 'if' || $token eq 'unless' ) {
                    push @{ $rand_or_list[$depth][4] }, $i;
                    if ( ( $i == $i_line_start || $i == $i_line_end )
                        && $rOpts_break_at_old_logical_breakpoints )
                    {
                        set_forced_breakpoint($i);
                    }
                } ## end elsif ( $token eq 'if' ||...)
            } ## end elsif ( $type eq 'k' )
            elsif ( $is_assignment{$type} ) {
                $i_equals[$depth] = $i;
            }

            if ($type_sequence) {

                # handle any postponed closing breakpoints
                if ( $token =~ /^[\)\]\}\:]$/ ) {
                    if ( $type eq ':' ) {
                        $last_colon_sequence_number = $type_sequence;

                        # retain break at a ':' line break
                        if ( ( $i == $i_line_start || $i == $i_line_end )
                            && $rOpts_break_at_old_ternary_breakpoints )
                        {

                            set_forced_breakpoint($i);

                            # break at previous '='
                            if ( $i_equals[$depth] > 0 ) {
                                set_forced_breakpoint( $i_equals[$depth] );
                                $i_equals[$depth] = -1;
                            }
                        } ## end if ( ( $i == $i_line_start...))
                    } ## end if ( $type eq ':' )
                    if ( defined( $postponed_breakpoint{$type_sequence} ) ) {
                        my $inc = ( $type eq ':' ) ? 0 : 1;
                        set_forced_breakpoint( $i - $inc );
                        delete $postponed_breakpoint{$type_sequence};
                    }
                } ## end if ( $token =~ /^[\)\]\}\:]$/[{[(])

                # set breaks at ?/: if they will get separated (and are
                # not a ?/: chain), or if the '?' is at the end of the
                # line
                elsif ( $token eq '?' ) {
                    my $i_colon = $mate_index_to_go[$i];
                    if (
                        $i_colon <= 0  # the ':' is not in this batch
                        || $i == 0     # this '?' is the first token of the line
                        || $i ==
                        $max_index_to_go    # or this '?' is the last token
                      )
                    {

                        # don't break at a '?' if preceded by ':' on
                        # this line of previous ?/: pair on this line.
                        # This is an attempt to preserve a chain of ?/:
                        # expressions (elsif2.t).  And don't break if
                        # this has a side comment.
                        set_forced_breakpoint($i)
                          unless (
                            $type_sequence == (
                                $last_colon_sequence_number +
                                  TYPE_SEQUENCE_INCREMENT
                            )
                            || $tokens_to_go[$max_index_to_go] eq '#'
                          );
                        set_closing_breakpoint($i);
                    } ## end if ( $i_colon <= 0  ||...)
                } ## end elsif ( $token eq '?' )
            } ## end if ($type_sequence)

#print "LISTX sees: i=$i type=$type  tok=$token  block=$block_type depth=$depth\n";

            #------------------------------------------------------------
            # Handle Increasing Depth..
            #
            # prepare for a new list when depth increases
            # token $i is a '(','{', or '['
            #------------------------------------------------------------
            if ( $depth > $current_depth ) {

                $breakpoint_stack[$depth]       = $forced_breakpoint_count;
                $breakpoint_undo_stack[$depth]  = $forced_breakpoint_undo_count;
                $has_broken_sublist[$depth]     = 0;
                $identifier_count_stack[$depth] = 0;
                $index_before_arrow[$depth]     = -1;
                $interrupted_list[$depth]       = 0;
                $item_count_stack[$depth]       = 0;
                $last_comma_index[$depth]       = undef;
                $last_dot_index[$depth]         = undef;
                $last_nonblank_type[$depth]     = $last_nonblank_type;
                $old_breakpoint_count_stack[$depth]    = $old_breakpoint_count;
                $opening_structure_index_stack[$depth] = $i;
                $rand_or_list[$depth]                  = [];
                $rfor_semicolon_list[$depth]           = [];
                $i_equals[$depth]                      = -1;
                $want_comma_break[$depth]              = 0;
                $container_type[$depth] =
                  ( $last_nonblank_type =~ /^(k|=>|&&|\|\||\?|\:|\.)$/ )
                  ? $last_nonblank_token
                  : "";
                $has_old_logical_breakpoints[$depth] = 0;

                # if line ends here then signal closing token to break
                if ( $next_nonblank_type eq 'b' || $next_nonblank_type eq '#' )
                {
                    set_closing_breakpoint($i);
                }

                # Not all lists of values should be vertically aligned..
                $dont_align[$depth] =

                  # code BLOCKS are handled at a higher level
                  ( $block_type ne "" )

                  # certain paren lists
                  || ( $type eq '(' ) && (

                    # it does not usually look good to align a list of
                    # identifiers in a parameter list, as in:
                    #    my($var1, $var2, ...)
                    # (This test should probably be refined, for now I'm just
                    # testing for any keyword)
                    ( $last_nonblank_type eq 'k' )

                    # a trailing '(' usually indicates a non-list
                    || ( $next_nonblank_type eq '(' )
                  );

                # patch to outdent opening brace of long if/for/..
                # statements (like this one).  See similar coding in
                # set_continuation breaks.  We have also catch it here for
                # short line fragments which otherwise will not go through
                # set_continuation_breaks.
                if (
                    $block_type

                    # if we have the ')' but not its '(' in this batch..
                    && ( $last_nonblank_token eq ')' )
                    && $mate_index_to_go[$i_last_nonblank_token] < 0

                    # and user wants brace to left
                    && !$rOpts->{'opening-brace-always-on-right'}

                    && ( $type eq '{' )     # should be true
                    && ( $token eq '{' )    # should be true
                  )
                {
                    set_forced_breakpoint( $i - 1 );
                } ## end if ( $block_type && ( ...))
            } ## end if ( $depth > $current_depth)

            #------------------------------------------------------------
            # Handle Decreasing Depth..
            #
            # finish off any old list when depth decreases
            # token $i is a ')','}', or ']'
            #------------------------------------------------------------
            elsif ( $depth < $current_depth ) {

                check_for_new_minimum_depth($depth);

                # force all outer logical containers to break after we see on
                # old breakpoint
                $has_old_logical_breakpoints[$depth] ||=
                  $has_old_logical_breakpoints[$current_depth];

                # Patch to break between ') {' if the paren list is broken.
                # There is similar logic in set_continuation_breaks for
                # non-broken lists.
                if (   $token eq ')'
                    && $next_nonblank_block_type
                    && $interrupted_list[$current_depth]
                    && $next_nonblank_type eq '{'
                    && !$rOpts->{'opening-brace-always-on-right'} )
                {
                    set_forced_breakpoint($i);
                } ## end if ( $token eq ')' && ...

#print "LISTY sees: i=$i type=$type  tok=$token  block=$block_type depth=$depth next=$next_nonblank_type next_block=$next_nonblank_block_type inter=$interrupted_list[$current_depth]\n";

                # set breaks at commas if necessary
                my ( $bp_count, $do_not_break_apart ) =
                  set_comma_breakpoints($current_depth);

                my $i_opening = $opening_structure_index_stack[$current_depth];
                my $saw_opening_structure = ( $i_opening >= 0 );

                # this term is long if we had to break at interior commas..
                my $is_long_term = $bp_count > 0;

                # If this is a short container with one or more comma arrows,
                # then we will mark it as a long term to open it if requested.
                # $rOpts_comma_arrow_breakpoints =
                #    0 - open only if comma precedes closing brace
                #    1 - stable: except for one line blocks
                #    2 - try to form 1 line blocks
                #    3 - ignore =>
                #    4 - always open up if vt=0
                #    5 - stable: even for one line blocks if vt=0
                if (  !$is_long_term
                    && $tokens_to_go[$i_opening] =~ /^[\(\{\]L]$/
                    && $index_before_arrow[ $depth + 1 ] > 0
                    && !$opening_vertical_tightness{ $tokens_to_go[$i_opening] }
                  )
                {
                    $is_long_term = $rOpts_comma_arrow_breakpoints == 4
                      || ( $rOpts_comma_arrow_breakpoints == 0
                        && $last_nonblank_token eq ',' )
                      || ( $rOpts_comma_arrow_breakpoints == 5
                        && $old_breakpoint_to_go[$i_opening] );
                } ## end if ( !$is_long_term &&...)

                # mark term as long if the length between opening and closing
                # parens exceeds allowed line length
                if ( !$is_long_term && $saw_opening_structure ) {
                    my $i_opening_minus = find_token_starting_list($i_opening);

                    # Note: we have to allow for one extra space after a
                    # closing token so that we do not strand a comma or
                    # semicolon, hence the '>=' here (oneline.t)
                    $is_long_term =
                      excess_line_length( $i_opening_minus, $i ) >= 0;
                } ## end if ( !$is_long_term &&...)

                # We've set breaks after all comma-arrows.  Now we have to
                # undo them if this can be a one-line block
                # (the only breakpoints set will be due to comma-arrows)
                if (

                    # user doesn't require breaking after all comma-arrows
                    ( $rOpts_comma_arrow_breakpoints != 0 )
                    && ( $rOpts_comma_arrow_breakpoints != 4 )

                    # and if the opening structure is in this batch
                    && $saw_opening_structure

                    # and either on the same old line
                    && (
                        $old_breakpoint_count_stack[$current_depth] ==
                        $last_old_breakpoint_count

                        # or user wants to form long blocks with arrows
                        || $rOpts_comma_arrow_breakpoints == 2
                    )

                  # and we made some breakpoints between the opening and closing
                    && ( $breakpoint_undo_stack[$current_depth] <
                        $forced_breakpoint_undo_count )

                    # and this block is short enough to fit on one line
                    # Note: use < because need 1 more space for possible comma
                    && !$is_long_term

                  )
                {
                    undo_forced_breakpoint_stack(
                        $breakpoint_undo_stack[$current_depth] );
                } ## end if ( ( $rOpts_comma_arrow_breakpoints...))

                # now see if we have any comma breakpoints left
                my $has_comma_breakpoints =
                  ( $breakpoint_stack[$current_depth] !=
                      $forced_breakpoint_count );

                # update broken-sublist flag of the outer container
                $has_broken_sublist[$depth] =
                     $has_broken_sublist[$depth]
                  || $has_broken_sublist[$current_depth]
                  || $is_long_term
                  || $has_comma_breakpoints;

# Having come to the closing ')', '}', or ']', now we have to decide if we
# should 'open up' the structure by placing breaks at the opening and
# closing containers.  This is a tricky decision.  Here are some of the
# basic considerations:
#
# -If this is a BLOCK container, then any breakpoints will have already
# been set (and according to user preferences), so we need do nothing here.
#
# -If we have a comma-separated list for which we can align the list items,
# then we need to do so because otherwise the vertical aligner cannot
# currently do the alignment.
#
# -If this container does itself contain a container which has been broken
# open, then it should be broken open to properly show the structure.
#
# -If there is nothing to align, and no other reason to break apart,
# then do not do it.
#
# We will not break open the parens of a long but 'simple' logical expression.
# For example:
#
# This is an example of a simple logical expression and its formatting:
#
#     if ( $bigwasteofspace1 && $bigwasteofspace2
#         || $bigwasteofspace3 && $bigwasteofspace4 )
#
# Most people would prefer this than the 'spacey' version:
#
#     if (
#         $bigwasteofspace1 && $bigwasteofspace2
#         || $bigwasteofspace3 && $bigwasteofspace4
#     )
#
# To illustrate the rules for breaking logical expressions, consider:
#
#             FULLY DENSE:
#             if ( $opt_excl
#                 and ( exists $ids_excl_uc{$id_uc}
#                     or grep $id_uc =~ /$_/, @ids_excl_uc ))
#
# This is on the verge of being difficult to read.  The current default is to
# open it up like this:
#
#             DEFAULT:
#             if (
#                 $opt_excl
#                 and ( exists $ids_excl_uc{$id_uc}
#                     or grep $id_uc =~ /$_/, @ids_excl_uc )
#               )
#
# This is a compromise which tries to avoid being too dense and to spacey.
# A more spaced version would be:
#
#             SPACEY:
#             if (
#                 $opt_excl
#                 and (
#                     exists $ids_excl_uc{$id_uc}
#                     or grep $id_uc =~ /$_/, @ids_excl_uc
#                 )
#               )
#
# Some people might prefer the spacey version -- an option could be added.  The
# innermost expression contains a long block '( exists $ids_...  ')'.
#
# Here is how the logic goes: We will force a break at the 'or' that the
# innermost expression contains, but we will not break apart its opening and
# closing containers because (1) it contains no multi-line sub-containers itself,
# and (2) there is no alignment to be gained by breaking it open like this
#
#             and (
#                 exists $ids_excl_uc{$id_uc}
#                 or grep $id_uc =~ /$_/, @ids_excl_uc
#             )
#
# (although this looks perfectly ok and might be good for long expressions).  The
# outer 'if' container, though, contains a broken sub-container, so it will be
# broken open to avoid too much density.  Also, since it contains no 'or's, there
# will be a forced break at its 'and'.

                # set some flags telling something about this container..
                my $is_simple_logical_expression = 0;
                if (   $item_count_stack[$current_depth] == 0
                    && $saw_opening_structure
                    && $tokens_to_go[$i_opening] eq '('
                    && $is_logical_container{ $container_type[$current_depth] }
                  )
                {

                    # This seems to be a simple logical expression with
                    # no existing breakpoints.  Set a flag to prevent
                    # opening it up.
                    if ( !$has_comma_breakpoints ) {
                        $is_simple_logical_expression = 1;
                    }

                    # This seems to be a simple logical expression with
                    # breakpoints (broken sublists, for example).  Break
                    # at all 'or's and '||'s.
                    else {
                        set_logical_breakpoints($current_depth);
                    }
                } ## end if ( $item_count_stack...)

                if ( $is_long_term
                    && @{ $rfor_semicolon_list[$current_depth] } )
                {
                    set_for_semicolon_breakpoints($current_depth);

                    # open up a long 'for' or 'foreach' container to allow
                    # leading term alignment unless -lp is used.
                    $has_comma_breakpoints = 1
                      unless $rOpts_line_up_parentheses;
                } ## end if ( $is_long_term && ...)

                if (

                    # breaks for code BLOCKS are handled at a higher level
                    !$block_type

                    # we do not need to break at the top level of an 'if'
                    # type expression
                    && !$is_simple_logical_expression

                    ## modification to keep ': (' containers vertically tight;
                    ## but probably better to let user set -vt=1 to avoid
                    ## inconsistency with other paren types
                    ## && ($container_type[$current_depth] ne ':')

                    # otherwise, we require one of these reasons for breaking:
                    && (

                        # - this term has forced line breaks
                        $has_comma_breakpoints

                       # - the opening container is separated from this batch
                       #   for some reason (comment, blank line, code block)
                       # - this is a non-paren container spanning multiple lines
                        || !$saw_opening_structure

                        # - this is a long block contained in another breakable
                        #   container
                        || (   $is_long_term
                            && $container_environment_to_go[$i_opening] ne
                            'BLOCK' )
                    )
                  )
                {

                    # For -lp option, we must put a breakpoint before
                    # the token which has been identified as starting
                    # this indentation level.  This is necessary for
                    # proper alignment.
                    if ( $rOpts_line_up_parentheses && $saw_opening_structure )
                    {
                        my $item = $leading_spaces_to_go[ $i_opening + 1 ];
                        if (   $i_opening + 1 < $max_index_to_go
                            && $types_to_go[ $i_opening + 1 ] eq 'b' )
                        {
                            $item = $leading_spaces_to_go[ $i_opening + 2 ];
                        }
                        if ( defined($item) ) {
                            my $i_start_2 = $item->get_STARTING_INDEX();
                            if (
                                defined($i_start_2)

                                # we are breaking after an opening brace, paren,
                                # so don't break before it too
                                && $i_start_2 ne $i_opening
                              )
                            {

                                # Only break for breakpoints at the same
                                # indentation level as the opening paren
                                my $test1 = $nesting_depth_to_go[$i_opening];
                                my $test2 = $nesting_depth_to_go[$i_start_2];
                                if ( $test2 == $test1 ) {
                                    set_forced_breakpoint( $i_start_2 - 1 );
                                }
                            } ## end if ( defined($i_start_2...))
                        } ## end if ( defined($item) )
                    } ## end if ( $rOpts_line_up_parentheses...)

                    # break after opening structure.
                    # note: break before closing structure will be automatic
                    if ( $minimum_depth <= $current_depth ) {

                        set_forced_breakpoint($i_opening)
                          unless ( $do_not_break_apart
                            || is_unbreakable_container($current_depth) );

                        # break at ',' of lower depth level before opening token
                        if ( $last_comma_index[$depth] ) {
                            set_forced_breakpoint( $last_comma_index[$depth] );
                        }

                        # break at '.' of lower depth level before opening token
                        if ( $last_dot_index[$depth] ) {
                            set_forced_breakpoint( $last_dot_index[$depth] );
                        }

                        # break before opening structure if preeced by another
                        # closing structure and a comma.  This is normally
                        # done by the previous closing brace, but not
                        # if it was a one-line block.
                        if ( $i_opening > 2 ) {
                            my $i_prev =
                              ( $types_to_go[ $i_opening - 1 ] eq 'b' )
                              ? $i_opening - 2
                              : $i_opening - 1;

                            if (   $types_to_go[$i_prev] eq ','
                                && $types_to_go[ $i_prev - 1 ] =~ /^[\)\}]$/ )
                            {
                                set_forced_breakpoint($i_prev);
                            }

                            # also break before something like ':('  or '?('
                            # if appropriate.
                            elsif (
                                $types_to_go[$i_prev] =~ /^([k\:\?]|&&|\|\|)$/ )
                            {
                                my $token_prev = $tokens_to_go[$i_prev];
                                if ( $want_break_before{$token_prev} ) {
                                    set_forced_breakpoint($i_prev);
                                }
                            } ## end elsif ( $types_to_go[$i_prev...])
                        } ## end if ( $i_opening > 2 )
                    } ## end if ( $minimum_depth <=...)

                    # break after comma following closing structure
                    if ( $next_type eq ',' ) {
                        set_forced_breakpoint( $i + 1 );
                    }

                    # break before an '=' following closing structure
                    if (
                        $is_assignment{$next_nonblank_type}
                        && ( $breakpoint_stack[$current_depth] !=
                            $forced_breakpoint_count )
                      )
                    {
                        set_forced_breakpoint($i);
                    } ## end if ( $is_assignment{$next_nonblank_type...})

                    # break at any comma before the opening structure Added
                    # for -lp, but seems to be good in general.  It isn't
                    # obvious how far back to look; the '5' below seems to
                    # work well and will catch the comma in something like
                    #  push @list, myfunc( $param, $param, ..

                    my $icomma = $last_comma_index[$depth];
                    if ( defined($icomma) && ( $i_opening - $icomma ) < 5 ) {
                        unless ( $forced_breakpoint_to_go[$icomma] ) {
                            set_forced_breakpoint($icomma);
                        }
                    }
                }    # end logic to open up a container

                # Break open a logical container open if it was already open
                elsif ($is_simple_logical_expression
                    && $has_old_logical_breakpoints[$current_depth] )
                {
                    set_logical_breakpoints($current_depth);
                }

                # Handle long container which does not get opened up
                elsif ($is_long_term) {

                    # must set fake breakpoint to alert outer containers that
                    # they are complex
                    set_fake_breakpoint();
                } ## end elsif ($is_long_term)

            } ## end elsif ( $depth < $current_depth)

            #------------------------------------------------------------
            # Handle this token
            #------------------------------------------------------------

            $current_depth = $depth;

            # handle comma-arrow
            if ( $type eq '=>' ) {
                next if ( $last_nonblank_type eq '=>' );
                next if $rOpts_break_at_old_comma_breakpoints;
                next if $rOpts_comma_arrow_breakpoints == 3;
                $want_comma_break[$depth]   = 1;
                $index_before_arrow[$depth] = $i_last_nonblank_token;
                next;
            } ## end if ( $type eq '=>' )

            elsif ( $type eq '.' ) {
                $last_dot_index[$depth] = $i;
            }

            # Turn off alignment if we are sure that this is not a list
            # environment.  To be safe, we will do this if we see certain
            # non-list tokens, such as ';', and also the environment is
            # not a list.  Note that '=' could be in any of the = operators
            # (lextest.t). We can't just use the reported environment
            # because it can be incorrect in some cases.
            elsif ( ( $type =~ /^[\;\<\>\~]$/ || $is_assignment{$type} )
                && $container_environment_to_go[$i] ne 'LIST' )
            {
                $dont_align[$depth]         = 1;
                $want_comma_break[$depth]   = 0;
                $index_before_arrow[$depth] = -1;
            } ## end elsif ( ( $type =~ /^[\;\<\>\~]$/...))

            # now just handle any commas
            next unless ( $type eq ',' );

            $last_dot_index[$depth]   = undef;
            $last_comma_index[$depth] = $i;

            # break here if this comma follows a '=>'
            # but not if there is a side comment after the comma
            if ( $want_comma_break[$depth] ) {

                if ( $next_nonblank_type =~ /^[\)\}\]R]$/ ) {
                    if ($rOpts_comma_arrow_breakpoints) {
                        $want_comma_break[$depth] = 0;
                        ##$index_before_arrow[$depth] = -1;
                        next;
                    }
                }

                set_forced_breakpoint($i) unless ( $next_nonblank_type eq '#' );

                # break before the previous token if it looks safe
                # Example of something that we will not try to break before:
                #   DBI::SQL_SMALLINT() => $ado_consts->{adSmallInt},
                # Also we don't want to break at a binary operator (like +):
                # $c->createOval(
                #    $x + $R, $y +
                #    $R => $x - $R,
                #    $y - $R, -fill   => 'black',
                # );
                my $ibreak = $index_before_arrow[$depth] - 1;
                if (   $ibreak > 0
                    && $tokens_to_go[ $ibreak + 1 ] !~ /^[\)\}\]]$/ )
                {
                    if ( $tokens_to_go[$ibreak] eq '-' ) { $ibreak-- }
                    if ( $types_to_go[$ibreak] eq 'b' )  { $ibreak-- }
                    if ( $types_to_go[$ibreak] =~ /^[,wiZCUG\(\{\[]$/ ) {

                        # don't break pointer calls, such as the following:
                        #  File::Spec->curdir  => 1,
                        # (This is tokenized as adjacent 'w' tokens)
                        if ( $tokens_to_go[ $ibreak + 1 ] !~ /^->/ ) {
                            set_forced_breakpoint($ibreak);
                        }
                    } ## end if ( $types_to_go[$ibreak...])
                } ## end if ( $ibreak > 0 && $tokens_to_go...)

                $want_comma_break[$depth]   = 0;
                $index_before_arrow[$depth] = -1;

                # handle list which mixes '=>'s and ','s:
                # treat any list items so far as an interrupted list
                $interrupted_list[$depth] = 1;
                next;
            } ## end if ( $want_comma_break...)

            # break after all commas above starting depth
            if ( $depth < $starting_depth && !$dont_align[$depth] ) {
                set_forced_breakpoint($i) unless ( $next_nonblank_type eq '#' );
                next;
            }

            # add this comma to the list..
            my $item_count = $item_count_stack[$depth];
            if ( $item_count == 0 ) {

                # but do not form a list with no opening structure
                # for example:

                #            open INFILE_COPY, ">$input_file_copy"
                #              or die ("very long message");

                if ( ( $opening_structure_index_stack[$depth] < 0 )
                    && $container_environment_to_go[$i] eq 'BLOCK' )
                {
                    $dont_align[$depth] = 1;
                }
            } ## end if ( $item_count == 0 )

            $comma_index[$depth][$item_count] = $i;
            ++$item_count_stack[$depth];
            if ( $last_nonblank_type =~ /^[iR\]]$/ ) {
                $identifier_count_stack[$depth]++;
            }
        } ## end while ( ++$i <= $max_index_to_go)

        #-------------------------------------------
        # end of loop over all tokens in this batch
        #-------------------------------------------

        # set breaks for any unfinished lists ..
        for ( my $dd = $current_depth ; $dd >= $minimum_depth ; $dd-- ) {

            $interrupted_list[$dd] = 1;
            $has_broken_sublist[$dd] = 1 if ( $dd < $current_depth );
            set_comma_breakpoints($dd);
            set_logical_breakpoints($dd)
              if ( $has_old_logical_breakpoints[$dd] );
            set_for_semicolon_breakpoints($dd);

            # break open container...
            my $i_opening = $opening_structure_index_stack[$dd];
            set_forced_breakpoint($i_opening)
              unless (
                is_unbreakable_container($dd)

                # Avoid a break which would place an isolated ' or "
                # on a line
                || (   $type eq 'Q'
                    && $i_opening >= $max_index_to_go - 2
                    && $token =~ /^['"]$/ )
              );
        } ## end for ( my $dd = $current_depth...)

        # Return a flag indicating if the input file had some good breakpoints.
        # This flag will be used to force a break in a line shorter than the
        # allowed line length.
        if ( $has_old_logical_breakpoints[$current_depth] ) {
            $saw_good_breakpoint = 1;
        }

        # A complex line with one break at an = has a good breakpoint.
        # This is not complex ($total_depth_variation=0):
        # $res1
        #   = 10;
        #
        # This is complex ($total_depth_variation=6):
        # $res2 =
        #  (is_boundp("a", 'self-insert') && is_boundp("b", 'self-insert'));
        elsif ($i_old_assignment_break
            && $total_depth_variation > 4
            && $old_breakpoint_count == 1 )
        {
            $saw_good_breakpoint = 1;
        } ## end elsif ( $i_old_assignment_break...)

        return $saw_good_breakpoint;
    } ## end sub scan_list
}    # end scan_list

sub find_token_starting_list {

    # When testing to see if a block will fit on one line, some
    # previous token(s) may also need to be on the line; particularly
    # if this is a sub call.  So we will look back at least one
    # token. NOTE: This isn't perfect, but not critical, because
    # if we mis-identify a block, it will be wrapped and therefore
    # fixed the next time it is formatted.
    my $i_opening_paren = shift;
    my $i_opening_minus = $i_opening_paren;
    my $im1             = $i_opening_paren - 1;
    my $im2             = $i_opening_paren - 2;
    my $im3             = $i_opening_paren - 3;
    my $typem1          = $types_to_go[$im1];
    my $typem2          = $im2 >= 0 ? $types_to_go[$im2] : 'b';
    if ( $typem1 eq ',' || ( $typem1 eq 'b' && $typem2 eq ',' ) ) {
        $i_opening_minus = $i_opening_paren;
    }
    elsif ( $tokens_to_go[$i_opening_paren] eq '(' ) {
        $i_opening_minus = $im1 if $im1 >= 0;

        # walk back to improve length estimate
        for ( my $j = $im1 ; $j >= 0 ; $j-- ) {
            last if ( $types_to_go[$j] =~ /^[\(\[\{L\}\]\)Rb,]$/ );
            $i_opening_minus = $j;
        }
        if ( $types_to_go[$i_opening_minus] eq 'b' ) { $i_opening_minus++ }
    }
    elsif ( $typem1 eq 'k' ) { $i_opening_minus = $im1 }
    elsif ( $typem1 eq 'b' && $im2 >= 0 && $types_to_go[$im2] eq 'k' ) {
        $i_opening_minus = $im2;
    }
    return $i_opening_minus;
}

{    # begin set_comma_breakpoints_do

    my %is_keyword_with_special_leading_term;

    BEGIN {

        # These keywords have prototypes which allow a special leading item
        # followed by a list
        @_ =
          qw(formline grep kill map printf sprintf push chmod join pack unshift);
        @is_keyword_with_special_leading_term{@_} = (1) x scalar(@_);
    }

    sub set_comma_breakpoints_do {

        # Given a list with some commas, set breakpoints at some of the
        # commas, if necessary, to make it easy to read.  This list is
        # an example:
        my (
            $depth,               $i_opening_paren,  $i_closing_paren,
            $item_count,          $identifier_count, $rcomma_index,
            $next_nonblank_type,  $list_type,        $interrupted,
            $rdo_not_break_apart, $must_break_open,
        ) = @_;

        # nothing to do if no commas seen
        return if ( $item_count < 1 );
        my $i_first_comma     = $$rcomma_index[0];
        my $i_true_last_comma = $$rcomma_index[ $item_count - 1 ];
        my $i_last_comma      = $i_true_last_comma;
        if ( $i_last_comma >= $max_index_to_go ) {
            $i_last_comma = $$rcomma_index[ --$item_count - 1 ];
            return if ( $item_count < 1 );
        }

        #---------------------------------------------------------------
        # find lengths of all items in the list to calculate page layout
        #---------------------------------------------------------------
        my $comma_count = $item_count;
        my @item_lengths;
        my @i_term_begin;
        my @i_term_end;
        my @i_term_comma;
        my $i_prev_plus;
        my @max_length = ( 0, 0 );
        my $first_term_length;
        my $i      = $i_opening_paren;
        my $is_odd = 1;

        for ( my $j = 0 ; $j < $comma_count ; $j++ ) {
            $is_odd      = 1 - $is_odd;
            $i_prev_plus = $i + 1;
            $i           = $$rcomma_index[$j];

            my $i_term_end =
              ( $types_to_go[ $i - 1 ] eq 'b' ) ? $i - 2 : $i - 1;
            my $i_term_begin =
              ( $types_to_go[$i_prev_plus] eq 'b' )
              ? $i_prev_plus + 1
              : $i_prev_plus;
            push @i_term_begin, $i_term_begin;
            push @i_term_end,   $i_term_end;
            push @i_term_comma, $i;

            # note: currently adding 2 to all lengths (for comma and space)
            my $length =
              2 + token_sequence_length( $i_term_begin, $i_term_end );
            push @item_lengths, $length;

            if ( $j == 0 ) {
                $first_term_length = $length;
            }
            else {

                if ( $length > $max_length[$is_odd] ) {
                    $max_length[$is_odd] = $length;
                }
            }
        }

        # now we have to make a distinction between the comma count and item
        # count, because the item count will be one greater than the comma
        # count if the last item is not terminated with a comma
        my $i_b =
          ( $types_to_go[ $i_last_comma + 1 ] eq 'b' )
          ? $i_last_comma + 1
          : $i_last_comma;
        my $i_e =
          ( $types_to_go[ $i_closing_paren - 1 ] eq 'b' )
          ? $i_closing_paren - 2
          : $i_closing_paren - 1;
        my $i_effective_last_comma = $i_last_comma;

        my $last_item_length = token_sequence_length( $i_b + 1, $i_e );

        if ( $last_item_length > 0 ) {

            # add 2 to length because other lengths include a comma and a blank
            $last_item_length += 2;
            push @item_lengths, $last_item_length;
            push @i_term_begin, $i_b + 1;
            push @i_term_end,   $i_e;
            push @i_term_comma, undef;

            my $i_odd = $item_count % 2;

            if ( $last_item_length > $max_length[$i_odd] ) {
                $max_length[$i_odd] = $last_item_length;
            }

            $item_count++;
            $i_effective_last_comma = $i_e + 1;

            if ( $types_to_go[ $i_b + 1 ] =~ /^[iR\]]$/ ) {
                $identifier_count++;
            }
        }

        #---------------------------------------------------------------
        # End of length calculations
        #---------------------------------------------------------------

        #---------------------------------------------------------------
        # Compound List Rule 1:
        # Break at (almost) every comma for a list containing a broken
        # sublist.  This has higher priority than the Interrupted List
        # Rule.
        #---------------------------------------------------------------
        if ( $has_broken_sublist[$depth] ) {

            # Break at every comma except for a comma between two
            # simple, small terms.  This prevents long vertical
            # columns of, say, just 0's.
            my $small_length = 10;    # 2 + actual maximum length wanted

            # We'll insert a break in long runs of small terms to
            # allow alignment in uniform tables.
            my $skipped_count = 0;
            my $columns       = table_columns_available($i_first_comma);
            my $fields        = int( $columns / $small_length );
            if (   $rOpts_maximum_fields_per_table
                && $fields > $rOpts_maximum_fields_per_table )
            {
                $fields = $rOpts_maximum_fields_per_table;
            }
            my $max_skipped_count = $fields - 1;

            my $is_simple_last_term = 0;
            my $is_simple_next_term = 0;
            foreach my $j ( 0 .. $item_count ) {
                $is_simple_last_term = $is_simple_next_term;
                $is_simple_next_term = 0;
                if (   $j < $item_count
                    && $i_term_end[$j] == $i_term_begin[$j]
                    && $item_lengths[$j] <= $small_length )
                {
                    $is_simple_next_term = 1;
                }
                next if $j == 0;
                if (   $is_simple_last_term
                    && $is_simple_next_term
                    && $skipped_count < $max_skipped_count )
                {
                    $skipped_count++;
                }
                else {
                    $skipped_count = 0;
                    my $i = $i_term_comma[ $j - 1 ];
                    last unless defined $i;
                    set_forced_breakpoint($i);
                }
            }

            # always break at the last comma if this list is
            # interrupted; we wouldn't want to leave a terminal '{', for
            # example.
            if ($interrupted) { set_forced_breakpoint($i_true_last_comma) }
            return;
        }

#my ( $a, $b, $c ) = caller();
#print "LISTX: in set_list $a $c interupt=$interrupted count=$item_count
#i_first = $i_first_comma  i_last=$i_last_comma max=$max_index_to_go\n";
#print "depth=$depth has_broken=$has_broken_sublist[$depth] is_multi=$is_multiline opening_paren=($i_opening_paren) \n";

        #---------------------------------------------------------------
        # Interrupted List Rule:
        # A list is is forced to use old breakpoints if it was interrupted
        # by side comments or blank lines, or requested by user.
        #---------------------------------------------------------------
        if (   $rOpts_break_at_old_comma_breakpoints
            || $interrupted
            || $i_opening_paren < 0 )
        {
            copy_old_breakpoints( $i_first_comma, $i_true_last_comma );
            return;
        }

        #---------------------------------------------------------------
        # Looks like a list of items.  We have to look at it and size it up.
        #---------------------------------------------------------------

        my $opening_token = $tokens_to_go[$i_opening_paren];
        my $opening_environment =
          $container_environment_to_go[$i_opening_paren];

        #-------------------------------------------------------------------
        # Return if this will fit on one line
        #-------------------------------------------------------------------

        my $i_opening_minus = find_token_starting_list($i_opening_paren);
        return
          unless excess_line_length( $i_opening_minus, $i_closing_paren ) > 0;

        #-------------------------------------------------------------------
        # Now we know that this block spans multiple lines; we have to set
        # at least one breakpoint -- real or fake -- as a signal to break
        # open any outer containers.
        #-------------------------------------------------------------------
        set_fake_breakpoint();

        # be sure we do not extend beyond the current list length
        if ( $i_effective_last_comma >= $max_index_to_go ) {
            $i_effective_last_comma = $max_index_to_go - 1;
        }

        # Set a flag indicating if we need to break open to keep -lp
        # items aligned.  This is necessary if any of the list terms
        # exceeds the available space after the '('.
        my $need_lp_break_open = $must_break_open;
        if ( $rOpts_line_up_parentheses && !$must_break_open ) {
            my $columns_if_unbroken =
              maximum_line_length($i_opening_minus) -
              total_line_length( $i_opening_minus, $i_opening_paren );
            $need_lp_break_open =
                 ( $max_length[0] > $columns_if_unbroken )
              || ( $max_length[1] > $columns_if_unbroken )
              || ( $first_term_length > $columns_if_unbroken );
        }

        # Specify if the list must have an even number of fields or not.
        # It is generally safest to assume an even number, because the
        # list items might be a hash list.  But if we can be sure that
        # it is not a hash, then we can allow an odd number for more
        # flexibility.
        my $odd_or_even = 2;    # 1 = odd field count ok, 2 = want even count

        if (   $identifier_count >= $item_count - 1
            || $is_assignment{$next_nonblank_type}
            || ( $list_type && $list_type ne '=>' && $list_type !~ /^[\:\?]$/ )
          )
        {
            $odd_or_even = 1;
        }

        # do we have a long first term which should be
        # left on a line by itself?
        my $use_separate_first_term = (
            $odd_or_even == 1       # only if we can use 1 field/line
              && $item_count > 3    # need several items
              && $first_term_length >
              2 * $max_length[0] - 2    # need long first term
              && $first_term_length >
              2 * $max_length[1] - 2    # need long first term
        );

        # or do we know from the type of list that the first term should
        # be placed alone?
        if ( !$use_separate_first_term ) {
            if ( $is_keyword_with_special_leading_term{$list_type} ) {
                $use_separate_first_term = 1;

                # should the container be broken open?
                if ( $item_count < 3 ) {
                    if ( $i_first_comma - $i_opening_paren < 4 ) {
                        $$rdo_not_break_apart = 1;
                    }
                }
                elsif ($first_term_length < 20
                    && $i_first_comma - $i_opening_paren < 4 )
                {
                    my $columns = table_columns_available($i_first_comma);
                    if ( $first_term_length < $columns ) {
                        $$rdo_not_break_apart = 1;
                    }
                }
            }
        }

        # if so,
        if ($use_separate_first_term) {

            # ..set a break and update starting values
            $use_separate_first_term = 1;
            set_forced_breakpoint($i_first_comma);
            $i_opening_paren = $i_first_comma;
            $i_first_comma   = $$rcomma_index[1];
            $item_count--;
            return if $comma_count == 1;
            shift @item_lengths;
            shift @i_term_begin;
            shift @i_term_end;
            shift @i_term_comma;
        }

        # if not, update the metrics to include the first term
        else {
            if ( $first_term_length > $max_length[0] ) {
                $max_length[0] = $first_term_length;
            }
        }

        # Field width parameters
        my $pair_width = ( $max_length[0] + $max_length[1] );
        my $max_width =
          ( $max_length[0] > $max_length[1] ) ? $max_length[0] : $max_length[1];

        # Number of free columns across the page width for laying out tables
        my $columns = table_columns_available($i_first_comma);

        # Estimated maximum number of fields which fit this space
        # This will be our first guess
        my $number_of_fields_max =
          maximum_number_of_fields( $columns, $odd_or_even, $max_width,
            $pair_width );
        my $number_of_fields = $number_of_fields_max;

        # Find the best-looking number of fields
        # and make this our second guess if possible
        my ( $number_of_fields_best, $ri_ragged_break_list,
            $new_identifier_count )
          = study_list_complexity( \@i_term_begin, \@i_term_end, \@item_lengths,
            $max_width );

        if (   $number_of_fields_best != 0
            && $number_of_fields_best < $number_of_fields_max )
        {
            $number_of_fields = $number_of_fields_best;
        }

        # ----------------------------------------------------------------------
        # If we are crowded and the -lp option is being used, try to
        # undo some indentation
        # ----------------------------------------------------------------------
        if (
            $rOpts_line_up_parentheses
            && (
                $number_of_fields == 0
                || (   $number_of_fields == 1
                    && $number_of_fields != $number_of_fields_best )
            )
          )
        {
            my $available_spaces = get_AVAILABLE_SPACES_to_go($i_first_comma);
            if ( $available_spaces > 0 ) {

                my $spaces_wanted = $max_width - $columns;    # for 1 field

                if ( $number_of_fields_best == 0 ) {
                    $number_of_fields_best =
                      get_maximum_fields_wanted( \@item_lengths );
                }

                if ( $number_of_fields_best != 1 ) {
                    my $spaces_wanted_2 =
                      1 + $pair_width - $columns;             # for 2 fields
                    if ( $available_spaces > $spaces_wanted_2 ) {
                        $spaces_wanted = $spaces_wanted_2;
                    }
                }

                if ( $spaces_wanted > 0 ) {
                    my $deleted_spaces =
                      reduce_lp_indentation( $i_first_comma, $spaces_wanted );

                    # redo the math
                    if ( $deleted_spaces > 0 ) {
                        $columns = table_columns_available($i_first_comma);
                        $number_of_fields_max =
                          maximum_number_of_fields( $columns, $odd_or_even,
                            $max_width, $pair_width );
                        $number_of_fields = $number_of_fields_max;

                        if (   $number_of_fields_best == 1
                            && $number_of_fields >= 1 )
                        {
                            $number_of_fields = $number_of_fields_best;
                        }
                    }
                }
            }
        }

        # try for one column if two won't work
        if ( $number_of_fields <= 0 ) {
            $number_of_fields = int( $columns / $max_width );
        }

        # The user can place an upper bound on the number of fields,
        # which can be useful for doing maintenance on tables
        if (   $rOpts_maximum_fields_per_table
            && $number_of_fields > $rOpts_maximum_fields_per_table )
        {
            $number_of_fields = $rOpts_maximum_fields_per_table;
        }

        # How many columns (characters) and lines would this container take
        # if no additional whitespace were added?
        my $packed_columns = token_sequence_length( $i_opening_paren + 1,
            $i_effective_last_comma + 1 );
        if ( $columns <= 0 ) { $columns = 1 }    # avoid divide by zero
        my $packed_lines = 1 + int( $packed_columns / $columns );

        # are we an item contained in an outer list?
        my $in_hierarchical_list = $next_nonblank_type =~ /^[\}\,]$/;

        if ( $number_of_fields <= 0 ) {

#         #---------------------------------------------------------------
#         # We're in trouble.  We can't find a single field width that works.
#         # There is no simple answer here; we may have a single long list
#         # item, or many.
#         #---------------------------------------------------------------
#
#         In many cases, it may be best to not force a break if there is just one
#         comma, because the standard continuation break logic will do a better
#         job without it.
#
#         In the common case that all but one of the terms can fit
#         on a single line, it may look better not to break open the
#         containing parens.  Consider, for example
#
#             $color =
#               join ( '/',
#                 sort { $color_value{$::a} <=> $color_value{$::b}; }
#                 keys %colors );
#
#         which will look like this with the container broken:
#
#             $color = join (
#                 '/',
#                 sort { $color_value{$::a} <=> $color_value{$::b}; } keys %colors
#             );
#
#         Here is an example of this rule for a long last term:
#
#             log_message( 0, 256, 128,
#                 "Number of routes in adj-RIB-in to be considered: $peercount" );
#
#         And here is an example with a long first term:
#
#         $s = sprintf(
# "%2d wallclock secs (%$f usr %$f sys + %$f cusr %$f csys = %$f CPU)",
#             $r, $pu, $ps, $cu, $cs, $tt
#           )
#           if $style eq 'all';

            my $i_last_comma = $$rcomma_index[ $comma_count - 1 ];
            my $long_last_term = excess_line_length( 0, $i_last_comma ) <= 0;
            my $long_first_term =
              excess_line_length( $i_first_comma + 1, $max_index_to_go ) <= 0;

            # break at every comma ...
            if (

                # if requested by user or is best looking
                $number_of_fields_best == 1

                # or if this is a sublist of a larger list
                || $in_hierarchical_list

                # or if multiple commas and we dont have a long first or last
                # term
                || ( $comma_count > 1
                    && !( $long_last_term || $long_first_term ) )
              )
            {
                foreach ( 0 .. $comma_count - 1 ) {
                    set_forced_breakpoint( $$rcomma_index[$_] );
                }
            }
            elsif ($long_last_term) {

                set_forced_breakpoint($i_last_comma);
                $$rdo_not_break_apart = 1 unless $must_break_open;
            }
            elsif ($long_first_term) {

                set_forced_breakpoint($i_first_comma);
            }
            else {

                # let breaks be defined by default bond strength logic
            }
            return;
        }

        # --------------------------------------------------------
        # We have a tentative field count that seems to work.
        # How many lines will this require?
        # --------------------------------------------------------
        my $formatted_lines = $item_count / ($number_of_fields);
        if ( $formatted_lines != int $formatted_lines ) {
            $formatted_lines = 1 + int $formatted_lines;
        }

        # So far we've been trying to fill out to the right margin.  But
        # compact tables are easier to read, so let's see if we can use fewer
        # fields without increasing the number of lines.
        $number_of_fields =
          compactify_table( $item_count, $number_of_fields, $formatted_lines,
            $odd_or_even );

        # How many spaces across the page will we fill?
        my $columns_per_line =
          ( int $number_of_fields / 2 ) * $pair_width +
          ( $number_of_fields % 2 ) * $max_width;

        my $formatted_columns;

        if ( $number_of_fields > 1 ) {
            $formatted_columns =
              ( $pair_width * ( int( $item_count / 2 ) ) +
                  ( $item_count % 2 ) * $max_width );
        }
        else {
            $formatted_columns = $max_width * $item_count;
        }
        if ( $formatted_columns < $packed_columns ) {
            $formatted_columns = $packed_columns;
        }

        my $unused_columns = $formatted_columns - $packed_columns;

        # set some empirical parameters to help decide if we should try to
        # align; high sparsity does not look good, especially with few lines
        my $sparsity = ($unused_columns) / ($formatted_columns);
        my $max_allowed_sparsity =
            ( $item_count < 3 )    ? 0.1
          : ( $packed_lines == 1 ) ? 0.15
          : ( $packed_lines == 2 ) ? 0.4
          :                          0.7;

        # Begin check for shortcut methods, which avoid treating a list
        # as a table for relatively small parenthesized lists.  These
        # are usually easier to read if not formatted as tables.
        if (
            $packed_lines <= 2                    # probably can fit in 2 lines
            && $item_count < 9                    # doesn't have too many items
            && $opening_environment eq 'BLOCK'    # not a sub-container
            && $opening_token eq '('              # is paren list
          )
        {

            # Shortcut method 1: for -lp and just one comma:
            # This is a no-brainer, just break at the comma.
            if (
                $rOpts_line_up_parentheses    # -lp
                && $item_count == 2           # two items, one comma
                && !$must_break_open
              )
            {
                my $i_break = $$rcomma_index[0];
                set_forced_breakpoint($i_break);
                $$rdo_not_break_apart = 1;
                set_non_alignment_flags( $comma_count, $rcomma_index );
                return;

            }

            # method 2 is for most small ragged lists which might look
            # best if not displayed as a table.
            if (
                ( $number_of_fields == 2 && $item_count == 3 )
                || (
                    $new_identifier_count > 0    # isn't all quotes
                    && $sparsity > 0.15
                )    # would be fairly spaced gaps if aligned
              )
            {

                my $break_count = set_ragged_breakpoints( \@i_term_comma,
                    $ri_ragged_break_list );
                ++$break_count if ($use_separate_first_term);

                # NOTE: we should really use the true break count here,
                # which can be greater if there are large terms and
                # little space, but usually this will work well enough.
                unless ($must_break_open) {

                    if ( $break_count <= 1 ) {
                        $$rdo_not_break_apart = 1;
                    }
                    elsif ( $rOpts_line_up_parentheses && !$need_lp_break_open )
                    {
                        $$rdo_not_break_apart = 1;
                    }
                }
                set_non_alignment_flags( $comma_count, $rcomma_index );
                return;
            }

        }    # end shortcut methods

        # debug stuff

        FORMATTER_DEBUG_FLAG_SPARSE && do {
            print STDOUT
"SPARSE:cols=$columns commas=$comma_count items:$item_count ids=$identifier_count pairwidth=$pair_width fields=$number_of_fields lines packed: $packed_lines packed_cols=$packed_columns fmtd:$formatted_lines cols /line:$columns_per_line  unused:$unused_columns fmtd:$formatted_columns sparsity=$sparsity allow=$max_allowed_sparsity\n";

        };

        #---------------------------------------------------------------
        # Compound List Rule 2:
        # If this list is too long for one line, and it is an item of a
        # larger list, then we must format it, regardless of sparsity
        # (ian.t).  One reason that we have to do this is to trigger
        # Compound List Rule 1, above, which causes breaks at all commas of
        # all outer lists.  In this way, the structure will be properly
        # displayed.
        #---------------------------------------------------------------

        # Decide if this list is too long for one line unless broken
        my $total_columns = table_columns_available($i_opening_paren);
        my $too_long      = $packed_columns > $total_columns;

        # For a paren list, include the length of the token just before the
        # '(' because this is likely a sub call, and we would have to
        # include the sub name on the same line as the list.  This is still
        # imprecise, but not too bad.  (steve.t)
        if ( !$too_long && $i_opening_paren > 0 && $opening_token eq '(' ) {

            $too_long = excess_line_length( $i_opening_minus,
                $i_effective_last_comma + 1 ) > 0;
        }

        # FIXME: For an item after a '=>', try to include the length of the
        # thing before the '=>'.  This is crude and should be improved by
        # actually looking back token by token.
        if ( !$too_long && $i_opening_paren > 0 && $list_type eq '=>' ) {
            my $i_opening_minus = $i_opening_paren - 4;
            if ( $i_opening_minus >= 0 ) {
                $too_long = excess_line_length( $i_opening_minus,
                    $i_effective_last_comma + 1 ) > 0;
            }
        }

        # Always break lists contained in '[' and '{' if too long for 1 line,
        # and always break lists which are too long and part of a more complex
        # structure.
        my $must_break_open_container = $must_break_open
          || ( $too_long
            && ( $in_hierarchical_list || $opening_token ne '(' ) );

#print "LISTX: next=$next_nonblank_type  avail cols=$columns packed=$packed_columns must format = $must_break_open_container too-long=$too_long  opening=$opening_token list_type=$list_type formatted_lines=$formatted_lines  packed=$packed_lines max_sparsity= $max_allowed_sparsity sparsity=$sparsity \n";

        #---------------------------------------------------------------
        # The main decision:
        # Now decide if we will align the data into aligned columns.  Do not
        # attempt to align columns if this is a tiny table or it would be
        # too spaced.  It seems that the more packed lines we have, the
        # sparser the list that can be allowed and still look ok.
        #---------------------------------------------------------------

        if (   ( $formatted_lines < 3 && $packed_lines < $formatted_lines )
            || ( $formatted_lines < 2 )
            || ( $unused_columns > $max_allowed_sparsity * $formatted_columns )
          )
        {

            #---------------------------------------------------------------
            # too sparse: would look ugly if aligned in a table;
            #---------------------------------------------------------------

            # use old breakpoints if this is a 'big' list
            # FIXME: goal is to improve set_ragged_breakpoints so that
            # this is not necessary.
            if ( $packed_lines > 2 && $item_count > 10 ) {
                write_logfile_entry("List sparse: using old breakpoints\n");
                copy_old_breakpoints( $i_first_comma, $i_last_comma );
            }

            # let the continuation logic handle it if 2 lines
            else {

                my $break_count = set_ragged_breakpoints( \@i_term_comma,
                    $ri_ragged_break_list );
                ++$break_count if ($use_separate_first_term);

                unless ($must_break_open_container) {
                    if ( $break_count <= 1 ) {
                        $$rdo_not_break_apart = 1;
                    }
                    elsif ( $rOpts_line_up_parentheses && !$need_lp_break_open )
                    {
                        $$rdo_not_break_apart = 1;
                    }
                }
                set_non_alignment_flags( $comma_count, $rcomma_index );
            }
            return;
        }

        #---------------------------------------------------------------
        # go ahead and format as a table
        #---------------------------------------------------------------
        write_logfile_entry(
            "List: auto formatting with $number_of_fields fields/row\n");

        my $j_first_break =
          $use_separate_first_term ? $number_of_fields : $number_of_fields - 1;

        for (
            my $j = $j_first_break ;
            $j < $comma_count ;
            $j += $number_of_fields
          )
        {
            my $i = $$rcomma_index[$j];
            set_forced_breakpoint($i);
        }
        return;
    }
}

sub set_non_alignment_flags {

    # set flag which indicates that these commas should not be
    # aligned
    my ( $comma_count, $rcomma_index ) = @_;
    foreach ( 0 .. $comma_count - 1 ) {
        $matching_token_to_go[ $$rcomma_index[$_] ] = 1;
    }
}

sub study_list_complexity {

    # Look for complex tables which should be formatted with one term per line.
    # Returns the following:
    #
    #  \@i_ragged_break_list = list of good breakpoints to avoid lines
    #    which are hard to read
    #  $number_of_fields_best = suggested number of fields based on
    #    complexity; = 0 if any number may be used.
    #
    my ( $ri_term_begin, $ri_term_end, $ritem_lengths, $max_width ) = @_;
    my $item_count            = @{$ri_term_begin};
    my $complex_item_count    = 0;
    my $number_of_fields_best = $rOpts_maximum_fields_per_table;
    my $i_max                 = @{$ritem_lengths} - 1;
    ##my @item_complexity;

    my $i_last_last_break = -3;
    my $i_last_break      = -2;
    my @i_ragged_break_list;

    my $definitely_complex = 30;
    my $definitely_simple  = 12;
    my $quote_count        = 0;

    for my $i ( 0 .. $i_max ) {
        my $ib = $ri_term_begin->[$i];
        my $ie = $ri_term_end->[$i];

        # define complexity: start with the actual term length
        my $weighted_length = ( $ritem_lengths->[$i] - 2 );

        ##TBD: join types here and check for variations
        ##my $str=join "", @tokens_to_go[$ib..$ie];

        my $is_quote = 0;
        if ( $types_to_go[$ib] =~ /^[qQ]$/ ) {
            $is_quote = 1;
            $quote_count++;
        }
        elsif ( $types_to_go[$ib] =~ /^[w\-]$/ ) {
            $quote_count++;
        }

        if ( $ib eq $ie ) {
            if ( $is_quote && $tokens_to_go[$ib] =~ /\s/ ) {
                $complex_item_count++;
                $weighted_length *= 2;
            }
            else {
            }
        }
        else {
            if ( grep { $_ eq 'b' } @types_to_go[ $ib .. $ie ] ) {
                $complex_item_count++;
                $weighted_length *= 2;
            }
            if ( grep { $_ eq '..' } @types_to_go[ $ib .. $ie ] ) {
                $weighted_length += 4;
            }
        }

        # add weight for extra tokens.
        $weighted_length += 2 * ( $ie - $ib );

##        my $BUB = join '', @tokens_to_go[$ib..$ie];
##        print "# COMPLEXITY:$weighted_length   $BUB\n";

##push @item_complexity, $weighted_length;

        # now mark a ragged break after this item it if it is 'long and
        # complex':
        if ( $weighted_length >= $definitely_complex ) {

            # if we broke after the previous term
            # then break before it too
            if (   $i_last_break == $i - 1
                && $i > 1
                && $i_last_last_break != $i - 2 )
            {

                ## FIXME: don't strand a small term
                pop @i_ragged_break_list;
                push @i_ragged_break_list, $i - 2;
                push @i_ragged_break_list, $i - 1;
            }

            push @i_ragged_break_list, $i;
            $i_last_last_break = $i_last_break;
            $i_last_break      = $i;
        }

        # don't break before a small last term -- it will
        # not look good on a line by itself.
        elsif ($i == $i_max
            && $i_last_break == $i - 1
            && $weighted_length <= $definitely_simple )
        {
            pop @i_ragged_break_list;
        }
    }

    my $identifier_count = $i_max + 1 - $quote_count;

    # Need more tuning here..
    if (   $max_width > 12
        && $complex_item_count > $item_count / 2
        && $number_of_fields_best != 2 )
    {
        $number_of_fields_best = 1;
    }

    return ( $number_of_fields_best, \@i_ragged_break_list, $identifier_count );
}

sub get_maximum_fields_wanted {

    # Not all tables look good with more than one field of items.
    # This routine looks at a table and decides if it should be
    # formatted with just one field or not.
    # This coding is still under development.
    my ($ritem_lengths) = @_;

    my $number_of_fields_best = 0;

    # For just a few items, we tentatively assume just 1 field.
    my $item_count = @{$ritem_lengths};
    if ( $item_count <= 5 ) {
        $number_of_fields_best = 1;
    }

    # For larger tables, look at it both ways and see what looks best
    else {

        my $is_odd            = 1;
        my @max_length        = ( 0, 0 );
        my @last_length_2     = ( undef, undef );
        my @first_length_2    = ( undef, undef );
        my $last_length       = undef;
        my $total_variation_1 = 0;
        my $total_variation_2 = 0;
        my @total_variation_2 = ( 0, 0 );
        for ( my $j = 0 ; $j < $item_count ; $j++ ) {

            $is_odd = 1 - $is_odd;
            my $length = $ritem_lengths->[$j];
            if ( $length > $max_length[$is_odd] ) {
                $max_length[$is_odd] = $length;
            }

            if ( defined($last_length) ) {
                my $dl = abs( $length - $last_length );
                $total_variation_1 += $dl;
            }
            $last_length = $length;

            my $ll = $last_length_2[$is_odd];
            if ( defined($ll) ) {
                my $dl = abs( $length - $ll );
                $total_variation_2[$is_odd] += $dl;
            }
            else {
                $first_length_2[$is_odd] = $length;
            }
            $last_length_2[$is_odd] = $length;
        }
        $total_variation_2 = $total_variation_2[0] + $total_variation_2[1];

        my $factor = ( $item_count > 10 ) ? 1 : ( $item_count > 5 ) ? 0.75 : 0;
        unless ( $total_variation_2 < $factor * $total_variation_1 ) {
            $number_of_fields_best = 1;
        }
    }
    return ($number_of_fields_best);
}

sub table_columns_available {
    my $i_first_comma = shift;
    my $columns =
      maximum_line_length($i_first_comma) -
      leading_spaces_to_go($i_first_comma);

    # Patch: the vertical formatter does not line up lines whose lengths
    # exactly equal the available line length because of allowances
    # that must be made for side comments.  Therefore, the number of
    # available columns is reduced by 1 character.
    $columns -= 1;
    return $columns;
}

sub maximum_number_of_fields {

    # how many fields will fit in the available space?
    my ( $columns, $odd_or_even, $max_width, $pair_width ) = @_;
    my $max_pairs        = int( $columns / $pair_width );
    my $number_of_fields = $max_pairs * 2;
    if (   $odd_or_even == 1
        && $max_pairs * $pair_width + $max_width <= $columns )
    {
        $number_of_fields++;
    }
    return $number_of_fields;
}

sub compactify_table {

    # given a table with a certain number of fields and a certain number
    # of lines, see if reducing the number of fields will make it look
    # better.
    my ( $item_count, $number_of_fields, $formatted_lines, $odd_or_even ) = @_;
    if ( $number_of_fields >= $odd_or_even * 2 && $formatted_lines > 0 ) {
        my $min_fields;

        for (
            $min_fields = $number_of_fields ;
            $min_fields >= $odd_or_even
            && $min_fields * $formatted_lines >= $item_count ;
            $min_fields -= $odd_or_even
          )
        {
            $number_of_fields = $min_fields;
        }
    }
    return $number_of_fields;
}

sub set_ragged_breakpoints {

    # Set breakpoints in a list that cannot be formatted nicely as a
    # table.
    my ( $ri_term_comma, $ri_ragged_break_list ) = @_;

    my $break_count = 0;
    foreach (@$ri_ragged_break_list) {
        my $j = $ri_term_comma->[$_];
        if ($j) {
            set_forced_breakpoint($j);
            $break_count++;
        }
    }
    return $break_count;
}

sub copy_old_breakpoints {
    my ( $i_first_comma, $i_last_comma ) = @_;
    for my $i ( $i_first_comma .. $i_last_comma ) {
        if ( $old_breakpoint_to_go[$i] ) {
            set_forced_breakpoint($i);
        }
    }
}

sub set_nobreaks {
    my ( $i, $j ) = @_;
    if ( $i >= 0 && $i <= $j && $j <= $max_index_to_go ) {

        FORMATTER_DEBUG_FLAG_NOBREAK && do {
            my ( $a, $b, $c ) = caller();
            print STDOUT
"NOBREAK: forced_breakpoint $forced_breakpoint_count from $a $c with i=$i max=$max_index_to_go type=$types_to_go[$i]\n";
        };

        @nobreak_to_go[ $i .. $j ] = (1) x ( $j - $i + 1 );
    }

    # shouldn't happen; non-critical error
    else {
        FORMATTER_DEBUG_FLAG_NOBREAK && do {
            my ( $a, $b, $c ) = caller();
            print STDOUT
              "NOBREAK ERROR: from $a $c with i=$i j=$j max=$max_index_to_go\n";
        };
    }
}

sub set_fake_breakpoint {

    # Just bump up the breakpoint count as a signal that there are breaks.
    # This is useful if we have breaks but may want to postpone deciding where
    # to make them.
    $forced_breakpoint_count++;
}

sub set_forced_breakpoint {
    my $i = shift;

    return unless defined $i && $i >= 0;

    # when called with certain tokens, use bond strengths to decide
    # if we break before or after it
    my $token = $tokens_to_go[$i];

    if ( $token =~ /^([\=\.\,\:\?]|and|or|xor|&&|\|\|)$/ ) {
        if ( $want_break_before{$token} && $i >= 0 ) { $i-- }
    }

    # breaks are forced before 'if' and 'unless'
    elsif ( $is_if_unless{$token} ) { $i-- }

    if ( $i >= 0 && $i <= $max_index_to_go ) {
        my $i_nonblank = ( $types_to_go[$i] ne 'b' ) ? $i : $i - 1;

        FORMATTER_DEBUG_FLAG_FORCE && do {
            my ( $a, $b, $c ) = caller();
            print STDOUT
"FORCE $forced_breakpoint_count from $a $c with i=$i_nonblank max=$max_index_to_go tok=$tokens_to_go[$i_nonblank] type=$types_to_go[$i_nonblank] nobr=$nobreak_to_go[$i_nonblank]\n";
        };

        if ( $i_nonblank >= 0 && $nobreak_to_go[$i_nonblank] == 0 ) {
            $forced_breakpoint_to_go[$i_nonblank] = 1;

            if ( $i_nonblank > $index_max_forced_break ) {
                $index_max_forced_break = $i_nonblank;
            }
            $forced_breakpoint_count++;
            $forced_breakpoint_undo_stack[ $forced_breakpoint_undo_count++ ] =
              $i_nonblank;

            # if we break at an opening container..break at the closing
            if ( $tokens_to_go[$i_nonblank] =~ /^[\{\[\(\?]$/ ) {
                set_closing_breakpoint($i_nonblank);
            }
        }
    }
}

sub clear_breakpoint_undo_stack {
    $forced_breakpoint_undo_count = 0;
}

sub undo_forced_breakpoint_stack {

    my $i_start = shift;
    if ( $i_start < 0 ) {
        $i_start = 0;
        my ( $a, $b, $c ) = caller();
        warning(
"Program Bug: undo_forced_breakpoint_stack from $a $c has i=$i_start "
        );
    }

    while ( $forced_breakpoint_undo_count > $i_start ) {
        my $i =
          $forced_breakpoint_undo_stack[ --$forced_breakpoint_undo_count ];
        if ( $i >= 0 && $i <= $max_index_to_go ) {
            $forced_breakpoint_to_go[$i] = 0;
            $forced_breakpoint_count--;

            FORMATTER_DEBUG_FLAG_UNDOBP && do {
                my ( $a, $b, $c ) = caller();
                print STDOUT
"UNDOBP: undo forced_breakpoint i=$i $forced_breakpoint_undo_count from $a $c max=$max_index_to_go\n";
            };
        }

        # shouldn't happen, but not a critical error
        else {
            FORMATTER_DEBUG_FLAG_UNDOBP && do {
                my ( $a, $b, $c ) = caller();
                print STDOUT
"Program Bug: undo_forced_breakpoint from $a $c has i=$i but max=$max_index_to_go";
            };
        }
    }
}

{    # begin recombine_breakpoints

    my %is_amp_amp;
    my %is_ternary;
    my %is_math_op;
    my %is_plus_minus;
    my %is_mult_div;

    BEGIN {

        @_ = qw( && || );
        @is_amp_amp{@_} = (1) x scalar(@_);

        @_ = qw( ? : );
        @is_ternary{@_} = (1) x scalar(@_);

        @_ = qw( + - * / );
        @is_math_op{@_} = (1) x scalar(@_);

        @_ = qw( + - );
        @is_plus_minus{@_} = (1) x scalar(@_);

        @_ = qw( * / );
        @is_mult_div{@_} = (1) x scalar(@_);
    }

    sub recombine_breakpoints {

        # sub set_continuation_breaks is very liberal in setting line breaks
        # for long lines, always setting breaks at good breakpoints, even
        # when that creates small lines.  Sometimes small line fragments
        # are produced which would look better if they were combined.
        # That's the task of this routine.
        #
        # We are given indexes to the current lines:
        # $ri_beg = ref to array of BEGinning indexes of each line
        # $ri_end = ref to array of ENDing indexes of each line
        my ( $ri_beg, $ri_end ) = @_;

        # Make a list of all good joining tokens between the lines
        # n-1 and n.
        my @joint;
        my $nmax = @$ri_end - 1;
        for my $n ( 1 .. $nmax ) {
            my $ibeg_1 = $$ri_beg[ $n - 1 ];
            my $iend_1 = $$ri_end[ $n - 1 ];
            my $iend_2 = $$ri_end[$n];
            my $ibeg_2 = $$ri_beg[$n];

            my ( $itok, $itokp, $itokm );

            foreach my $itest ( $iend_1, $ibeg_2 ) {
                my $type = $types_to_go[$itest];
                if (   $is_math_op{$type}
                    || $is_amp_amp{$type}
                    || $is_assignment{$type}
                    || $type eq ':' )
                {
                    $itok = $itest;
                }
            }
            $joint[$n] = [$itok];
        }

        my $more_to_do = 1;

        # We keep looping over all of the lines of this batch
        # until there are no more possible recombinations
        my $nmax_last = @$ri_end;
        while ($more_to_do) {
            my $n_best = 0;
            my $bs_best;
            my $n;
            my $nmax = @$ri_end - 1;

            # Safety check for infinite loop
            unless ( $nmax < $nmax_last ) {

                # Shouldn't happen because splice below decreases nmax on each
                # pass.
                Perl::Tidy::Die
                  "Program bug-infinite loop in recombine breakpoints\n";
            }
            $nmax_last  = $nmax;
            $more_to_do = 0;
            my $previous_outdentable_closing_paren;
            my $leading_amp_count = 0;
            my $this_line_is_semicolon_terminated;

            # loop over all remaining lines in this batch
            for $n ( 1 .. $nmax ) {

                #----------------------------------------------------------
                # If we join the current pair of lines,
                # line $n-1 will become the left part of the joined line
                # line $n will become the right part of the joined line
                #
                # Here are Indexes of the endpoint tokens of the two lines:
                #
                #  -----line $n-1--- | -----line $n-----
                #  $ibeg_1   $iend_1 | $ibeg_2   $iend_2
                #                    ^
                #                    |
                # We want to decide if we should remove the line break
                # between the tokens at $iend_1 and $ibeg_2
                #
                # We will apply a number of ad-hoc tests to see if joining
                # here will look ok.  The code will just issue a 'next'
                # command if the join doesn't look good.  If we get through
                # the gauntlet of tests, the lines will be recombined.
                #----------------------------------------------------------
                #
                # beginning and ending tokens of the lines we are working on
                my $ibeg_1    = $$ri_beg[ $n - 1 ];
                my $iend_1    = $$ri_end[ $n - 1 ];
                my $iend_2    = $$ri_end[$n];
                my $ibeg_2    = $$ri_beg[$n];
                my $ibeg_nmax = $$ri_beg[$nmax];

                my $type_iend_1 = $types_to_go[$iend_1];
                my $type_iend_2 = $types_to_go[$iend_2];
                my $type_ibeg_1 = $types_to_go[$ibeg_1];
                my $type_ibeg_2 = $types_to_go[$ibeg_2];

                # some beginning indexes of other lines, which may not exist
                my $ibeg_0 = $n > 1          ? $$ri_beg[ $n - 2 ] : -1;
                my $ibeg_3 = $n < $nmax      ? $$ri_beg[ $n + 1 ] : -1;
                my $ibeg_4 = $n + 2 <= $nmax ? $$ri_beg[ $n + 2 ] : -1;

                my $bs_tweak = 0;

                #my $depth_increase=( $nesting_depth_to_go[$ibeg_2] -
                #        $nesting_depth_to_go[$ibeg_1] );

                FORMATTER_DEBUG_FLAG_RECOMBINE && do {
                    print STDERR
"RECOMBINE: n=$n imid=$iend_1 if=$ibeg_1 type=$type_ibeg_1 =$tokens_to_go[$ibeg_1] next_type=$type_ibeg_2 next_tok=$tokens_to_go[$ibeg_2]\n";
                };

                # If line $n is the last line, we set some flags and
                # do any special checks for it
                if ( $n == $nmax ) {

                    # a terminal '{' should stay where it is
                    next if $type_ibeg_2 eq '{';

                    # set flag if statement $n ends in ';'
                    $this_line_is_semicolon_terminated = $type_iend_2 eq ';'

                      # with possible side comment
                      || ( $type_iend_2 eq '#'
                        && $iend_2 - $ibeg_2 >= 2
                        && $types_to_go[ $iend_2 - 2 ] eq ';'
                        && $types_to_go[ $iend_2 - 1 ] eq 'b' );
                }

                #----------------------------------------------------------
                # Recombine Section 1:
                # Examine the special token joining this line pair, if any.
                # Put as many tests in this section to avoid duplicate code and
                # to make formatting independent of whether breaks are to the
                # left or right of an operator.
                #----------------------------------------------------------

                my ($itok) = @{ $joint[$n] };
                if ($itok) {

                    # FIXME: Patch - may not be necessary
                    my $iend_1 =
                        $type_iend_1 eq 'b'
                      ? $iend_1 - 1
                      : $iend_1;

                    my $iend_2 =
                        $type_iend_2 eq 'b'
                      ? $iend_2 - 1
                      : $iend_2;
                    ## END PATCH

                    my $type = $types_to_go[$itok];

                    if ( $type eq ':' ) {

                   # do not join at a colon unless it disobeys the break request
                        if ( $itok eq $iend_1 ) {
                            next unless $want_break_before{$type};
                        }
                        else {
                            $leading_amp_count++;
                            next if $want_break_before{$type};
                        }
                    } ## end if ':'

                    # handle math operators + - * /
                    elsif ( $is_math_op{$type} ) {

                        # Combine these lines if this line is a single
                        # number, or if it is a short term with same
                        # operator as the previous line.  For example, in
                        # the following code we will combine all of the
                        # short terms $A, $B, $C, $D, $E, $F, together
                        # instead of leaving them one per line:
                        #  my $time =
                        #    $A * $B * $C * $D * $E * $F *
                        #    ( 2. * $eps * $sigma * $area ) *
                        #    ( 1. / $tcold**3 - 1. / $thot**3 );

                        # This can be important in math-intensive code.

                        my $good_combo;

                        my $itokp  = min( $inext_to_go[$itok],  $iend_2 );
                        my $itokpp = min( $inext_to_go[$itokp], $iend_2 );
                        my $itokm  = max( $iprev_to_go[$itok],  $ibeg_1 );
                        my $itokmm = max( $iprev_to_go[$itokm], $ibeg_1 );

                        # check for a number on the right
                        if ( $types_to_go[$itokp] eq 'n' ) {

                            # ok if nothing else on right
                            if ( $itokp == $iend_2 ) {
                                $good_combo = 1;
                            }
                            else {

                                # look one more token to right..
                                # okay if math operator or some termination
                                $good_combo =
                                  ( ( $itokpp == $iend_2 )
                                      && $is_math_op{ $types_to_go[$itokpp] } )
                                  || $types_to_go[$itokpp] =~ /^[#,;]$/;
                            }
                        }

                        # check for a number on the left
                        if ( !$good_combo && $types_to_go[$itokm] eq 'n' ) {

                            # okay if nothing else to left
                            if ( $itokm == $ibeg_1 ) {
                                $good_combo = 1;
                            }

                            # otherwise look one more token to left
                            else {

                                # okay if math operator, comma, or assignment
                                $good_combo = ( $itokmm == $ibeg_1 )
                                  && ( $is_math_op{ $types_to_go[$itokmm] }
                                    || $types_to_go[$itokmm] =~ /^[,]$/
                                    || $is_assignment{ $types_to_go[$itokmm] }
                                  );
                            }
                        }

                        # look for a single short token either side of the
                        # operator
                        if ( !$good_combo ) {

                            # Slight adjustment factor to make results
                            # independent of break before or after operator in
                            # long summed lists.  (An operator and a space make
                            # two spaces).
                            my $two = ( $itok eq $iend_1 ) ? 2 : 0;

                            $good_combo =

                              # numbers or id's on both sides of this joint
                              $types_to_go[$itokp] =~ /^[in]$/
                              && $types_to_go[$itokm] =~ /^[in]$/

                              # one of the two lines must be short:
                              && (
                                (
                                    # no more than 2 nonblank tokens right of
                                    # joint
                                    $itokpp == $iend_2

                                    # short
                                    && token_sequence_length( $itokp, $iend_2 )
                                    < $two +
                                    $rOpts_short_concatenation_item_length
                                )
                                || (
                                    # no more than 2 nonblank tokens left of
                                    # joint
                                    $itokmm == $ibeg_1

                                    # short
                                    && token_sequence_length( $ibeg_1, $itokm )
                                    < 2 - $two +
                                    $rOpts_short_concatenation_item_length
                                )

                              )

                              # keep pure terms; don't mix +- with */
                              && !(
                                $is_plus_minus{$type}
                                && (   $is_mult_div{ $types_to_go[$itokmm] }
                                    || $is_mult_div{ $types_to_go[$itokpp] } )
                              )
                              && !(
                                $is_mult_div{$type}
                                && (   $is_plus_minus{ $types_to_go[$itokmm] }
                                    || $is_plus_minus{ $types_to_go[$itokpp] } )
                              )

                              ;
                        }

                        # it is also good to combine if we can reduce to 2 lines
                        if ( !$good_combo ) {

                            # index on other line where same token would be in a
                            # long chain.
                            my $iother =
                              ( $itok == $iend_1 ) ? $iend_2 : $ibeg_1;

                            $good_combo =
                                 $n == 2
                              && $n == $nmax
                              && $types_to_go[$iother] ne $type;
                        }

                        next unless ($good_combo);

                    } ## end math

                    elsif ( $is_amp_amp{$type} ) {
                        ##TBD
                    } ## end &&, ||

                    elsif ( $is_assignment{$type} ) {
                        ##TBD
                    } ## end assignment
                }

                #----------------------------------------------------------
                # Recombine Section 2:
                # Examine token at $iend_1 (right end of first line of pair)
                #----------------------------------------------------------

                # an isolated '}' may join with a ';' terminated segment
                if ( $type_iend_1 eq '}' ) {

                    # Check for cases where combining a semicolon terminated
                    # statement with a previous isolated closing paren will
                    # allow the combined line to be outdented.  This is
                    # generally a good move.  For example, we can join up
                    # the last two lines here:
                    #  (
                    #      $dev,  $ino,   $mode,  $nlink, $uid,     $gid, $rdev,
                    #      $size, $atime, $mtime, $ctime, $blksize, $blocks
                    #    )
                    #    = stat($file);
                    #
                    # to get:
                    #  (
                    #      $dev,  $ino,   $mode,  $nlink, $uid,     $gid, $rdev,
                    #      $size, $atime, $mtime, $ctime, $blksize, $blocks
                    #  ) = stat($file);
                    #
                    # which makes the parens line up.
                    #
                    # Another example, from Joe Matarazzo, probably looks best
                    # with the 'or' clause appended to the trailing paren:
                    #  $self->some_method(
                    #      PARAM1 => 'foo',
                    #      PARAM2 => 'bar'
                    #  ) or die "Some_method didn't work";
                    #
                    # But we do not want to do this for something like the -lp
                    # option where the paren is not outdentable because the
                    # trailing clause will be far to the right.
                    #
                    # The logic here is synchronized with the logic in sub
                    # sub set_adjusted_indentation, which actually does
                    # the outdenting.
                    #
                    $previous_outdentable_closing_paren =
                      $this_line_is_semicolon_terminated

                      # only one token on last line
                      && $ibeg_1 == $iend_1

                      # must be structural paren
                      && $tokens_to_go[$iend_1] eq ')'

                      # style must allow outdenting,
                      && !$closing_token_indentation{')'}

                      # only leading '&&', '||', and ':' if no others seen
                      # (but note: our count made below could be wrong
                      # due to intervening comments)
                      && ( $leading_amp_count == 0
                        || $type_ibeg_2 !~ /^(:|\&\&|\|\|)$/ )

                      # but leading colons probably line up with with a
                      # previous colon or question (count could be wrong).
                      && $type_ibeg_2 ne ':'

                      # only one step in depth allowed.  this line must not
                      # begin with a ')' itself.
                      && ( $nesting_depth_to_go[$iend_1] ==
                        $nesting_depth_to_go[$iend_2] + 1 );

                    # YVES patch 2 of 2:
                    # Allow cuddled eval chains, like this:
                    #   eval {
                    #       #STUFF;
                    #       1; # return true
                    #   } or do {
                    #       #handle error
                    #   };
                    # This patch works together with a patch in
                    # setting adjusted indentation (where the closing eval
                    # brace is outdented if possible).
                    # The problem is that an 'eval' block has continuation
                    # indentation and it looks better to undo it in some
                    # cases.  If we do not use this patch we would get:
                    #   eval {
                    #       #STUFF;
                    #       1; # return true
                    #       }
                    #       or do {
                    #       #handle error
                    #     };
                    # The alternative, for uncuddled style, is to create
                    # a patch in set_adjusted_indentation which undoes
                    # the indentation of a leading line like 'or do {'.
                    # This doesn't work well with -icb through
                    if (
                           $block_type_to_go[$iend_1] eq 'eval'
                        && !$rOpts->{'line-up-parentheses'}
                        && !$rOpts->{'indent-closing-brace'}
                        && $tokens_to_go[$iend_2] eq '{'
                        && (
                            ( $type_ibeg_2 =~ /^(|\&\&|\|\|)$/ )
                            || (   $type_ibeg_2 eq 'k'
                                && $is_and_or{ $tokens_to_go[$ibeg_2] } )
                            || $is_if_unless{ $tokens_to_go[$ibeg_2] }
                        )
                      )
                    {
                        $previous_outdentable_closing_paren ||= 1;
                    }

                    next
                      unless (
                        $previous_outdentable_closing_paren

                        # handle '.' and '?' specially below
                        || ( $type_ibeg_2 =~ /^[\.\?]$/ )
                      );
                }

                # YVES
                # honor breaks at opening brace
                # Added to prevent recombining something like this:
                #  } || eval { package main;
                elsif ( $type_iend_1 eq '{' ) {
                    next if $forced_breakpoint_to_go[$iend_1];
                }

                # do not recombine lines with ending &&, ||,
                elsif ( $is_amp_amp{$type_iend_1} ) {
                    next unless $want_break_before{$type_iend_1};
                }

                # Identify and recombine a broken ?/: chain
                elsif ( $type_iend_1 eq '?' ) {

                    # Do not recombine different levels
                    next
                      if ( $levels_to_go[$ibeg_1] ne $levels_to_go[$ibeg_2] );

                    # do not recombine unless next line ends in :
                    next unless $type_iend_2 eq ':';
                }

                # for lines ending in a comma...
                elsif ( $type_iend_1 eq ',' ) {

                    # Do not recombine at comma which is following the
                    # input bias.
                    # TODO: might be best to make a special flag
                    next if ( $old_breakpoint_to_go[$iend_1] );

                 # an isolated '},' may join with an identifier + ';'
                 # this is useful for the class of a 'bless' statement (bless.t)
                    if (   $type_ibeg_1 eq '}'
                        && $type_ibeg_2 eq 'i' )
                    {
                        next
                          unless ( ( $ibeg_1 == ( $iend_1 - 1 ) )
                            && ( $iend_2 == ( $ibeg_2 + 1 ) )
                            && $this_line_is_semicolon_terminated );

                        # override breakpoint
                        $forced_breakpoint_to_go[$iend_1] = 0;
                    }

                    # but otherwise ..
                    else {

                        # do not recombine after a comma unless this will leave
                        # just 1 more line
                        next unless ( $n + 1 >= $nmax );

                    # do not recombine if there is a change in indentation depth
                        next
                          if (
                            $levels_to_go[$iend_1] != $levels_to_go[$iend_2] );

                        # do not recombine a "complex expression" after a
                        # comma.  "complex" means no parens.
                        my $saw_paren;
                        foreach my $ii ( $ibeg_2 .. $iend_2 ) {
                            if ( $tokens_to_go[$ii] eq '(' ) {
                                $saw_paren = 1;
                                last;
                            }
                        }
                        next if $saw_paren;
                    }
                }

                # opening paren..
                elsif ( $type_iend_1 eq '(' ) {

                    # No longer doing this
                }

                elsif ( $type_iend_1 eq ')' ) {

                    # No longer doing this
                }

                # keep a terminal for-semicolon
                elsif ( $type_iend_1 eq 'f' ) {
                    next;
                }

                # if '=' at end of line ...
                elsif ( $is_assignment{$type_iend_1} ) {

                    # keep break after = if it was in input stream
                    # this helps prevent 'blinkers'
                    next if $old_breakpoint_to_go[$iend_1]

                      # don't strand an isolated '='
                      && $iend_1 != $ibeg_1;

                    my $is_short_quote =
                      (      $type_ibeg_2 eq 'Q'
                          && $ibeg_2 == $iend_2
                          && token_sequence_length( $ibeg_2, $ibeg_2 ) <
                          $rOpts_short_concatenation_item_length );
                    my $is_ternary =
                      ( $type_ibeg_1 eq '?'
                          && ( $ibeg_3 >= 0 && $types_to_go[$ibeg_3] eq ':' ) );

                    # always join an isolated '=', a short quote, or if this
                    # will put ?/: at start of adjacent lines
                    if (   $ibeg_1 != $iend_1
                        && !$is_short_quote
                        && !$is_ternary )
                    {
                        next
                          unless (
                            (

                                # unless we can reduce this to two lines
                                $nmax < $n + 2

                             # or three lines, the last with a leading semicolon
                                || (   $nmax == $n + 2
                                    && $types_to_go[$ibeg_nmax] eq ';' )

                                # or the next line ends with a here doc
                                || $type_iend_2 eq 'h'

                               # or the next line ends in an open paren or brace
                               # and the break hasn't been forced [dima.t]
                                || (  !$forced_breakpoint_to_go[$iend_1]
                                    && $type_iend_2 eq '{' )
                            )

                            # do not recombine if the two lines might align well
                            # this is a very approximate test for this
                            && (   $ibeg_3 >= 0
                                && $type_ibeg_2 ne $types_to_go[$ibeg_3] )
                          );

                        if (

                            # Recombine if we can make two lines
                            $nmax >= $n + 2

                            # -lp users often prefer this:
                            #  my $title = function($env, $env, $sysarea,
                            #                       "bubba Borrower Entry");
                            #  so we will recombine if -lp is used we have
                            #  ending comma
                            && (  !$rOpts_line_up_parentheses
                                || $type_iend_2 ne ',' )
                          )
                        {

                           # otherwise, scan the rhs line up to last token for
                           # complexity.  Note that we are not counting the last
                           # token in case it is an opening paren.
                            my $tv    = 0;
                            my $depth = $nesting_depth_to_go[$ibeg_2];
                            for ( my $i = $ibeg_2 + 1 ; $i < $iend_2 ; $i++ ) {
                                if ( $nesting_depth_to_go[$i] != $depth ) {
                                    $tv++;
                                    last if ( $tv > 1 );
                                }
                                $depth = $nesting_depth_to_go[$i];
                            }

                         # ok to recombine if no level changes before last token
                            if ( $tv > 0 ) {

                                # otherwise, do not recombine if more than two
                                # level changes.
                                next if ( $tv > 1 );

                              # check total complexity of the two adjacent lines
                              # that will occur if we do this join
                                my $istop =
                                  ( $n < $nmax ) ? $$ri_end[ $n + 1 ] : $iend_2;
                                for ( my $i = $iend_2 ; $i <= $istop ; $i++ ) {
                                    if ( $nesting_depth_to_go[$i] != $depth ) {
                                        $tv++;
                                        last if ( $tv > 2 );
                                    }
                                    $depth = $nesting_depth_to_go[$i];
                                }

                        # do not recombine if total is more than 2 level changes
                                next if ( $tv > 2 );
                            }
                        }
                    }

                    unless ( $tokens_to_go[$ibeg_2] =~ /^[\{\(\[]$/ ) {
                        $forced_breakpoint_to_go[$iend_1] = 0;
                    }
                }

                # for keywords..
                elsif ( $type_iend_1 eq 'k' ) {

                    # make major control keywords stand out
                    # (recombine.t)
                    next
                      if (

                        #/^(last|next|redo|return)$/
                        $is_last_next_redo_return{ $tokens_to_go[$iend_1] }

                        # but only if followed by multiple lines
                        && $n < $nmax
                      );

                    if ( $is_and_or{ $tokens_to_go[$iend_1] } ) {
                        next
                          unless $want_break_before{ $tokens_to_go[$iend_1] };
                    }
                }

                #----------------------------------------------------------
                # Recombine Section 3:
                # Examine token at $ibeg_2 (left end of second line of pair)
                #----------------------------------------------------------

                # join lines identified above as capable of
                # causing an outdented line with leading closing paren
                # Note that we are skipping the rest of this section
                if ($previous_outdentable_closing_paren) {
                    $forced_breakpoint_to_go[$iend_1] = 0;
                }

                # handle lines with leading &&, ||
                elsif ( $is_amp_amp{$type_ibeg_2} ) {

                    $leading_amp_count++;

                    # ok to recombine if it follows a ? or :
                    # and is followed by an open paren..
                    my $ok =
                      (      $is_ternary{$type_ibeg_1}
                          && $tokens_to_go[$iend_2] eq '(' )

                    # or is followed by a ? or : at same depth
                    #
                    # We are looking for something like this. We can
                    # recombine the && line with the line above to make the
                    # structure more clear:
                    #  return
                    #    exists $G->{Attr}->{V}
                    #    && exists $G->{Attr}->{V}->{$u}
                    #    ? %{ $G->{Attr}->{V}->{$u} }
                    #    : ();
                    #
                    # We should probably leave something like this alone:
                    #  return
                    #       exists $G->{Attr}->{E}
                    #    && exists $G->{Attr}->{E}->{$u}
                    #    && exists $G->{Attr}->{E}->{$u}->{$v}
                    #    ? %{ $G->{Attr}->{E}->{$u}->{$v} }
                    #    : ();
                    # so that we either have all of the &&'s (or ||'s)
                    # on one line, as in the first example, or break at
                    # each one as in the second example.  However, it
                    # sometimes makes things worse to check for this because
                    # it prevents multiple recombinations.  So this is not done.
                      || ( $ibeg_3 >= 0
                        && $is_ternary{ $types_to_go[$ibeg_3] }
                        && $nesting_depth_to_go[$ibeg_3] ==
                        $nesting_depth_to_go[$ibeg_2] );

                    next if !$ok && $want_break_before{$type_ibeg_2};
                    $forced_breakpoint_to_go[$iend_1] = 0;

                    # tweak the bond strength to give this joint priority
                    # over ? and :
                    $bs_tweak = 0.25;
                }

                # Identify and recombine a broken ?/: chain
                elsif ( $type_ibeg_2 eq '?' ) {

                    # Do not recombine different levels
                    my $lev = $levels_to_go[$ibeg_2];
                    next if ( $lev ne $levels_to_go[$ibeg_1] );

                    # Do not recombine a '?' if either next line or
                    # previous line does not start with a ':'.  The reasons
                    # are that (1) no alignment of the ? will be possible
                    # and (2) the expression is somewhat complex, so the
                    # '?' is harder to see in the interior of the line.
                    my $follows_colon = $ibeg_1 >= 0 && $type_ibeg_1 eq ':';
                    my $precedes_colon =
                      $ibeg_3 >= 0 && $types_to_go[$ibeg_3] eq ':';
                    next unless ( $follows_colon || $precedes_colon );

                    # we will always combining a ? line following a : line
                    if ( !$follows_colon ) {

                        # ...otherwise recombine only if it looks like a chain.
                        # we will just look at a few nearby lines to see if
                        # this looks like a chain.
                        my $local_count = 0;
                        foreach my $ii ( $ibeg_0, $ibeg_1, $ibeg_3, $ibeg_4 ) {
                            $local_count++
                              if $ii >= 0
                              && $types_to_go[$ii] eq ':'
                              && $levels_to_go[$ii] == $lev;
                        }
                        next unless ( $local_count > 1 );
                    }
                    $forced_breakpoint_to_go[$iend_1] = 0;
                }

                # do not recombine lines with leading '.'
                elsif ( $type_ibeg_2 eq '.' ) {
                    my $i_next_nonblank = min( $inext_to_go[$ibeg_2], $iend_2 );
                    next
                      unless (

                   # ... unless there is just one and we can reduce
                   # this to two lines if we do.  For example, this
                   #
                   #
                   #  $bodyA .=
                   #    '($dummy, $pat) = &get_next_tex_cmd;' . '$args .= $pat;'
                   #
                   #  looks better than this:
                   #  $bodyA .= '($dummy, $pat) = &get_next_tex_cmd;'
                   #    . '$args .= $pat;'

                        (
                               $n == 2
                            && $n == $nmax
                            && $type_ibeg_1 ne $type_ibeg_2
                        )

                        #  ... or this would strand a short quote , like this
                        #                . "some long qoute"
                        #                . "\n";

                        || (   $types_to_go[$i_next_nonblank] eq 'Q'
                            && $i_next_nonblank >= $iend_2 - 1
                            && $token_lengths_to_go[$i_next_nonblank] <
                            $rOpts_short_concatenation_item_length )
                      );
                }

                # handle leading keyword..
                elsif ( $type_ibeg_2 eq 'k' ) {

                    # handle leading "or"
                    if ( $tokens_to_go[$ibeg_2] eq 'or' ) {
                        next
                          unless (
                            $this_line_is_semicolon_terminated
                            && (

                                # following 'if' or 'unless' or 'or'
                                $type_ibeg_1 eq 'k'
                                && $is_if_unless{ $tokens_to_go[$ibeg_1] }

                                # important: only combine a very simple or
                                # statement because the step below may have
                                # combined a trailing 'and' with this or,
                                # and we do not want to then combine
                                # everything together
                                && ( $iend_2 - $ibeg_2 <= 7 )
                            )
                          );
                    }

                    # handle leading 'and'
                    elsif ( $tokens_to_go[$ibeg_2] eq 'and' ) {

                        # Decide if we will combine a single terminal 'and'
                        # after an 'if' or 'unless'.

                        #     This looks best with the 'and' on the same
                        #     line as the 'if':
                        #
                        #         $a = 1
                        #           if $seconds and $nu < 2;
                        #
                        #     But this looks better as shown:
                        #
                        #         $a = 1
                        #           if !$this->{Parents}{$_}
                        #           or $this->{Parents}{$_} eq $_;
                        #
                        next
                          unless (
                            $this_line_is_semicolon_terminated
                            && (

                                # following 'if' or 'unless' or 'or'
                                $type_ibeg_1 eq 'k'
                                && (   $is_if_unless{ $tokens_to_go[$ibeg_1] }
                                    || $tokens_to_go[$ibeg_1] eq 'or' )
                            )
                          );
                    }

                    # handle leading "if" and "unless"
                    elsif ( $is_if_unless{ $tokens_to_go[$ibeg_2] } ) {

                      # FIXME: This is still experimental..may not be too useful
                        next
                          unless (
                            $this_line_is_semicolon_terminated

                            #  previous line begins with 'and' or 'or'
                            && $type_ibeg_1 eq 'k'
                            && $is_and_or{ $tokens_to_go[$ibeg_1] }

                          );
                    }

                    # handle all other leading keywords
                    else {

                        # keywords look best at start of lines,
                        # but combine things like "1 while"
                        unless ( $is_assignment{$type_iend_1} ) {
                            next
                              if ( ( $type_iend_1 ne 'k' )
                                && ( $tokens_to_go[$ibeg_2] ne 'while' ) );
                        }
                    }
                }

                # similar treatment of && and || as above for 'and' and 'or':
                # NOTE: This block of code is currently bypassed because
                # of a previous block but is retained for possible future use.
                elsif ( $is_amp_amp{$type_ibeg_2} ) {

                    # maybe looking at something like:
                    # unless $TEXTONLY || $item =~ m%</?(hr>|p>|a|img)%i;

                    next
                      unless (
                        $this_line_is_semicolon_terminated

                        # previous line begins with an 'if' or 'unless' keyword
                        && $type_ibeg_1 eq 'k'
                        && $is_if_unless{ $tokens_to_go[$ibeg_1] }

                      );
                }

                # handle line with leading = or similar
                elsif ( $is_assignment{$type_ibeg_2} ) {
                    next unless ( $n == 1 || $n == $nmax );
                    next if $old_breakpoint_to_go[$iend_1];
                    next
                      unless (

                        # unless we can reduce this to two lines
                        $nmax == 2

                        # or three lines, the last with a leading semicolon
                        || ( $nmax == 3 && $types_to_go[$ibeg_nmax] eq ';' )

                        # or the next line ends with a here doc
                        || $type_iend_2 eq 'h'

                        # or this is a short line ending in ;
                        || ( $n == $nmax && $this_line_is_semicolon_terminated )
                      );
                    $forced_breakpoint_to_go[$iend_1] = 0;
                }

                #----------------------------------------------------------
                # Recombine Section 4:
                # Combine the lines if we arrive here and it is possible
                #----------------------------------------------------------

                # honor hard breakpoints
                next if ( $forced_breakpoint_to_go[$iend_1] > 0 );

                my $bs = $bond_strength_to_go[$iend_1] + $bs_tweak;

                # combined line cannot be too long
                my $excess = excess_line_length( $ibeg_1, $iend_2 );
                next if ( $excess > 0 );

                # Require a few extra spaces before recombining lines if we are
                # at an old breakpoint unless this is a simple list or terminal
                # line.  The goal is to avoid oscillating between two
                # quasi-stable end states.  For example this snippet caused
                # problems:
##    my $this =
##    bless {
##        TText => "[" . ( join ',', map { "\"$_\"" } split "\n", $_ ) . "]"
##      },
##      $type;
                next
                  if ( $old_breakpoint_to_go[$iend_1]
                    && !$this_line_is_semicolon_terminated
                    && $n < $nmax
                    && $excess + 4 > 0
                    && $type_iend_2 ne ',' );

                # do not recombine if we would skip in indentation levels
                if ( $n < $nmax ) {
                    my $if_next = $$ri_beg[ $n + 1 ];
                    next
                      if (
                           $levels_to_go[$ibeg_1] < $levels_to_go[$ibeg_2]
                        && $levels_to_go[$ibeg_2] < $levels_to_go[$if_next]

                        # but an isolated 'if (' is undesirable
                        && !(
                               $n == 1
                            && $iend_1 - $ibeg_1 <= 2
                            && $type_ibeg_1 eq 'k'
                            && $tokens_to_go[$ibeg_1] eq 'if'
                            && $tokens_to_go[$iend_1] ne '('
                        )
                      );
                }

                # honor no-break's
                next if ( $bs >= NO_BREAK - 1 );

                # remember the pair with the greatest bond strength
                if ( !$n_best ) {
                    $n_best  = $n;
                    $bs_best = $bs;
                }
                else {

                    if ( $bs > $bs_best ) {
                        $n_best  = $n;
                        $bs_best = $bs;
                    }
                }
            }

            # recombine the pair with the greatest bond strength
            if ($n_best) {
                splice @$ri_beg, $n_best, 1;
                splice @$ri_end, $n_best - 1, 1;
                splice @joint, $n_best, 1;

                # keep going if we are still making progress
                $more_to_do++;
            }
        }
        return ( $ri_beg, $ri_end );
    }
}    # end recombine_breakpoints

sub break_all_chain_tokens {

    # scan the current breakpoints looking for breaks at certain "chain
    # operators" (. : && || + etc) which often occur repeatedly in a long
    # statement.  If we see a break at any one, break at all similar tokens
    # within the same container.
    #
    my ( $ri_left, $ri_right ) = @_;

    my %saw_chain_type;
    my %left_chain_type;
    my %right_chain_type;
    my %interior_chain_type;
    my $nmax = @$ri_right - 1;

    # scan the left and right end tokens of all lines
    my $count = 0;
    for my $n ( 0 .. $nmax ) {
        my $il    = $$ri_left[$n];
        my $ir    = $$ri_right[$n];
        my $typel = $types_to_go[$il];
        my $typer = $types_to_go[$ir];
        $typel = '+' if ( $typel eq '-' );    # treat + and - the same
        $typer = '+' if ( $typer eq '-' );
        $typel = '*' if ( $typel eq '/' );    # treat * and / the same
        $typer = '*' if ( $typer eq '/' );
        my $tokenl = $tokens_to_go[$il];
        my $tokenr = $tokens_to_go[$ir];

        if ( $is_chain_operator{$tokenl} && $want_break_before{$typel} ) {
            next if ( $typel eq '?' );
            push @{ $left_chain_type{$typel} }, $il;
            $saw_chain_type{$typel} = 1;
            $count++;
        }
        if ( $is_chain_operator{$tokenr} && !$want_break_before{$typer} ) {
            next if ( $typer eq '?' );
            push @{ $right_chain_type{$typer} }, $ir;
            $saw_chain_type{$typer} = 1;
            $count++;
        }
    }
    return unless $count;

    # now look for any interior tokens of the same types
    $count = 0;
    for my $n ( 0 .. $nmax ) {
        my $il = $$ri_left[$n];
        my $ir = $$ri_right[$n];
        for ( my $i = $il + 1 ; $i < $ir ; $i++ ) {
            my $type = $types_to_go[$i];
            $type = '+' if ( $type eq '-' );
            $type = '*' if ( $type eq '/' );
            if ( $saw_chain_type{$type} ) {
                push @{ $interior_chain_type{$type} }, $i;
                $count++;
            }
        }
    }
    return unless $count;

    # now make a list of all new break points
    my @insert_list;

    # loop over all chain types
    foreach my $type ( keys %saw_chain_type ) {

        # quit if just ONE continuation line with leading .  For example--
        # print LATEXFILE '\framebox{\parbox[c][' . $h . '][t]{' . $w . '}{'
        #  . $contents;
        last if ( $nmax == 1 && $type =~ /^[\.\+]$/ );

        # loop over all interior chain tokens
        foreach my $itest ( @{ $interior_chain_type{$type} } ) {

            # loop over all left end tokens of same type
            if ( $left_chain_type{$type} ) {
                next if $nobreak_to_go[ $itest - 1 ];
                foreach my $i ( @{ $left_chain_type{$type} } ) {
                    next unless in_same_container( $i, $itest );
                    push @insert_list, $itest - 1;

                    # Break at matching ? if this : is at a different level.
                    # For example, the ? before $THRf_DEAD in the following
                    # should get a break if its : gets a break.
                    #
                    # my $flags =
                    #     ( $_ & 1 ) ? ( $_ & 4 ) ? $THRf_DEAD : $THRf_ZOMBIE
                    #   : ( $_ & 4 ) ? $THRf_R_DETACHED
                    #   :              $THRf_R_JOINABLE;
                    if (   $type eq ':'
                        && $levels_to_go[$i] != $levels_to_go[$itest] )
                    {
                        my $i_question = $mate_index_to_go[$itest];
                        if ( $i_question > 0 ) {
                            push @insert_list, $i_question - 1;
                        }
                    }
                    last;
                }
            }

            # loop over all right end tokens of same type
            if ( $right_chain_type{$type} ) {
                next if $nobreak_to_go[$itest];
                foreach my $i ( @{ $right_chain_type{$type} } ) {
                    next unless in_same_container( $i, $itest );
                    push @insert_list, $itest;

                    # break at matching ? if this : is at a different level
                    if (   $type eq ':'
                        && $levels_to_go[$i] != $levels_to_go[$itest] )
                    {
                        my $i_question = $mate_index_to_go[$itest];
                        if ( $i_question >= 0 ) {
                            push @insert_list, $i_question;
                        }
                    }
                    last;
                }
            }
        }
    }

    # insert any new break points
    if (@insert_list) {
        insert_additional_breaks( \@insert_list, $ri_left, $ri_right );
    }
}

sub break_equals {

    # Look for assignment operators that could use a breakpoint.
    # For example, in the following snippet
    #
    #    $HOME = $ENV{HOME}
    #      || $ENV{LOGDIR}
    #      || $pw[7]
    #      || die "no home directory for user $<";
    #
    # we could break at the = to get this, which is a little nicer:
    #    $HOME =
    #         $ENV{HOME}
    #      || $ENV{LOGDIR}
    #      || $pw[7]
    #      || die "no home directory for user $<";
    #
    # The logic here follows the logic in set_logical_padding, which
    # will add the padding in the second line to improve alignment.
    #
    my ( $ri_left, $ri_right ) = @_;
    my $nmax = @$ri_right - 1;
    return unless ( $nmax >= 2 );

    # scan the left ends of first two lines
    my $tokbeg = "";
    my $depth_beg;
    for my $n ( 1 .. 2 ) {
        my $il     = $$ri_left[$n];
        my $typel  = $types_to_go[$il];
        my $tokenl = $tokens_to_go[$il];

        my $has_leading_op = ( $tokenl =~ /^\w/ )
          ? $is_chain_operator{$tokenl}    # + - * / : ? && ||
          : $is_chain_operator{$typel};    # and, or
        return unless ($has_leading_op);
        if ( $n > 1 ) {
            return
              unless ( $tokenl eq $tokbeg
                && $nesting_depth_to_go[$il] eq $depth_beg );
        }
        $tokbeg    = $tokenl;
        $depth_beg = $nesting_depth_to_go[$il];
    }

    # now look for any interior tokens of the same types
    my $il = $$ri_left[0];
    my $ir = $$ri_right[0];

    # now make a list of all new break points
    my @insert_list;
    for ( my $i = $ir - 1 ; $i > $il ; $i-- ) {
        my $type = $types_to_go[$i];
        if (   $is_assignment{$type}
            && $nesting_depth_to_go[$i] eq $depth_beg )
        {
            if ( $want_break_before{$type} ) {
                push @insert_list, $i - 1;
            }
            else {
                push @insert_list, $i;
            }
        }
    }

    # Break after a 'return' followed by a chain of operators
    #  return ( $^O !~ /win32|dos/i )
    #    && ( $^O ne 'VMS' )
    #    && ( $^O ne 'OS2' )
    #    && ( $^O ne 'MacOS' );
    # To give:
    #  return
    #       ( $^O !~ /win32|dos/i )
    #    && ( $^O ne 'VMS' )
    #    && ( $^O ne 'OS2' )
    #    && ( $^O ne 'MacOS' );
    my $i = 0;
    if (   $types_to_go[$i] eq 'k'
        && $tokens_to_go[$i] eq 'return'
        && $ir > $il
        && $nesting_depth_to_go[$i] eq $depth_beg )
    {
        push @insert_list, $i;
    }

    return unless (@insert_list);

    # One final check...
    # scan second and third lines and be sure there are no assignments
    # we want to avoid breaking at an = to make something like this:
    #    unless ( $icon =
    #           $html_icons{"$type-$state"}
    #        or $icon = $html_icons{$type}
    #        or $icon = $html_icons{$state} )
    for my $n ( 1 .. 2 ) {
        my $il = $$ri_left[$n];
        my $ir = $$ri_right[$n];
        for ( my $i = $il + 1 ; $i <= $ir ; $i++ ) {
            my $type = $types_to_go[$i];
            return
              if ( $is_assignment{$type}
                && $nesting_depth_to_go[$i] eq $depth_beg );
        }
    }

    # ok, insert any new break point
    if (@insert_list) {
        insert_additional_breaks( \@insert_list, $ri_left, $ri_right );
    }
}

sub insert_final_breaks {

    my ( $ri_left, $ri_right ) = @_;

    my $nmax = @$ri_right - 1;

    # scan the left and right end tokens of all lines
    my $count         = 0;
    my $i_first_colon = -1;
    for my $n ( 0 .. $nmax ) {
        my $il    = $$ri_left[$n];
        my $ir    = $$ri_right[$n];
        my $typel = $types_to_go[$il];
        my $typer = $types_to_go[$ir];
        return if ( $typel eq '?' );
        return if ( $typer eq '?' );
        if    ( $typel eq ':' ) { $i_first_colon = $il; last; }
        elsif ( $typer eq ':' ) { $i_first_colon = $ir; last; }
    }

    # For long ternary chains,
    # if the first : we see has its # ? is in the interior
    # of a preceding line, then see if there are any good
    # breakpoints before the ?.
    if ( $i_first_colon > 0 ) {
        my $i_question = $mate_index_to_go[$i_first_colon];
        if ( $i_question > 0 ) {
            my @insert_list;
            for ( my $ii = $i_question - 1 ; $ii >= 0 ; $ii -= 1 ) {
                my $token = $tokens_to_go[$ii];
                my $type  = $types_to_go[$ii];

                # For now, a good break is either a comma or a 'return'.
                if ( ( $type eq ',' || $type eq 'k' && $token eq 'return' )
                    && in_same_container( $ii, $i_question ) )
                {
                    push @insert_list, $ii;
                    last;
                }
            }

            # insert any new break points
            if (@insert_list) {
                insert_additional_breaks( \@insert_list, $ri_left, $ri_right );
            }
        }
    }
}

sub in_same_container {

    # check to see if tokens at i1 and i2 are in the
    # same container, and not separated by a comma, ? or :
    my ( $i1, $i2 ) = @_;
    my $type  = $types_to_go[$i1];
    my $depth = $nesting_depth_to_go[$i1];
    return unless ( $nesting_depth_to_go[$i2] == $depth );
    if ( $i2 < $i1 ) { ( $i1, $i2 ) = ( $i2, $i1 ) }

    ###########################################################
    # This is potentially a very slow routine and not critical.
    # For safety just give up for large differences.
    # See test file 'infinite_loop.txt'
    # TODO: replace this loop with a data structure
    ###########################################################
    return if ( $i2 - $i1 > 200 );

    for ( my $i = $i1 + 1 ; $i < $i2 ; $i++ ) {
        next   if ( $nesting_depth_to_go[$i] > $depth );
        return if ( $nesting_depth_to_go[$i] < $depth );

        my $tok = $tokens_to_go[$i];
        $tok = ',' if $tok eq '=>';    # treat => same as ,

        # Example: we would not want to break at any of these .'s
        #  : "<A HREF=\"#item_" . htmlify( 0, $s2 ) . "\">$str</A>"
        if ( $type ne ':' ) {
            return if ( $tok =~ /^[\,\:\?]$/ ) || $tok eq '||' || $tok eq 'or';
        }
        else {
            return if ( $tok =~ /^[\,]$/ );
        }
    }
    return 1;
}

sub set_continuation_breaks {

    # Define an array of indexes for inserting newline characters to
    # keep the line lengths below the maximum desired length.  There is
    # an implied break after the last token, so it need not be included.

    # Method:
    # This routine is part of series of routines which adjust line
    # lengths.  It is only called if a statement is longer than the
    # maximum line length, or if a preliminary scanning located
    # desirable break points.   Sub scan_list has already looked at
    # these tokens and set breakpoints (in array
    # $forced_breakpoint_to_go[$i]) where it wants breaks (for example
    # after commas, after opening parens, and before closing parens).
    # This routine will honor these breakpoints and also add additional
    # breakpoints as necessary to keep the line length below the maximum
    # requested.  It bases its decision on where the 'bond strength' is
    # lowest.

    # Output: returns references to the arrays:
    #  @i_first
    #  @i_last
    # which contain the indexes $i of the first and last tokens on each
    # line.

    # In addition, the array:
    #   $forced_breakpoint_to_go[$i]
    # may be updated to be =1 for any index $i after which there must be
    # a break.  This signals later routines not to undo the breakpoint.

    my $saw_good_break = shift;
    my @i_first        = ();      # the first index to output
    my @i_last         = ();      # the last index to output
    my @i_colon_breaks = ();      # needed to decide if we have to break at ?'s
    if ( $types_to_go[0] eq ':' ) { push @i_colon_breaks, 0 }

    set_bond_strengths();

    my $imin = 0;
    my $imax = $max_index_to_go;
    if ( $types_to_go[$imin] eq 'b' ) { $imin++ }
    if ( $types_to_go[$imax] eq 'b' ) { $imax-- }
    my $i_begin = $imin;          # index for starting next iteration

    my $leading_spaces          = leading_spaces_to_go($imin);
    my $line_count              = 0;
    my $last_break_strength     = NO_BREAK;
    my $i_last_break            = -1;
    my $max_bias                = 0.001;
    my $tiny_bias               = 0.0001;
    my $leading_alignment_token = "";
    my $leading_alignment_type  = "";

    # see if any ?/:'s are in order
    my $colons_in_order = 1;
    my $last_tok        = "";
    my @colon_list  = grep /^[\?\:]$/, @types_to_go[ 0 .. $max_index_to_go ];
    my $colon_count = @colon_list;
    foreach (@colon_list) {
        if ( $_ eq $last_tok ) { $colons_in_order = 0; last }
        $last_tok = $_;
    }

    # This is a sufficient but not necessary condition for colon chain
    my $is_colon_chain = ( $colons_in_order && @colon_list > 2 );

    #-------------------------------------------------------
    # BEGINNING of main loop to set continuation breakpoints
    # Keep iterating until we reach the end
    #-------------------------------------------------------
    while ( $i_begin <= $imax ) {
        my $lowest_strength        = NO_BREAK;
        my $starting_sum           = $summed_lengths_to_go[$i_begin];
        my $i_lowest               = -1;
        my $i_test                 = -1;
        my $lowest_next_token      = '';
        my $lowest_next_type       = 'b';
        my $i_lowest_next_nonblank = -1;

        #-------------------------------------------------------
        # BEGINNING of inner loop to find the best next breakpoint
        #-------------------------------------------------------
        for ( $i_test = $i_begin ; $i_test <= $imax ; $i_test++ ) {
            my $type                     = $types_to_go[$i_test];
            my $token                    = $tokens_to_go[$i_test];
            my $next_type                = $types_to_go[ $i_test + 1 ];
            my $next_token               = $tokens_to_go[ $i_test + 1 ];
            my $i_next_nonblank          = $inext_to_go[$i_test];
            my $next_nonblank_type       = $types_to_go[$i_next_nonblank];
            my $next_nonblank_token      = $tokens_to_go[$i_next_nonblank];
            my $next_nonblank_block_type = $block_type_to_go[$i_next_nonblank];
            my $strength                 = $bond_strength_to_go[$i_test];
            my $maximum_line_length      = maximum_line_length($i_begin);

            # use old breaks as a tie-breaker.  For example to
            # prevent blinkers with -pbp in this code:

##@keywords{
##    qw/ARG OUTPUT PROTO CONSTRUCTOR RETURNS DESC PARAMS SEEALSO EXAMPLE/}
##    = ();

            # At the same time try to prevent a leading * in this code
            # with the default formatting:
            #
##                return
##                    factorial( $a + $b - 1 ) / factorial( $a - 1 ) / factorial( $b - 1 )
##                  * ( $x**( $a - 1 ) )
##                  * ( ( 1 - $x )**( $b - 1 ) );

            # reduce strength a bit to break ties at an old breakpoint ...
            if (
                $old_breakpoint_to_go[$i_test]

                # which is a 'good' breakpoint, meaning ...
                # we don't want to break before it
                && !$want_break_before{$type}

                # and either we want to break before the next token
                # or the next token is not short (i.e. not a '*', '/' etc.)
                && $i_next_nonblank <= $imax
                && (   $want_break_before{$next_nonblank_type}
                    || $token_lengths_to_go[$i_next_nonblank] > 2
                    || $next_nonblank_type =~ /^[\,\(\[\{L]$/ )
              )
            {
                $strength -= $tiny_bias;
            }

            # otherwise increase strength a bit if this token would be at the
            # maximum line length.  This is necessary to avoid blinking
            # in the above example when the -iob flag is added.
            else {
                my $len =
                  $leading_spaces +
                  $summed_lengths_to_go[ $i_test + 1 ] -
                  $starting_sum;
                if ( $len >= $maximum_line_length ) {
                    $strength += $tiny_bias;
                }
            }

            my $must_break = 0;

            # Force an immediate break at certain operators
            # with lower level than the start of the line,
            # unless we've already seen a better break.
            #
            ##############################################
            # Note on an issue with a preceding ?
            ##############################################
            # We don't include a ? in the above list, but there may
            # be a break at a previous ? if the line is long.
            # Because of this we do not want to force a break if
            # there is a previous ? on this line.  For now the best way
            # to do this is to not break if we have seen a lower strength
            # point, which is probably a ?.
            #
            # Example of unwanted breaks we are avoiding at a '.' following a ?
            # from pod2html using perltidy -gnu:
            # )
            # ? "\n&lt;A NAME=\""
            # . $value
            # . "\"&gt;\n$text&lt;/A&gt;\n"
            # : "\n$type$pod2.html\#" . $value . "\"&gt;$text&lt;\/A&gt;\n";
            if (
                (
                    $next_nonblank_type =~ /^(\.|\&\&|\|\|)$/
                    || (   $next_nonblank_type eq 'k'
                        && $next_nonblank_token =~ /^(and|or)$/ )
                )
                && ( $nesting_depth_to_go[$i_begin] >
                    $nesting_depth_to_go[$i_next_nonblank] )
                && ( $strength <= $lowest_strength )
              )
            {
                set_forced_breakpoint($i_next_nonblank);
            }

            if (

                # Try to put a break where requested by scan_list
                $forced_breakpoint_to_go[$i_test]

                # break between ) { in a continued line so that the '{' can
                # be outdented
                # See similar logic in scan_list which catches instances
                # where a line is just something like ') {'.  We have to
                # be careful because the corresponding block keyword might
                # not be on the first line, such as 'for' here:
                #
                # eval {
                #     for ("a") {
                #         for $x ( 1, 2 ) { local $_ = "b"; s/(.*)/+$1/ }
                #     }
                # };
                #
                || (   $line_count
                    && ( $token eq ')' )
                    && ( $next_nonblank_type eq '{' )
                    && ($next_nonblank_block_type)
                    && ( $next_nonblank_block_type ne $tokens_to_go[$i_begin] )
                    && !$rOpts->{'opening-brace-always-on-right'} )

                # There is an implied forced break at a terminal opening brace
                || ( ( $type eq '{' ) && ( $i_test == $imax ) )
              )
            {

                # Forced breakpoints must sometimes be overridden, for example
                # because of a side comment causing a NO_BREAK.  It is easier
                # to catch this here than when they are set.
                if ( $strength < NO_BREAK - 1 ) {
                    $strength   = $lowest_strength - $tiny_bias;
                    $must_break = 1;
                }
            }

            # quit if a break here would put a good terminal token on
            # the next line and we already have a possible break
            if (
                   !$must_break
                && ( $next_nonblank_type =~ /^[\;\,]$/ )
                && (
                    (
                        $leading_spaces +
                        $summed_lengths_to_go[ $i_next_nonblank + 1 ] -
                        $starting_sum
                    ) > $maximum_line_length
                )
              )
            {
                last if ( $i_lowest >= 0 );
            }

            # Avoid a break which would strand a single punctuation
            # token.  For example, we do not want to strand a leading
            # '.' which is followed by a long quoted string.
            # But note that we do want to do this with -extrude (l=1)
            # so please test any changes to this code on -extrude.
            if (
                   !$must_break
                && ( $i_test == $i_begin )
                && ( $i_test < $imax )
                && ( $token eq $type )
                && (
                    (
                        $leading_spaces +
                        $summed_lengths_to_go[ $i_test + 1 ] -
                        $starting_sum
                    ) < $maximum_line_length
                )
              )
            {
                $i_test = min( $imax, $inext_to_go[$i_test] );
                redo;
            }

            if ( ( $strength <= $lowest_strength ) && ( $strength < NO_BREAK ) )
            {

                # break at previous best break if it would have produced
                # a leading alignment of certain common tokens, and it
                # is different from the latest candidate break
                last
                  if ($leading_alignment_type);

                # Force at least one breakpoint if old code had good
                # break It is only called if a breakpoint is required or
                # desired.  This will probably need some adjustments
                # over time.  A goal is to try to be sure that, if a new
                # side comment is introduced into formated text, then
                # the same breakpoints will occur.  scbreak.t
                last
                  if (
                    $i_test == $imax              # we are at the end
                    && !$forced_breakpoint_count  #
                    && $saw_good_break            # old line had good break
                    && $type =~ /^[#;\{]$/        # and this line ends in
                                                  # ';' or side comment
                    && $i_last_break < 0          # and we haven't made a break
                    && $i_lowest >= 0             # and we saw a possible break
                    && $i_lowest < $imax - 1      # (but not just before this ;)
                    && $strength - $lowest_strength < 0.5 * WEAK # and it's good
                  );

                # Do not skip past an important break point in a short final
                # segment.  For example, without this check we would miss the
                # break at the final / in the following code:
                #
                #  $depth_stop =
                #    ( $tau * $mass_pellet * $q_0 *
                #        ( 1. - exp( -$t_stop / $tau ) ) -
                #        4. * $pi * $factor * $k_ice *
                #        ( $t_melt - $t_ice ) *
                #        $r_pellet *
                #        $t_stop ) /
                #    ( $rho_ice * $Qs * $pi * $r_pellet**2 );
                #
                if (   $line_count > 2
                    && $i_lowest < $i_test
                    && $i_test > $imax - 2
                    && $nesting_depth_to_go[$i_begin] >
                    $nesting_depth_to_go[$i_lowest]
                    && $lowest_strength < $last_break_strength - .5 * WEAK )
                {
                    # Make this break for math operators for now
                    my $ir = $inext_to_go[$i_lowest];
                    my $il = $iprev_to_go[$ir];
                    last
                      if ( $types_to_go[$il] =~ /^[\/\*\+\-\%]$/
                        || $types_to_go[$ir] =~ /^[\/\*\+\-\%]$/ );
                }

                # Update the minimum bond strength location
                $lowest_strength        = $strength;
                $i_lowest               = $i_test;
                $lowest_next_token      = $next_nonblank_token;
                $lowest_next_type       = $next_nonblank_type;
                $i_lowest_next_nonblank = $i_next_nonblank;
                last if $must_break;

                # set flags to remember if a break here will produce a
                # leading alignment of certain common tokens
                if (   $line_count > 0
                    && $i_test < $imax
                    && ( $lowest_strength - $last_break_strength <= $max_bias )
                  )
                {
                    my $i_last_end = $iprev_to_go[$i_begin];
                    my $tok_beg    = $tokens_to_go[$i_begin];
                    my $type_beg   = $types_to_go[$i_begin];
                    if (

                        # check for leading alignment of certain tokens
                        (
                               $tok_beg eq $next_nonblank_token
                            && $is_chain_operator{$tok_beg}
                            && (   $type_beg eq 'k'
                                || $type_beg eq $tok_beg )
                            && $nesting_depth_to_go[$i_begin] >=
                            $nesting_depth_to_go[$i_next_nonblank]
                        )

                        || (   $tokens_to_go[$i_last_end] eq $token
                            && $is_chain_operator{$token}
                            && ( $type eq 'k' || $type eq $token )
                            && $nesting_depth_to_go[$i_last_end] >=
                            $nesting_depth_to_go[$i_test] )
                      )
                    {
                        $leading_alignment_token = $next_nonblank_token;
                        $leading_alignment_type  = $next_nonblank_type;
                    }
                }
            }

            my $too_long = ( $i_test >= $imax );
            if ( !$too_long ) {
                my $next_length =
                  $leading_spaces +
                  $summed_lengths_to_go[ $i_test + 2 ] -
                  $starting_sum;
                $too_long = $next_length > $maximum_line_length;

                # To prevent blinkers we will avoid leaving a token exactly at
                # the line length limit unless it is the last token or one of
                # several "good" types.
                #
                # The following code was a blinker with -pbp before this
                # modification:
##                    $last_nonblank_token eq '('
##                        && $is_indirect_object_taker{ $paren_type
##                            [$paren_depth] }
                # The issue causing the problem is that if the
                # term [$paren_depth] gets broken across a line then
                # the whitespace routine doesn't see both opening and closing
                # brackets and will format like '[ $paren_depth ]'.  This
                # leads to an oscillation in length depending if we break
                # before the closing bracket or not.
                if (  !$too_long
                    && $i_test + 1 < $imax
                    && $next_nonblank_type !~ /^[,\}\]\)R]$/ )
                {
                    $too_long = $next_length >= $maximum_line_length;
                }
            }

            FORMATTER_DEBUG_FLAG_BREAK
              && do {
                my $ltok     = $token;
                my $rtok     = $next_nonblank_token ? $next_nonblank_token : "";
                my $i_testp2 = $i_test + 2;
                if ( $i_testp2 > $max_index_to_go + 1 ) {
                    $i_testp2 = $max_index_to_go + 1;
                }
                if ( length($ltok) > 6 ) { $ltok = substr( $ltok, 0, 8 ) }
                if ( length($rtok) > 6 ) { $rtok = substr( $rtok, 0, 8 ) }
                print STDOUT
"BREAK: i=$i_test imax=$imax $types_to_go[$i_test] $next_nonblank_type sp=($leading_spaces) lnext= $summed_lengths_to_go[$i_testp2] 2long=$too_long str=$strength    $ltok $rtok\n";
              };

            # allow one extra terminal token after exceeding line length
            # if it would strand this token.
            if (   $rOpts_fuzzy_line_length
                && $too_long
                && $i_lowest == $i_test
                && $token_lengths_to_go[$i_test] > 1
                && $next_nonblank_type =~ /^[\;\,]$/ )
            {
                $too_long = 0;
            }

            last
              if (
                ( $i_test == $imax )    # we're done if no more tokens,
                || (
                    ( $i_lowest >= 0 )    # or no more space and we have a break
                    && $too_long
                )
              );
        }

        #-------------------------------------------------------
        # END of inner loop to find the best next breakpoint
        # Now decide exactly where to put the breakpoint
        #-------------------------------------------------------

        # it's always ok to break at imax if no other break was found
        if ( $i_lowest < 0 ) { $i_lowest = $imax }

        # semi-final index calculation
        my $i_next_nonblank     = $inext_to_go[$i_lowest];
        my $next_nonblank_type  = $types_to_go[$i_next_nonblank];
        my $next_nonblank_token = $tokens_to_go[$i_next_nonblank];

        #-------------------------------------------------------
        # ?/: rule 1 : if a break here will separate a '?' on this
        # line from its closing ':', then break at the '?' instead.
        #-------------------------------------------------------
        my $i;
        foreach $i ( $i_begin + 1 .. $i_lowest - 1 ) {
            next unless ( $tokens_to_go[$i] eq '?' );

            # do not break if probable sequence of ?/: statements
            next if ($is_colon_chain);

            # do not break if statement is broken by side comment
            next
              if (
                $tokens_to_go[$max_index_to_go] eq '#'
                && terminal_type( \@types_to_go, \@block_type_to_go, 0,
                    $max_index_to_go ) !~ /^[\;\}]$/
              );

            # no break needed if matching : is also on the line
            next
              if ( $mate_index_to_go[$i] >= 0
                && $mate_index_to_go[$i] <= $i_next_nonblank );

            $i_lowest = $i;
            if ( $want_break_before{'?'} ) { $i_lowest-- }
            last;
        }

        #-------------------------------------------------------
        # END of inner loop to find the best next breakpoint:
        # Break the line after the token with index i=$i_lowest
        #-------------------------------------------------------

        # final index calculation
        $i_next_nonblank     = $inext_to_go[$i_lowest];
        $next_nonblank_type  = $types_to_go[$i_next_nonblank];
        $next_nonblank_token = $tokens_to_go[$i_next_nonblank];

        FORMATTER_DEBUG_FLAG_BREAK
          && print STDOUT
          "BREAK: best is i = $i_lowest strength = $lowest_strength\n";

        #-------------------------------------------------------
        # ?/: rule 2 : if we break at a '?', then break at its ':'
        #
        # Note: this rule is also in sub scan_list to handle a break
        # at the start and end of a line (in case breaks are dictated
        # by side comments).
        #-------------------------------------------------------
        if ( $next_nonblank_type eq '?' ) {
            set_closing_breakpoint($i_next_nonblank);
        }
        elsif ( $types_to_go[$i_lowest] eq '?' ) {
            set_closing_breakpoint($i_lowest);
        }

        #-------------------------------------------------------
        # ?/: rule 3 : if we break at a ':' then we save
        # its location for further work below.  We may need to go
        # back and break at its '?'.
        #-------------------------------------------------------
        if ( $next_nonblank_type eq ':' ) {
            push @i_colon_breaks, $i_next_nonblank;
        }
        elsif ( $types_to_go[$i_lowest] eq ':' ) {
            push @i_colon_breaks, $i_lowest;
        }

        # here we should set breaks for all '?'/':' pairs which are
        # separated by this line

        $line_count++;

        # save this line segment, after trimming blanks at the ends
        push( @i_first,
            ( $types_to_go[$i_begin] eq 'b' ) ? $i_begin + 1 : $i_begin );
        push( @i_last,
            ( $types_to_go[$i_lowest] eq 'b' ) ? $i_lowest - 1 : $i_lowest );

        # set a forced breakpoint at a container opening, if necessary, to
        # signal a break at a closing container.  Excepting '(' for now.
        if ( $tokens_to_go[$i_lowest] =~ /^[\{\[]$/
            && !$forced_breakpoint_to_go[$i_lowest] )
        {
            set_closing_breakpoint($i_lowest);
        }

        # get ready to go again
        $i_begin                 = $i_lowest + 1;
        $last_break_strength     = $lowest_strength;
        $i_last_break            = $i_lowest;
        $leading_alignment_token = "";
        $leading_alignment_type  = "";
        $lowest_next_token       = '';
        $lowest_next_type        = 'b';

        if ( ( $i_begin <= $imax ) && ( $types_to_go[$i_begin] eq 'b' ) ) {
            $i_begin++;
        }

        # update indentation size
        if ( $i_begin <= $imax ) {
            $leading_spaces = leading_spaces_to_go($i_begin);
        }
    }

    #-------------------------------------------------------
    # END of main loop to set continuation breakpoints
    # Now go back and make any necessary corrections
    #-------------------------------------------------------

    #-------------------------------------------------------
    # ?/: rule 4 -- if we broke at a ':', then break at
    # corresponding '?' unless this is a chain of ?: expressions
    #-------------------------------------------------------
    if (@i_colon_breaks) {

        # using a simple method for deciding if we are in a ?/: chain --
        # this is a chain if it has multiple ?/: pairs all in order;
        # otherwise not.
        # Note that if line starts in a ':' we count that above as a break
        my $is_chain = ( $colons_in_order && @i_colon_breaks > 1 );

        unless ($is_chain) {
            my @insert_list = ();
            foreach (@i_colon_breaks) {
                my $i_question = $mate_index_to_go[$_];
                if ( $i_question >= 0 ) {
                    if ( $want_break_before{'?'} ) {
                        $i_question = $iprev_to_go[$i_question];
                    }

                    if ( $i_question >= 0 ) {
                        push @insert_list, $i_question;
                    }
                }
                insert_additional_breaks( \@insert_list, \@i_first, \@i_last );
            }
        }
    }
    return ( \@i_first, \@i_last, $colon_count );
}

sub insert_additional_breaks {

    # this routine will add line breaks at requested locations after
    # sub set_continuation_breaks has made preliminary breaks.

    my ( $ri_break_list, $ri_first, $ri_last ) = @_;
    my $i_f;
    my $i_l;
    my $line_number = 0;
    my $i_break_left;
    foreach $i_break_left ( sort { $a <=> $b } @$ri_break_list ) {

        $i_f = $$ri_first[$line_number];
        $i_l = $$ri_last[$line_number];
        while ( $i_break_left >= $i_l ) {
            $line_number++;

            # shouldn't happen unless caller passes bad indexes
            if ( $line_number >= @$ri_last ) {
                warning(
"Non-fatal program bug: couldn't set break at $i_break_left\n"
                );
                report_definite_bug();
                return;
            }
            $i_f = $$ri_first[$line_number];
            $i_l = $$ri_last[$line_number];
        }

        # Do not leave a blank at the end of a line; back up if necessary
        if ( $types_to_go[$i_break_left] eq 'b' ) { $i_break_left-- }

        my $i_break_right = $inext_to_go[$i_break_left];
        if (   $i_break_left >= $i_f
            && $i_break_left < $i_l
            && $i_break_right > $i_f
            && $i_break_right <= $i_l )
        {
            splice( @$ri_first, $line_number, 1, ( $i_f, $i_break_right ) );
            splice( @$ri_last, $line_number, 1, ( $i_break_left, $i_l ) );
        }
    }
}

sub set_closing_breakpoint {

    # set a breakpoint at a matching closing token
    # at present, this is only used to break at a ':' which matches a '?'
    my $i_break = shift;

    if ( $mate_index_to_go[$i_break] >= 0 ) {

        # CAUTION: infinite recursion possible here:
        #   set_closing_breakpoint calls set_forced_breakpoint, and
        #   set_forced_breakpoint call set_closing_breakpoint
        #   ( test files attrib.t, BasicLyx.pm.html).
        # Don't reduce the '2' in the statement below
        if ( $mate_index_to_go[$i_break] > $i_break + 2 ) {

            # break before } ] and ), but sub set_forced_breakpoint will decide
            # to break before or after a ? and :
            my $inc = ( $tokens_to_go[$i_break] eq '?' ) ? 0 : 1;
            set_forced_breakpoint( $mate_index_to_go[$i_break] - $inc );
        }
    }
    else {
        my $type_sequence = $type_sequence_to_go[$i_break];
        if ($type_sequence) {
            my $closing_token = $matching_token{ $tokens_to_go[$i_break] };
            $postponed_breakpoint{$type_sequence} = 1;
        }
    }
}

sub compare_indentation_levels {

    # check to see if output line tabbing agrees with input line
    # this can be very useful for debugging a script which has an extra
    # or missing brace
    my ( $guessed_indentation_level, $structural_indentation_level ) = @_;
    if ( $guessed_indentation_level ne $structural_indentation_level ) {
        $last_tabbing_disagreement = $input_line_number;

        if ($in_tabbing_disagreement) {
        }
        else {
            $tabbing_disagreement_count++;

            if ( $tabbing_disagreement_count <= MAX_NAG_MESSAGES ) {
                write_logfile_entry(
"Start indentation disagreement: input=$guessed_indentation_level; output=$structural_indentation_level\n"
                );
            }
            $in_tabbing_disagreement    = $input_line_number;
            $first_tabbing_disagreement = $in_tabbing_disagreement
              unless ($first_tabbing_disagreement);
        }
    }
    else {

        if ($in_tabbing_disagreement) {

            if ( $tabbing_disagreement_count <= MAX_NAG_MESSAGES ) {
                write_logfile_entry(
"End indentation disagreement from input line $in_tabbing_disagreement\n"
                );

                if ( $tabbing_disagreement_count == MAX_NAG_MESSAGES ) {
                    write_logfile_entry(
                        "No further tabbing disagreements will be noted\n");
                }
            }
            $in_tabbing_disagreement = 0;
        }
    }
}

#####################################################################
#
# the Perl::Tidy::IndentationItem class supplies items which contain
# how much whitespace should be used at the start of a line
#
#####################################################################

package Perl::Tidy::IndentationItem;

# Indexes for indentation items
use constant SPACES             => 0;     # total leading white spaces
use constant LEVEL              => 1;     # the indentation 'level'
use constant CI_LEVEL           => 2;     # the 'continuation level'
use constant AVAILABLE_SPACES   => 3;     # how many left spaces available
                                          # for this level
use constant CLOSED             => 4;     # index where we saw closing '}'
use constant COMMA_COUNT        => 5;     # how many commas at this level?
use constant SEQUENCE_NUMBER    => 6;     # output batch number
use constant INDEX              => 7;     # index in output batch list
use constant HAVE_CHILD         => 8;     # any dependents?
use constant RECOVERABLE_SPACES => 9;     # how many spaces to the right
                                          # we would like to move to get
                                          # alignment (negative if left)
use constant ALIGN_PAREN        => 10;    # do we want to try to align
                                          # with an opening structure?
use constant MARKED             => 11;    # if visited by corrector logic
use constant STACK_DEPTH        => 12;    # indentation nesting depth
use constant STARTING_INDEX     => 13;    # first token index of this level
use constant ARROW_COUNT        => 14;    # how many =>'s

sub new {

    # Create an 'indentation_item' which describes one level of leading
    # whitespace when the '-lp' indentation is used.  We return
    # a reference to an anonymous array of associated variables.
    # See above constants for storage scheme.
    my (
        $class,               $spaces,           $level,
        $ci_level,            $available_spaces, $index,
        $gnu_sequence_number, $align_paren,      $stack_depth,
        $starting_index,
    ) = @_;
    my $closed            = -1;
    my $arrow_count       = 0;
    my $comma_count       = 0;
    my $have_child        = 0;
    my $want_right_spaces = 0;
    my $marked            = 0;
    bless [
        $spaces,              $level,          $ci_level,
        $available_spaces,    $closed,         $comma_count,
        $gnu_sequence_number, $index,          $have_child,
        $want_right_spaces,   $align_paren,    $marked,
        $stack_depth,         $starting_index, $arrow_count,
    ], $class;
}

sub permanently_decrease_AVAILABLE_SPACES {

    # make a permanent reduction in the available indentation spaces
    # at one indentation item.  NOTE: if there are child nodes, their
    # total SPACES must be reduced by the caller.

    my ( $item, $spaces_needed ) = @_;
    my $available_spaces = $item->get_AVAILABLE_SPACES();
    my $deleted_spaces =
      ( $available_spaces > $spaces_needed )
      ? $spaces_needed
      : $available_spaces;
    $item->decrease_AVAILABLE_SPACES($deleted_spaces);
    $item->decrease_SPACES($deleted_spaces);
    $item->set_RECOVERABLE_SPACES(0);

    return $deleted_spaces;
}

sub tentatively_decrease_AVAILABLE_SPACES {

    # We are asked to tentatively delete $spaces_needed of indentation
    # for a indentation item.  We may want to undo this later.  NOTE: if
    # there are child nodes, their total SPACES must be reduced by the
    # caller.
    my ( $item, $spaces_needed ) = @_;
    my $available_spaces = $item->get_AVAILABLE_SPACES();
    my $deleted_spaces =
      ( $available_spaces > $spaces_needed )
      ? $spaces_needed
      : $available_spaces;
    $item->decrease_AVAILABLE_SPACES($deleted_spaces);
    $item->decrease_SPACES($deleted_spaces);
    $item->increase_RECOVERABLE_SPACES($deleted_spaces);
    return $deleted_spaces;
}

sub get_STACK_DEPTH {
    my $self = shift;
    return $self->[STACK_DEPTH];
}

sub get_SPACES {
    my $self = shift;
    return $self->[SPACES];
}

sub get_MARKED {
    my $self = shift;
    return $self->[MARKED];
}

sub set_MARKED {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[MARKED] = $value;
    }
    return $self->[MARKED];
}

sub get_AVAILABLE_SPACES {
    my $self = shift;
    return $self->[AVAILABLE_SPACES];
}

sub decrease_SPACES {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[SPACES] -= $value;
    }
    return $self->[SPACES];
}

sub decrease_AVAILABLE_SPACES {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[AVAILABLE_SPACES] -= $value;
    }
    return $self->[AVAILABLE_SPACES];
}

sub get_ALIGN_PAREN {
    my $self = shift;
    return $self->[ALIGN_PAREN];
}

sub get_RECOVERABLE_SPACES {
    my $self = shift;
    return $self->[RECOVERABLE_SPACES];
}

sub set_RECOVERABLE_SPACES {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[RECOVERABLE_SPACES] = $value;
    }
    return $self->[RECOVERABLE_SPACES];
}

sub increase_RECOVERABLE_SPACES {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[RECOVERABLE_SPACES] += $value;
    }
    return $self->[RECOVERABLE_SPACES];
}

sub get_CI_LEVEL {
    my $self = shift;
    return $self->[CI_LEVEL];
}

sub get_LEVEL {
    my $self = shift;
    return $self->[LEVEL];
}

sub get_SEQUENCE_NUMBER {
    my $self = shift;
    return $self->[SEQUENCE_NUMBER];
}

sub get_INDEX {
    my $self = shift;
    return $self->[INDEX];
}

sub get_STARTING_INDEX {
    my $self = shift;
    return $self->[STARTING_INDEX];
}

sub set_HAVE_CHILD {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[HAVE_CHILD] = $value;
    }
    return $self->[HAVE_CHILD];
}

sub get_HAVE_CHILD {
    my $self = shift;
    return $self->[HAVE_CHILD];
}

sub set_ARROW_COUNT {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[ARROW_COUNT] = $value;
    }
    return $self->[ARROW_COUNT];
}

sub get_ARROW_COUNT {
    my $self = shift;
    return $self->[ARROW_COUNT];
}

sub set_COMMA_COUNT {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[COMMA_COUNT] = $value;
    }
    return $self->[COMMA_COUNT];
}

sub get_COMMA_COUNT {
    my $self = shift;
    return $self->[COMMA_COUNT];
}

sub set_CLOSED {
    my ( $self, $value ) = @_;
    if ( defined($value) ) {
        $self->[CLOSED] = $value;
    }
    return $self->[CLOSED];
}

sub get_CLOSED {
    my $self = shift;
    return $self->[CLOSED];
}

#####################################################################
#
# the Perl::Tidy::VerticalAligner::Line class supplies an object to
# contain a single output line
#
#####################################################################

package Perl::Tidy::VerticalAligner::Line;

{

    use strict;
    use Carp;

    use constant JMAX                      => 0;
    use constant JMAX_ORIGINAL_LINE        => 1;
    use constant RTOKENS                   => 2;
    use constant RFIELDS                   => 3;
    use constant RPATTERNS                 => 4;
    use constant INDENTATION               => 5;
    use constant LEADING_SPACE_COUNT       => 6;
    use constant OUTDENT_LONG_LINES        => 7;
    use constant LIST_TYPE                 => 8;
    use constant IS_HANGING_SIDE_COMMENT   => 9;
    use constant RALIGNMENTS               => 10;
    use constant MAXIMUM_LINE_LENGTH       => 11;
    use constant RVERTICAL_TIGHTNESS_FLAGS => 12;

    my %_index_map;
    $_index_map{jmax}                      = JMAX;
    $_index_map{jmax_original_line}        = JMAX_ORIGINAL_LINE;
    $_index_map{rtokens}                   = RTOKENS;
    $_index_map{rfields}                   = RFIELDS;
    $_index_map{rpatterns}                 = RPATTERNS;
    $_index_map{indentation}               = INDENTATION;
    $_index_map{leading_space_count}       = LEADING_SPACE_COUNT;
    $_index_map{outdent_long_lines}        = OUTDENT_LONG_LINES;
    $_index_map{list_type}                 = LIST_TYPE;
    $_index_map{is_hanging_side_comment}   = IS_HANGING_SIDE_COMMENT;
    $_index_map{ralignments}               = RALIGNMENTS;
    $_index_map{maximum_line_length}       = MAXIMUM_LINE_LENGTH;
    $_index_map{rvertical_tightness_flags} = RVERTICAL_TIGHTNESS_FLAGS;

    my @_default_data = ();
    $_default_data[JMAX]                      = undef;
    $_default_data[JMAX_ORIGINAL_LINE]        = undef;
    $_default_data[RTOKENS]                   = undef;
    $_default_data[RFIELDS]                   = undef;
    $_default_data[RPATTERNS]                 = undef;
    $_default_data[INDENTATION]               = undef;
    $_default_data[LEADING_SPACE_COUNT]       = undef;
    $_default_data[OUTDENT_LONG_LINES]        = undef;
    $_default_data[LIST_TYPE]                 = undef;
    $_default_data[IS_HANGING_SIDE_COMMENT]   = undef;
    $_default_data[RALIGNMENTS]               = [];
    $_default_data[MAXIMUM_LINE_LENGTH]       = undef;
    $_default_data[RVERTICAL_TIGHTNESS_FLAGS] = undef;

    {

        # methods to count object population
        my $_count = 0;
        sub get_count        { $_count; }
        sub _increment_count { ++$_count }
        sub _decrement_count { --$_count }
    }

    # Constructor may be called as a class method
    sub new {
        my ( $caller, %arg ) = @_;
        my $caller_is_obj = ref($caller);
        my $class = $caller_is_obj || $caller;
        no strict "refs";
        my $self = bless [], $class;

        $self->[RALIGNMENTS] = [];

        my $index;
        foreach ( keys %_index_map ) {
            $index = $_index_map{$_};
            if    ( exists $arg{$_} ) { $self->[$index] = $arg{$_} }
            elsif ($caller_is_obj)    { $self->[$index] = $caller->[$index] }
            else { $self->[$index] = $_default_data[$index] }
        }

        $self->_increment_count();
        return $self;
    }

    sub DESTROY {
        $_[0]->_decrement_count();
    }

    sub get_jmax                      { $_[0]->[JMAX] }
    sub get_jmax_original_line        { $_[0]->[JMAX_ORIGINAL_LINE] }
    sub get_rtokens                   { $_[0]->[RTOKENS] }
    sub get_rfields                   { $_[0]->[RFIELDS] }
    sub get_rpatterns                 { $_[0]->[RPATTERNS] }
    sub get_indentation               { $_[0]->[INDENTATION] }
    sub get_leading_space_count       { $_[0]->[LEADING_SPACE_COUNT] }
    sub get_outdent_long_lines        { $_[0]->[OUTDENT_LONG_LINES] }
    sub get_list_type                 { $_[0]->[LIST_TYPE] }
    sub get_is_hanging_side_comment   { $_[0]->[IS_HANGING_SIDE_COMMENT] }
    sub get_rvertical_tightness_flags { $_[0]->[RVERTICAL_TIGHTNESS_FLAGS] }

    sub set_column     { $_[0]->[RALIGNMENTS]->[ $_[1] ]->set_column( $_[2] ) }
    sub get_alignment  { $_[0]->[RALIGNMENTS]->[ $_[1] ] }
    sub get_alignments { @{ $_[0]->[RALIGNMENTS] } }
    sub get_column     { $_[0]->[RALIGNMENTS]->[ $_[1] ]->get_column() }

    sub get_starting_column {
        $_[0]->[RALIGNMENTS]->[ $_[1] ]->get_starting_column();
    }

    sub increment_column {
        $_[0]->[RALIGNMENTS]->[ $_[1] ]->increment_column( $_[2] );
    }
    sub set_alignments { my $self = shift; @{ $self->[RALIGNMENTS] } = @_; }

    sub current_field_width {
        my $self = shift;
        my ($j) = @_;
        if ( $j == 0 ) {
            return $self->get_column($j);
        }
        else {
            return $self->get_column($j) - $self->get_column( $j - 1 );
        }
    }

    sub field_width_growth {
        my $self = shift;
        my $j    = shift;
        return $self->get_column($j) - $self->get_starting_column($j);
    }

    sub starting_field_width {
        my $self = shift;
        my $j    = shift;
        if ( $j == 0 ) {
            return $self->get_starting_column($j);
        }
        else {
            return $self->get_starting_column($j) -
              $self->get_starting_column( $j - 1 );
        }
    }

    sub increase_field_width {

        my $self = shift;
        my ( $j, $pad ) = @_;
        my $jmax = $self->get_jmax();
        for my $k ( $j .. $jmax ) {
            $self->increment_column( $k, $pad );
        }
    }

    sub get_available_space_on_right {
        my $self = shift;
        my $jmax = $self->get_jmax();
        return $self->[MAXIMUM_LINE_LENGTH] - $self->get_column($jmax);
    }

    sub set_jmax                    { $_[0]->[JMAX]                    = $_[1] }
    sub set_jmax_original_line      { $_[0]->[JMAX_ORIGINAL_LINE]      = $_[1] }
    sub set_rtokens                 { $_[0]->[RTOKENS]                 = $_[1] }
    sub set_rfields                 { $_[0]->[RFIELDS]                 = $_[1] }
    sub set_rpatterns               { $_[0]->[RPATTERNS]               = $_[1] }
    sub set_indentation             { $_[0]->[INDENTATION]             = $_[1] }
    sub set_leading_space_count     { $_[0]->[LEADING_SPACE_COUNT]     = $_[1] }
    sub set_outdent_long_lines      { $_[0]->[OUTDENT_LONG_LINES]      = $_[1] }
    sub set_list_type               { $_[0]->[LIST_TYPE]               = $_[1] }
    sub set_is_hanging_side_comment { $_[0]->[IS_HANGING_SIDE_COMMENT] = $_[1] }
    sub set_alignment               { $_[0]->[RALIGNMENTS]->[ $_[1] ]  = $_[2] }

}

#####################################################################
#
# the Perl::Tidy::VerticalAligner::Alignment class holds information
# on a single column being aligned
#
#####################################################################
package Perl::Tidy::VerticalAligner::Alignment;

{

    use strict;

    #use Carp;

    # Symbolic array indexes
    use constant COLUMN          => 0;    # the current column number
    use constant STARTING_COLUMN => 1;    # column number when created
    use constant MATCHING_TOKEN  => 2;    # what token we are matching
    use constant STARTING_LINE   => 3;    # the line index of creation
    use constant ENDING_LINE     => 4;    # the most recent line to use it
    use constant SAVED_COLUMN    => 5;    # the most recent line to use it
    use constant SERIAL_NUMBER   => 6;    # unique number for this alignment
                                          # (just its index in an array)

    # Correspondence between variables and array indexes
    my %_index_map;
    $_index_map{column}          = COLUMN;
    $_index_map{starting_column} = STARTING_COLUMN;
    $_index_map{matching_token}  = MATCHING_TOKEN;
    $_index_map{starting_line}   = STARTING_LINE;
    $_index_map{ending_line}     = ENDING_LINE;
    $_index_map{saved_column}    = SAVED_COLUMN;
    $_index_map{serial_number}   = SERIAL_NUMBER;

    my @_default_data = ();
    $_default_data[COLUMN]          = undef;
    $_default_data[STARTING_COLUMN] = undef;
    $_default_data[MATCHING_TOKEN]  = undef;
    $_default_data[STARTING_LINE]   = undef;
    $_default_data[ENDING_LINE]     = undef;
    $_default_data[SAVED_COLUMN]    = undef;
    $_default_data[SERIAL_NUMBER]   = undef;

    # class population count
    {
        my $_count = 0;
        sub get_count        { $_count; }
        sub _increment_count { ++$_count }
        sub _decrement_count { --$_count }
    }

    # constructor
    sub new {
        my ( $caller, %arg ) = @_;
        my $caller_is_obj = ref($caller);
        my $class = $caller_is_obj || $caller;
        no strict "refs";
        my $self = bless [], $class;

        foreach ( keys %_index_map ) {
            my $index = $_index_map{$_};
            if    ( exists $arg{$_} ) { $self->[$index] = $arg{$_} }
            elsif ($caller_is_obj)    { $self->[$index] = $caller->[$index] }
            else { $self->[$index] = $_default_data[$index] }
        }
        $self->_increment_count();
        return $self;
    }

    sub DESTROY {
        $_[0]->_decrement_count();
    }

    sub get_column          { return $_[0]->[COLUMN] }
    sub get_starting_column { return $_[0]->[STARTING_COLUMN] }
    sub get_matching_token  { return $_[0]->[MATCHING_TOKEN] }
    sub get_starting_line   { return $_[0]->[STARTING_LINE] }
    sub get_ending_line     { return $_[0]->[ENDING_LINE] }
    sub get_serial_number   { return $_[0]->[SERIAL_NUMBER] }

    sub set_column          { $_[0]->[COLUMN]          = $_[1] }
    sub set_starting_column { $_[0]->[STARTING_COLUMN] = $_[1] }
    sub set_matching_token  { $_[0]->[MATCHING_TOKEN]  = $_[1] }
    sub set_starting_line   { $_[0]->[STARTING_LINE]   = $_[1] }
    sub set_ending_line     { $_[0]->[ENDING_LINE]     = $_[1] }
    sub increment_column { $_[0]->[COLUMN] += $_[1] }

    sub save_column    { $_[0]->[SAVED_COLUMN] = $_[0]->[COLUMN] }
    sub restore_column { $_[0]->[COLUMN]       = $_[0]->[SAVED_COLUMN] }

}

package Perl::Tidy::VerticalAligner;

# The Perl::Tidy::VerticalAligner package collects output lines and
# attempts to line up certain common tokens, such as => and #, which are
# identified by the calling routine.
#
# There are two main routines: valign_input and flush.  Append acts as a
# storage buffer, collecting lines into a group which can be vertically
# aligned.  When alignment is no longer possible or desirable, it dumps
# the group to flush.
#
#     valign_input -----> flush
#
#     collects          writes
#     vertical          one
#     groups            group

BEGIN {

    # Caution: these debug flags produce a lot of output
    # They should all be 0 except when debugging small scripts

    use constant VALIGN_DEBUG_FLAG_APPEND  => 0;
    use constant VALIGN_DEBUG_FLAG_APPEND0 => 0;
    use constant VALIGN_DEBUG_FLAG_TERNARY => 0;
    use constant VALIGN_DEBUG_FLAG_TABS    => 0;

    my $debug_warning = sub {
        print STDOUT "VALIGN_DEBUGGING with key $_[0]\n";
    };

    VALIGN_DEBUG_FLAG_APPEND  && $debug_warning->('APPEND');
    VALIGN_DEBUG_FLAG_APPEND0 && $debug_warning->('APPEND0');
    VALIGN_DEBUG_FLAG_TERNARY && $debug_warning->('TERNARY');
    VALIGN_DEBUG_FLAG_TABS    && $debug_warning->('TABS');

}

use vars qw(
  $vertical_aligner_self
  $current_line
  $maximum_alignment_index
  $ralignment_list
  $maximum_jmax_seen
  $minimum_jmax_seen
  $previous_minimum_jmax_seen
  $previous_maximum_jmax_seen
  $maximum_line_index
  $group_level
  $group_type
  $group_maximum_gap
  $marginal_match
  $last_level_written
  $last_leading_space_count
  $extra_indent_ok
  $zero_count
  @group_lines
  $last_comment_column
  $last_side_comment_line_number
  $last_side_comment_length
  $last_side_comment_level
  $outdented_line_count
  $first_outdented_line_at
  $last_outdented_line_at
  $diagnostics_object
  $logger_object
  $file_writer_object
  @side_comment_history
  $comment_leading_space_count
  $is_matching_terminal_line
  $consecutive_block_comments

  $cached_line_text
  $cached_line_type
  $cached_line_flag
  $cached_seqno
  $cached_line_valid
  $cached_line_leading_space_count
  $cached_seqno_string

  $valign_buffer_filling
  @valign_buffer

  $seqno_string
  $last_nonblank_seqno_string

  $rOpts

  $rOpts_maximum_line_length
  $rOpts_variable_maximum_line_length
  $rOpts_continuation_indentation
  $rOpts_indent_columns
  $rOpts_tabs
  $rOpts_entab_leading_whitespace
  $rOpts_valign

  $rOpts_fixed_position_side_comment
  $rOpts_minimum_space_to_comment

);

sub initialize {

    my $class;

    ( $class, $rOpts, $file_writer_object, $logger_object, $diagnostics_object )
      = @_;

    # variables describing the entire space group:
    $ralignment_list            = [];
    $group_level                = 0;
    $last_level_written         = -1;
    $extra_indent_ok            = 0;    # can we move all lines to the right?
    $last_side_comment_length   = 0;
    $maximum_jmax_seen          = 0;
    $minimum_jmax_seen          = 0;
    $previous_minimum_jmax_seen = 0;
    $previous_maximum_jmax_seen = 0;

    # variables describing each line of the group
    @group_lines = ();                  # list of all lines in group

    $outdented_line_count          = 0;
    $first_outdented_line_at       = 0;
    $last_outdented_line_at        = 0;
    $last_side_comment_line_number = 0;
    $last_side_comment_level       = -1;
    $is_matching_terminal_line     = 0;

    # most recent 3 side comments; [ line number, column ]
    $side_comment_history[0] = [ -300, 0 ];
    $side_comment_history[1] = [ -200, 0 ];
    $side_comment_history[2] = [ -100, 0 ];

    # valign_output_step_B cache:
    $cached_line_text                = "";
    $cached_line_type                = 0;
    $cached_line_flag                = 0;
    $cached_seqno                    = 0;
    $cached_line_valid               = 0;
    $cached_line_leading_space_count = 0;
    $cached_seqno_string             = "";

    # string of sequence numbers joined together
    $seqno_string               = "";
    $last_nonblank_seqno_string = "";

    # frequently used parameters
    $rOpts_indent_columns           = $rOpts->{'indent-columns'};
    $rOpts_tabs                     = $rOpts->{'tabs'};
    $rOpts_entab_leading_whitespace = $rOpts->{'entab-leading-whitespace'};
    $rOpts_fixed_position_side_comment =
      $rOpts->{'fixed-position-side-comment'};
    $rOpts_minimum_space_to_comment = $rOpts->{'minimum-space-to-comment'};
    $rOpts_maximum_line_length      = $rOpts->{'maximum-line-length'};
    $rOpts_variable_maximum_line_length =
      $rOpts->{'variable-maximum-line-length'};
    $rOpts_valign = $rOpts->{'valign'};

    $consecutive_block_comments = 0;
    forget_side_comment();

    initialize_for_new_group();

    $vertical_aligner_self = {};
    bless $vertical_aligner_self, $class;
    return $vertical_aligner_self;
}

sub initialize_for_new_group {
    $maximum_line_index      = -1;      # lines in the current group
    $maximum_alignment_index = -1;      # alignments in current group
    $zero_count              = 0;       # count consecutive lines without tokens
    $current_line            = undef;   # line being matched for alignment
    $group_maximum_gap       = 0;       # largest gap introduced
    $group_type              = "";
    $marginal_match          = 0;
    $comment_leading_space_count = 0;
    $last_leading_space_count    = 0;
}

# interface to Perl::Tidy::Diagnostics routines
sub write_diagnostics {
    if ($diagnostics_object) {
        $diagnostics_object->write_diagnostics(@_);
    }
}

# interface to Perl::Tidy::Logger routines
sub warning {
    if ($logger_object) {
        $logger_object->warning(@_);
    }
}

sub write_logfile_entry {
    if ($logger_object) {
        $logger_object->write_logfile_entry(@_);
    }
}

sub report_definite_bug {
    if ($logger_object) {
        $logger_object->report_definite_bug();
    }
}

sub get_SPACES {

    # return the number of leading spaces associated with an indentation
    # variable $indentation is either a constant number of spaces or an
    # object with a get_SPACES method.
    my $indentation = shift;
    return ref($indentation) ? $indentation->get_SPACES() : $indentation;
}

sub get_RECOVERABLE_SPACES {

    # return the number of spaces (+ means shift right, - means shift left)
    # that we would like to shift a group of lines with the same indentation
    # to get them to line up with their opening parens
    my $indentation = shift;
    return ref($indentation) ? $indentation->get_RECOVERABLE_SPACES() : 0;
}

sub get_STACK_DEPTH {

    my $indentation = shift;
    return ref($indentation) ? $indentation->get_STACK_DEPTH() : 0;
}

sub make_alignment {
    my ( $col, $token ) = @_;

    # make one new alignment at column $col which aligns token $token
    ++$maximum_alignment_index;
    my $alignment = new Perl::Tidy::VerticalAligner::Alignment(
        column          => $col,
        starting_column => $col,
        matching_token  => $token,
        starting_line   => $maximum_line_index,
        ending_line     => $maximum_line_index,
        serial_number   => $maximum_alignment_index,
    );
    $ralignment_list->[$maximum_alignment_index] = $alignment;
    return $alignment;
}

sub dump_alignments {
    print STDOUT
"Current Alignments:\ni\ttoken\tstarting_column\tcolumn\tstarting_line\tending_line\n";
    for my $i ( 0 .. $maximum_alignment_index ) {
        my $column          = $ralignment_list->[$i]->get_column();
        my $starting_column = $ralignment_list->[$i]->get_starting_column();
        my $matching_token  = $ralignment_list->[$i]->get_matching_token();
        my $starting_line   = $ralignment_list->[$i]->get_starting_line();
        my $ending_line     = $ralignment_list->[$i]->get_ending_line();
        print STDOUT
"$i\t$matching_token\t$starting_column\t$column\t$starting_line\t$ending_line\n";
    }
}

sub save_alignment_columns {
    for my $i ( 0 .. $maximum_alignment_index ) {
        $ralignment_list->[$i]->save_column();
    }
}

sub restore_alignment_columns {
    for my $i ( 0 .. $maximum_alignment_index ) {
        $ralignment_list->[$i]->restore_column();
    }
}

sub forget_side_comment {
    $last_comment_column = 0;
}

sub maximum_line_length_for_level {

    # return maximum line length for line starting with a given level
    my $maximum_line_length = $rOpts_maximum_line_length;
    if ($rOpts_variable_maximum_line_length) {
        my $level = shift;
        if ( $level < 0 ) { $level = 0 }
        $maximum_line_length += $level * $rOpts_indent_columns;
    }
    return $maximum_line_length;
}

sub valign_input {

    # Place one line in the current vertical group.
    #
    # The input parameters are:
    #     $level = indentation level of this line
    #     $rfields = reference to array of fields
    #     $rpatterns = reference to array of patterns, one per field
    #     $rtokens   = reference to array of tokens starting fields 1,2,..
    #
    # Here is an example of what this package does.  In this example,
    # we are trying to line up both the '=>' and the '#'.
    #
    #         '18' => 'grave',    #   \`
    #         '19' => 'acute',    #   `'
    #         '20' => 'caron',    #   \v
    # <-tabs-><f1-><--field 2 ---><-f3->
    # |            |              |    |
    # |            |              |    |
    # col1        col2         col3 col4
    #
    # The calling routine has already broken the entire line into 3 fields as
    # indicated.  (So the work of identifying promising common tokens has
    # already been done).
    #
    # In this example, there will be 2 tokens being matched: '=>' and '#'.
    # They are the leading parts of fields 2 and 3, but we do need to know
    # what they are so that we can dump a group of lines when these tokens
    # change.
    #
    # The fields contain the actual characters of each field.  The patterns
    # are like the fields, but they contain mainly token types instead
    # of tokens, so they have fewer characters.  They are used to be
    # sure we are matching fields of similar type.
    #
    # In this example, there will be 4 column indexes being adjusted.  The
    # first one is always at zero.  The interior columns are at the start of
    # the matching tokens, and the last one tracks the maximum line length.
    #
    # Each time a new line comes in, it joins the current vertical
    # group if possible.  Otherwise it causes the current group to be dumped
    # and a new group is started.
    #
    # For each new group member, the column locations are increased, as
    # necessary, to make room for the new fields.  When the group is finally
    # output, these column numbers are used to compute the amount of spaces of
    # padding needed for each field.
    #
    # Programming note: the fields are assumed not to have any tab characters.
    # Tabs have been previously removed except for tabs in quoted strings and
    # side comments.  Tabs in these fields can mess up the column counting.
    # The log file warns the user if there are any such tabs.

    my (
        $level,               $level_end,
        $indentation,         $rfields,
        $rtokens,             $rpatterns,
        $is_forced_break,     $outdent_long_lines,
        $is_terminal_ternary, $is_terminal_statement,
        $do_not_pad,          $rvertical_tightness_flags,
        $level_jump,
    ) = @_;

    # number of fields is $jmax
    # number of tokens between fields is $jmax-1
    my $jmax = $#{$rfields};

    my $leading_space_count = get_SPACES($indentation);

    # set outdented flag to be sure we either align within statements or
    # across statement boundaries, but not both.
    my $is_outdented = $last_leading_space_count > $leading_space_count;
    $last_leading_space_count = $leading_space_count;

    # Patch: undo for hanging side comment
    my $is_hanging_side_comment =
      ( $jmax == 1 && $rtokens->[0] eq '#' && $rfields->[0] =~ /^\s*$/ );
    $is_outdented = 0 if $is_hanging_side_comment;

    # Forget side comment alignment after seeing 2 or more block comments
    my $is_block_comment = ( $jmax == 0 && $rfields->[0] =~ /^#/ );
    if ($is_block_comment) {
        $consecutive_block_comments++;
    }
    else {
        if ( $consecutive_block_comments > 1 ) { forget_side_comment() }
        $consecutive_block_comments = 0;
    }

    VALIGN_DEBUG_FLAG_APPEND0 && do {
        print STDOUT
"APPEND0: entering lines=$maximum_line_index new #fields= $jmax, leading_count=$leading_space_count last_cmt=$last_comment_column force=$is_forced_break\n";
    };

    # Validate cached line if necessary: If we can produce a container
    # with just 2 lines total by combining an existing cached opening
    # token with the closing token to follow, then we will mark both
    # cached flags as valid.
    if ($rvertical_tightness_flags) {
        if (   $maximum_line_index <= 0
            && $cached_line_type
            && $cached_seqno
            && $rvertical_tightness_flags->[2]
            && $rvertical_tightness_flags->[2] == $cached_seqno )
        {
            $rvertical_tightness_flags->[3] ||= 1;
            $cached_line_valid ||= 1;
        }
    }

    # do not join an opening block brace with an unbalanced line
    # unless requested with a flag value of 2
    if (   $cached_line_type == 3
        && $maximum_line_index < 0
        && $cached_line_flag < 2
        && $level_jump != 0 )
    {
        $cached_line_valid = 0;
    }

    # patch until new aligner is finished
    if ($do_not_pad) { my_flush() }

    # shouldn't happen:
    if ( $level < 0 ) { $level = 0 }

    # do not align code across indentation level changes
    # or if vertical alignment is turned off for debugging
    if ( $level != $group_level || $is_outdented || !$rOpts_valign ) {

        # we are allowed to shift a group of lines to the right if its
        # level is greater than the previous and next group
        $extra_indent_ok =
          ( $level < $group_level && $last_level_written < $group_level );

        my_flush();

        # If we know that this line will get flushed out by itself because
        # of level changes, we can leave the extra_indent_ok flag set.
        # That way, if we get an external flush call, we will still be
        # able to do some -lp alignment if necessary.
        $extra_indent_ok = ( $is_terminal_statement && $level > $group_level );

        $group_level = $level;

        # wait until after the above flush to get the leading space
        # count because it may have been changed if the -icp flag is in
        # effect
        $leading_space_count = get_SPACES($indentation);

    }

    # --------------------------------------------------------------------
    # Patch to collect outdentable block COMMENTS
    # --------------------------------------------------------------------
    my $is_blank_line = "";
    if ( $group_type eq 'COMMENT' ) {
        if (
            (
                   $is_block_comment
                && $outdent_long_lines
                && $leading_space_count == $comment_leading_space_count
            )
            || $is_blank_line
          )
        {
            $group_lines[ ++$maximum_line_index ] = $rfields->[0];
            return;
        }
        else {
            my_flush();
        }
    }

    # --------------------------------------------------------------------
    # add dummy fields for terminal ternary
    # --------------------------------------------------------------------
    my $j_terminal_match;
    if ( $is_terminal_ternary && $current_line ) {
        $j_terminal_match =
          fix_terminal_ternary( $rfields, $rtokens, $rpatterns );
        $jmax = @{$rfields} - 1;
    }

    # --------------------------------------------------------------------
    # add dummy fields for else statement
    # --------------------------------------------------------------------
    if (   $rfields->[0] =~ /^else\s*$/
        && $current_line
        && $level_jump == 0 )
    {
        $j_terminal_match = fix_terminal_else( $rfields, $rtokens, $rpatterns );
        $jmax = @{$rfields} - 1;
    }

    # --------------------------------------------------------------------
    # Step 1. Handle simple line of code with no fields to match.
    # --------------------------------------------------------------------
    if ( $jmax <= 0 ) {
        $zero_count++;

        if ( $maximum_line_index >= 0
            && !get_RECOVERABLE_SPACES( $group_lines[0]->get_indentation() ) )
        {

            # flush the current group if it has some aligned columns..
            if ( $group_lines[0]->get_jmax() > 1 ) { my_flush() }

            # flush current group if we are just collecting side comments..
            elsif (

                # ...and we haven't seen a comment lately
                ( $zero_count > 3 )

                # ..or if this new line doesn't fit to the left of the comments
                || ( ( $leading_space_count + length( $$rfields[0] ) ) >
                    $group_lines[0]->get_column(0) )
              )
            {
                my_flush();
            }
        }

        # patch to start new COMMENT group if this comment may be outdented
        if (   $is_block_comment
            && $outdent_long_lines
            && $maximum_line_index < 0 )
        {
            $group_type                           = 'COMMENT';
            $comment_leading_space_count          = $leading_space_count;
            $group_lines[ ++$maximum_line_index ] = $rfields->[0];
            return;
        }

        # just write this line directly if no current group, no side comment,
        # and no space recovery is needed.
        if ( $maximum_line_index < 0 && !get_RECOVERABLE_SPACES($indentation) )
        {
            valign_output_step_B( $leading_space_count, $$rfields[0], 0,
                $outdent_long_lines, $rvertical_tightness_flags, $level );
            return;
        }
    }
    else {
        $zero_count = 0;
    }

    # programming check: (shouldn't happen)
    # an error here implies an incorrect call was made
    if ( $jmax > 0 && ( $#{$rtokens} != ( $jmax - 1 ) ) ) {
        warning(
"Program bug in Perl::Tidy::VerticalAligner - number of tokens = $#{$rtokens} should be one less than number of fields: $#{$rfields})\n"
        );
        report_definite_bug();
    }

    # --------------------------------------------------------------------
    # create an object to hold this line
    # --------------------------------------------------------------------
    my $new_line = new Perl::Tidy::VerticalAligner::Line(
        jmax                      => $jmax,
        jmax_original_line        => $jmax,
        rtokens                   => $rtokens,
        rfields                   => $rfields,
        rpatterns                 => $rpatterns,
        indentation               => $indentation,
        leading_space_count       => $leading_space_count,
        outdent_long_lines        => $outdent_long_lines,
        list_type                 => "",
        is_hanging_side_comment   => $is_hanging_side_comment,
        maximum_line_length       => maximum_line_length_for_level($level),
        rvertical_tightness_flags => $rvertical_tightness_flags,
    );

    # Initialize a global flag saying if the last line of the group should
    # match end of group and also terminate the group.  There should be no
    # returns between here and where the flag is handled at the bottom.
    my $col_matching_terminal = 0;
    if ( defined($j_terminal_match) ) {

        # remember the column of the terminal ? or { to match with
        $col_matching_terminal = $current_line->get_column($j_terminal_match);

        # set global flag for sub decide_if_aligned
        $is_matching_terminal_line = 1;
    }

    # --------------------------------------------------------------------
    # It simplifies things to create a zero length side comment
    # if none exists.
    # --------------------------------------------------------------------
    make_side_comment( $new_line, $level_end );

    # --------------------------------------------------------------------
    # Decide if this is a simple list of items.
    # There are 3 list types: none, comma, comma-arrow.
    # We use this below to be less restrictive in deciding what to align.
    # --------------------------------------------------------------------
    if ($is_forced_break) {
        decide_if_list($new_line);
    }

    if ($current_line) {

        # --------------------------------------------------------------------
        # Allow hanging side comment to join current group, if any
        # This will help keep side comments aligned, because otherwise we
        # will have to start a new group, making alignment less likely.
        # --------------------------------------------------------------------
        join_hanging_comment( $new_line, $current_line )
          if $is_hanging_side_comment;

        # --------------------------------------------------------------------
        # If there is just one previous line, and it has more fields
        # than the new line, try to join fields together to get a match with
        # the new line.  At the present time, only a single leading '=' is
        # allowed to be compressed out.  This is useful in rare cases where
        # a table is forced to use old breakpoints because of side comments,
        # and the table starts out something like this:
        #   my %MonthChars = ('0', 'Jan',   # side comment
        #                     '1', 'Feb',
        #                     '2', 'Mar',
        # Eliminating the '=' field will allow the remaining fields to line up.
        # This situation does not occur if there are no side comments
        # because scan_list would put a break after the opening '('.
        # --------------------------------------------------------------------
        eliminate_old_fields( $new_line, $current_line );

        # --------------------------------------------------------------------
        # If the new line has more fields than the current group,
        # see if we can match the first fields and combine the remaining
        # fields of the new line.
        # --------------------------------------------------------------------
        eliminate_new_fields( $new_line, $current_line );

        # --------------------------------------------------------------------
        # Flush previous group unless all common tokens and patterns match..
        # --------------------------------------------------------------------
        check_match( $new_line, $current_line );

        # --------------------------------------------------------------------
        # See if there is space for this line in the current group (if any)
        # --------------------------------------------------------------------
        if ($current_line) {
            check_fit( $new_line, $current_line );
        }
    }

    # --------------------------------------------------------------------
    # Append this line to the current group (or start new group)
    # --------------------------------------------------------------------
    add_to_group($new_line);

    # Future update to allow this to vary:
    $current_line = $new_line if ( $maximum_line_index == 0 );

    # output this group if it ends in a terminal else or ternary line
    if ( defined($j_terminal_match) ) {

        # if there is only one line in the group (maybe due to failure to match
        # perfectly with previous lines), then align the ? or { of this
        # terminal line with the previous one unless that would make the line
        # too long
        if ( $maximum_line_index == 0 ) {
            my $col_now = $current_line->get_column($j_terminal_match);
            my $pad     = $col_matching_terminal - $col_now;
            my $padding_available =
              $current_line->get_available_space_on_right();
            if ( $pad > 0 && $pad <= $padding_available ) {
                $current_line->increase_field_width( $j_terminal_match, $pad );
            }
        }
        my_flush();
        $is_matching_terminal_line = 0;
    }

    # --------------------------------------------------------------------
    # Step 8. Some old debugging stuff
    # --------------------------------------------------------------------
    VALIGN_DEBUG_FLAG_APPEND && do {
        print STDOUT "APPEND fields:";
        dump_array(@$rfields);
        print STDOUT "APPEND tokens:";
        dump_array(@$rtokens);
        print STDOUT "APPEND patterns:";
        dump_array(@$rpatterns);
        dump_alignments();
    };

    return;
}

sub join_hanging_comment {

    my $line = shift;
    my $jmax = $line->get_jmax();
    return 0 unless $jmax == 1;    # must be 2 fields
    my $rtokens = $line->get_rtokens();
    return 0 unless $$rtokens[0] eq '#';    # the second field is a comment..
    my $rfields = $line->get_rfields();
    return 0 unless $$rfields[0] =~ /^\s*$/;    # the first field is empty...
    my $old_line            = shift;
    my $maximum_field_index = $old_line->get_jmax();
    return 0
      unless $maximum_field_index > $jmax;    # the current line has more fields
    my $rpatterns = $line->get_rpatterns();

    $line->set_is_hanging_side_comment(1);
    $jmax = $maximum_field_index;
    $line->set_jmax($jmax);
    $$rfields[$jmax]         = $$rfields[1];
    $$rtokens[ $jmax - 1 ]   = $$rtokens[0];
    $$rpatterns[ $jmax - 1 ] = $$rpatterns[0];
    for ( my $j = 1 ; $j < $jmax ; $j++ ) {
        $$rfields[$j]         = " ";  # NOTE: caused glitch unless 1 blank, why?
        $$rtokens[ $j - 1 ]   = "";
        $$rpatterns[ $j - 1 ] = "";
    }
    return 1;
}

sub eliminate_old_fields {

    my $new_line = shift;
    my $jmax     = $new_line->get_jmax();
    if ( $jmax > $maximum_jmax_seen ) { $maximum_jmax_seen = $jmax }
    if ( $jmax < $minimum_jmax_seen ) { $minimum_jmax_seen = $jmax }

    # there must be one previous line
    return unless ( $maximum_line_index == 0 );

    my $old_line            = shift;
    my $maximum_field_index = $old_line->get_jmax();

    ###############################################
    # this line must have fewer fields
    return unless $maximum_field_index > $jmax;
    ###############################################

    # Identify specific cases where field elimination is allowed:
    # case=1: both lines have comma-separated lists, and the first
    #         line has an equals
    # case=2: both lines have leading equals

    # case 1 is the default
    my $case = 1;

    # See if case 2: both lines have leading '='
    # We'll require smiliar leading patterns in this case
    my $old_rtokens   = $old_line->get_rtokens();
    my $rtokens       = $new_line->get_rtokens();
    my $rpatterns     = $new_line->get_rpatterns();
    my $old_rpatterns = $old_line->get_rpatterns();
    if (   $rtokens->[0] =~ /^=\d*$/
        && $old_rtokens->[0] eq $rtokens->[0]
        && $old_rpatterns->[0] eq $rpatterns->[0] )
    {
        $case = 2;
    }

    # not too many fewer fields in new line for case 1
    return unless ( $case != 1 || $maximum_field_index - 2 <= $jmax );

    # case 1 must have side comment
    my $old_rfields = $old_line->get_rfields();
    return
      if ( $case == 1
        && length( $$old_rfields[$maximum_field_index] ) == 0 );

    my $rfields = $new_line->get_rfields();

    my $hid_equals = 0;

    my @new_alignments        = ();
    my @new_fields            = ();
    my @new_matching_patterns = ();
    my @new_matching_tokens   = ();

    my $j = 0;
    my $k;
    my $current_field   = '';
    my $current_pattern = '';

    # loop over all old tokens
    my $in_match = 0;
    for ( $k = 0 ; $k < $maximum_field_index ; $k++ ) {
        $current_field   .= $$old_rfields[$k];
        $current_pattern .= $$old_rpatterns[$k];
        last if ( $j > $jmax - 1 );

        if ( $$old_rtokens[$k] eq $$rtokens[$j] ) {
            $in_match                  = 1;
            $new_fields[$j]            = $current_field;
            $new_matching_patterns[$j] = $current_pattern;
            $current_field             = '';
            $current_pattern           = '';
            $new_matching_tokens[$j]   = $$old_rtokens[$k];
            $new_alignments[$j]        = $old_line->get_alignment($k);
            $j++;
        }
        else {

            if ( $$old_rtokens[$k] =~ /^\=\d*$/ ) {
                last if ( $case == 2 );    # avoid problems with stuff
                                           # like:   $a=$b=$c=$d;
                $hid_equals = 1;
            }
            last
              if ( $in_match && $case == 1 )
              ;    # disallow gaps in matching field types in case 1
        }
    }

    # Modify the current state if we are successful.
    # We must exactly reach the ends of both lists for success.
    if (   ( $j == $jmax )
        && ( $current_field eq '' )
        && ( $case != 1 || $hid_equals ) )
    {
        $k = $maximum_field_index;
        $current_field   .= $$old_rfields[$k];
        $current_pattern .= $$old_rpatterns[$k];
        $new_fields[$j]            = $current_field;
        $new_matching_patterns[$j] = $current_pattern;

        $new_alignments[$j] = $old_line->get_alignment($k);
        $maximum_field_index = $j;

        $old_line->set_alignments(@new_alignments);
        $old_line->set_jmax($jmax);
        $old_line->set_rtokens( \@new_matching_tokens );
        $old_line->set_rfields( \@new_fields );
        $old_line->set_rpatterns( \@$rpatterns );
    }
}

# create an empty side comment if none exists
sub make_side_comment {
    my $new_line  = shift;
    my $level_end = shift;
    my $jmax      = $new_line->get_jmax();
    my $rtokens   = $new_line->get_rtokens();

    # if line does not have a side comment...
    if ( ( $jmax == 0 ) || ( $$rtokens[ $jmax - 1 ] ne '#' ) ) {
        my $rfields   = $new_line->get_rfields();
        my $rpatterns = $new_line->get_rpatterns();
        $$rtokens[$jmax]     = '#';
        $$rfields[ ++$jmax ] = '';
        $$rpatterns[$jmax]   = '#';
        $new_line->set_jmax($jmax);
        $new_line->set_jmax_original_line($jmax);
    }

    # line has a side comment..
    else {

        # don't remember old side comment location for very long
        my $line_number = $vertical_aligner_self->get_output_line_number();
        my $rfields     = $new_line->get_rfields();
        if (
            $line_number - $last_side_comment_line_number > 12

            # and don't remember comment location across block level changes
            || ( $level_end < $last_side_comment_level && $$rfields[0] =~ /^}/ )
          )
        {
            forget_side_comment();
        }
        $last_side_comment_line_number = $line_number;
        $last_side_comment_level       = $level_end;
    }
}

sub decide_if_list {

    my $line = shift;

    # A list will be taken to be a line with a forced break in which all
    # of the field separators are commas or comma-arrows (except for the
    # trailing #)

    # List separator tokens are things like ',3'   or '=>2',
    # where the trailing digit is the nesting depth.  Allow braces
    # to allow nested list items.
    my $rtokens    = $line->get_rtokens();
    my $test_token = $$rtokens[0];
    if ( $test_token =~ /^(\,|=>)/ ) {
        my $list_type = $test_token;
        my $jmax      = $line->get_jmax();

        foreach ( 1 .. $jmax - 2 ) {
            if ( $$rtokens[$_] !~ /^(\,|=>|\{)/ ) {
                $list_type = "";
                last;
            }
        }
        $line->set_list_type($list_type);
    }
}

sub eliminate_new_fields {

    return unless ( $maximum_line_index >= 0 );
    my ( $new_line, $old_line ) = @_;
    my $jmax = $new_line->get_jmax();

    my $old_rtokens = $old_line->get_rtokens();
    my $rtokens     = $new_line->get_rtokens();
    my $is_assignment =
      ( $rtokens->[0] =~ /^=\d*$/ && ( $old_rtokens->[0] eq $rtokens->[0] ) );

    # must be monotonic variation
    return unless ( $is_assignment || $previous_maximum_jmax_seen <= $jmax );

    # must be more fields in the new line
    my $maximum_field_index = $old_line->get_jmax();
    return unless ( $maximum_field_index < $jmax );

    unless ($is_assignment) {
        return
          unless ( $old_line->get_jmax_original_line() == $minimum_jmax_seen )
          ;    # only if monotonic

        # never combine fields of a comma list
        return
          unless ( $maximum_field_index > 1 )
          && ( $new_line->get_list_type() !~ /^,/ );
    }

    my $rfields       = $new_line->get_rfields();
    my $rpatterns     = $new_line->get_rpatterns();
    my $old_rpatterns = $old_line->get_rpatterns();

    # loop over all OLD tokens except comment and check match
    my $match = 1;
    my $k;
    for ( $k = 0 ; $k < $maximum_field_index - 1 ; $k++ ) {
        if (   ( $$old_rtokens[$k] ne $$rtokens[$k] )
            || ( $$old_rpatterns[$k] ne $$rpatterns[$k] ) )
        {
            $match = 0;
            last;
        }
    }

    # first tokens agree, so combine extra new tokens
    if ($match) {
        for $k ( $maximum_field_index .. $jmax - 1 ) {

            $$rfields[ $maximum_field_index - 1 ] .= $$rfields[$k];
            $$rfields[$k] = "";
            $$rpatterns[ $maximum_field_index - 1 ] .= $$rpatterns[$k];
            $$rpatterns[$k] = "";
        }

        $$rtokens[ $maximum_field_index - 1 ] = '#';
        $$rfields[$maximum_field_index]       = $$rfields[$jmax];
        $$rpatterns[$maximum_field_index]     = $$rpatterns[$jmax];
        $jmax                                 = $maximum_field_index;
    }
    $new_line->set_jmax($jmax);
}

sub fix_terminal_ternary {

    # Add empty fields as necessary to align a ternary term
    # like this:
    #
    #  my $leapyear =
    #      $year % 4   ? 0
    #    : $year % 100 ? 1
    #    : $year % 400 ? 0
    #    :               1;
    #
    # returns 1 if the terminal item should be indented

    my ( $rfields, $rtokens, $rpatterns ) = @_;

    my $jmax        = @{$rfields} - 1;
    my $old_line    = $group_lines[$maximum_line_index];
    my $rfields_old = $old_line->get_rfields();

    my $rpatterns_old       = $old_line->get_rpatterns();
    my $rtokens_old         = $old_line->get_rtokens();
    my $maximum_field_index = $old_line->get_jmax();

    # look for the question mark after the :
    my ($jquestion);
    my $depth_question;
    my $pad = "";
    for ( my $j = 0 ; $j < $maximum_field_index ; $j++ ) {
        my $tok = $rtokens_old->[$j];
        if ( $tok =~ /^\?(\d+)$/ ) {
            $depth_question = $1;

            # depth must be correct
            next unless ( $depth_question eq $group_level );

            $jquestion = $j;
            if ( $rfields_old->[ $j + 1 ] =~ /^(\?\s*)/ ) {
                $pad = " " x length($1);
            }
            else {
                return;    # shouldn't happen
            }
            last;
        }
    }
    return unless ( defined($jquestion) );    # shouldn't happen

    # Now splice the tokens and patterns of the previous line
    # into the else line to insure a match.  Add empty fields
    # as necessary.
    my $jadd = $jquestion;

    # Work on copies of the actual arrays in case we have
    # to return due to an error
    my @fields   = @{$rfields};
    my @patterns = @{$rpatterns};
    my @tokens   = @{$rtokens};

    VALIGN_DEBUG_FLAG_TERNARY && do {
        local $" = '><';
        print STDOUT "CURRENT FIELDS=<@{$rfields_old}>\n";
        print STDOUT "CURRENT TOKENS=<@{$rtokens_old}>\n";
        print STDOUT "CURRENT PATTERNS=<@{$rpatterns_old}>\n";
        print STDOUT "UNMODIFIED FIELDS=<@{$rfields}>\n";
        print STDOUT "UNMODIFIED TOKENS=<@{$rtokens}>\n";
        print STDOUT "UNMODIFIED PATTERNS=<@{$rpatterns}>\n";
    };

    # handle cases of leading colon on this line
    if ( $fields[0] =~ /^(:\s*)(.*)$/ ) {

        my ( $colon, $therest ) = ( $1, $2 );

        # Handle sub-case of first field with leading colon plus additional code
        # This is the usual situation as at the '1' below:
        #  ...
        #  : $year % 400 ? 0
        #  :               1;
        if ($therest) {

            # Split the first field after the leading colon and insert padding.
            # Note that this padding will remain even if the terminal value goes
            # out on a separate line.  This does not seem to look to bad, so no
            # mechanism has been included to undo it.
            my $field1 = shift @fields;
            unshift @fields, ( $colon, $pad . $therest );

            # change the leading pattern from : to ?
            return unless ( $patterns[0] =~ s/^\:/?/ );

            # install leading tokens and patterns of existing line
            unshift( @tokens,   @{$rtokens_old}[ 0 .. $jquestion ] );
            unshift( @patterns, @{$rpatterns_old}[ 0 .. $jquestion ] );

            # insert appropriate number of empty fields
            splice( @fields, 1, 0, ('') x $jadd ) if $jadd;
        }

        # handle sub-case of first field just equal to leading colon.
        # This can happen for example in the example below where
        # the leading '(' would create a new alignment token
        # : ( $name =~ /[]}]$/ ) ? ( $mname = $name )
        # :                        ( $mname = $name . '->' );
        else {

            return unless ( $jmax > 0 && $tokens[0] ne '#' ); # shouldn't happen

            # prepend a leading ? onto the second pattern
            $patterns[1] = "?b" . $patterns[1];

            # pad the second field
            $fields[1] = $pad . $fields[1];

            # install leading tokens and patterns of existing line, replacing
            # leading token and inserting appropriate number of empty fields
            splice( @tokens,   0, 1, @{$rtokens_old}[ 0 .. $jquestion ] );
            splice( @patterns, 1, 0, @{$rpatterns_old}[ 1 .. $jquestion ] );
            splice( @fields, 1, 0, ('') x $jadd ) if $jadd;
        }
    }

    # Handle case of no leading colon on this line.  This will
    # be the case when -wba=':' is used.  For example,
    #  $year % 400 ? 0 :
    #                1;
    else {

        # install leading tokens and patterns of existing line
        $patterns[0] = '?' . 'b' . $patterns[0];
        unshift( @tokens,   @{$rtokens_old}[ 0 .. $jquestion ] );
        unshift( @patterns, @{$rpatterns_old}[ 0 .. $jquestion ] );

        # insert appropriate number of empty fields
        $jadd = $jquestion + 1;
        $fields[0] = $pad . $fields[0];
        splice( @fields, 0, 0, ('') x $jadd ) if $jadd;
    }

    VALIGN_DEBUG_FLAG_TERNARY && do {
        local $" = '><';
        print STDOUT "MODIFIED TOKENS=<@tokens>\n";
        print STDOUT "MODIFIED PATTERNS=<@patterns>\n";
        print STDOUT "MODIFIED FIELDS=<@fields>\n";
    };

    # all ok .. update the arrays
    @{$rfields}   = @fields;
    @{$rtokens}   = @tokens;
    @{$rpatterns} = @patterns;

    # force a flush after this line
    return $jquestion;
}

sub fix_terminal_else {

    # Add empty fields as necessary to align a balanced terminal
    # else block to a previous if/elsif/unless block,
    # like this:
    #
    #  if   ( 1 || $x ) { print "ok 13\n"; }
    #  else             { print "not ok 13\n"; }
    #
    # returns 1 if the else block should be indented
    #
    my ( $rfields, $rtokens, $rpatterns ) = @_;
    my $jmax = @{$rfields} - 1;
    return unless ( $jmax > 0 );

    # check for balanced else block following if/elsif/unless
    my $rfields_old = $current_line->get_rfields();

    # TBD: add handling for 'case'
    return unless ( $rfields_old->[0] =~ /^(if|elsif|unless)\s*$/ );

    # look for the opening brace after the else, and extrace the depth
    my $tok_brace = $rtokens->[0];
    my $depth_brace;
    if ( $tok_brace =~ /^\{(\d+)/ ) { $depth_brace = $1; }

    # probably:  "else # side_comment"
    else { return }

    my $rpatterns_old       = $current_line->get_rpatterns();
    my $rtokens_old         = $current_line->get_rtokens();
    my $maximum_field_index = $current_line->get_jmax();

    # be sure the previous if/elsif is followed by an opening paren
    my $jparen    = 0;
    my $tok_paren = '(' . $depth_brace;
    my $tok_test  = $rtokens_old->[$jparen];
    return unless ( $tok_test eq $tok_paren );    # shouldn't happen

    # Now find the opening block brace
    my ($jbrace);
    for ( my $j = 1 ; $j < $maximum_field_index ; $j++ ) {
        my $tok = $rtokens_old->[$j];
        if ( $tok eq $tok_brace ) {
            $jbrace = $j;
            last;
        }
    }
    return unless ( defined($jbrace) );           # shouldn't happen

    # Now splice the tokens and patterns of the previous line
    # into the else line to insure a match.  Add empty fields
    # as necessary.
    my $jadd = $jbrace - $jparen;
    splice( @{$rtokens},   0, 0, @{$rtokens_old}[ $jparen .. $jbrace - 1 ] );
    splice( @{$rpatterns}, 1, 0, @{$rpatterns_old}[ $jparen + 1 .. $jbrace ] );
    splice( @{$rfields}, 1, 0, ('') x $jadd );

    # force a flush after this line if it does not follow a case
    return $jbrace
      unless ( $rfields_old->[0] =~ /^case\s*$/ );
}

{    # sub check_match
    my %is_good_alignment;

    BEGIN {

        # Vertically aligning on certain "good" tokens is usually okay
        # so we can be less restrictive in marginal cases.
        @_ = qw( { ? => = );
        push @_, (',');
        @is_good_alignment{@_} = (1) x scalar(@_);
    }

    sub check_match {

        # See if the current line matches the current vertical alignment group.
        # If not, flush the current group.
        my $new_line = shift;
        my $old_line = shift;

        # uses global variables:
        #  $previous_minimum_jmax_seen
        #  $maximum_jmax_seen
        #  $maximum_line_index
        #  $marginal_match
        my $jmax                = $new_line->get_jmax();
        my $maximum_field_index = $old_line->get_jmax();

        # flush if this line has too many fields
        if ( $jmax > $maximum_field_index ) { goto NO_MATCH }

        # flush if adding this line would make a non-monotonic field count
        if (
            ( $maximum_field_index > $jmax )    # this has too few fields
            && (
                ( $previous_minimum_jmax_seen <
                    $jmax )                     # and wouldn't be monotonic
                || ( $old_line->get_jmax_original_line() != $maximum_jmax_seen )
            )
          )
        {
            goto NO_MATCH;
        }

        # otherwise see if this line matches the current group
        my $jmax_original_line      = $new_line->get_jmax_original_line();
        my $is_hanging_side_comment = $new_line->get_is_hanging_side_comment();
        my $rtokens                 = $new_line->get_rtokens();
        my $rfields                 = $new_line->get_rfields();
        my $rpatterns               = $new_line->get_rpatterns();
        my $list_type               = $new_line->get_list_type();

        my $group_list_type = $old_line->get_list_type();
        my $old_rpatterns   = $old_line->get_rpatterns();
        my $old_rtokens     = $old_line->get_rtokens();

        my $jlimit = $jmax - 1;
        if ( $maximum_field_index > $jmax ) {
            $jlimit = $jmax_original_line;
            --$jlimit unless ( length( $new_line->get_rfields()->[$jmax] ) );
        }

        # handle comma-separated lists ..
        if ( $group_list_type && ( $list_type eq $group_list_type ) ) {
            for my $j ( 0 .. $jlimit ) {
                my $old_tok = $$old_rtokens[$j];
                next unless $old_tok;
                my $new_tok = $$rtokens[$j];
                next unless $new_tok;

                # lists always match ...
                # unless they would align any '=>'s with ','s
                goto NO_MATCH
                  if ( $old_tok =~ /^=>/ && $new_tok =~ /^,/
                    || $new_tok =~ /^=>/ && $old_tok =~ /^,/ );
            }
        }

        # do detailed check for everything else except hanging side comments
        elsif ( !$is_hanging_side_comment ) {

            my $leading_space_count = $new_line->get_leading_space_count();

            my $max_pad = 0;
            my $min_pad = 0;
            my $saw_good_alignment;

            for my $j ( 0 .. $jlimit ) {

                my $old_tok = $$old_rtokens[$j];
                my $new_tok = $$rtokens[$j];

                # Note on encoding used for alignment tokens:
                # -------------------------------------------
                # Tokens are "decorated" with information which can help
                # prevent unwanted alignments.  Consider for example the
                # following two lines:
                #   local ( $xn, $xd ) = split( '/', &'rnorm(@_) );
                #   local ( $i, $f ) = &'bdiv( $xn, $xd );
                # There are three alignment tokens in each line, a comma,
                # an =, and a comma.  In the first line these three tokens
                # are encoded as:
                #    ,4+local-18     =3      ,4+split-7
                # and in the second line they are encoded as
                #    ,4+local-18     =3      ,4+&'bdiv-8
                # Tokens always at least have token name and nesting
                # depth.  So in this example the ='s are at depth 3 and
                # the ,'s are at depth 4.  This prevents aligning tokens
                # of different depths.  Commas contain additional
                # information, as follows:
                # ,  {depth} + {container name} - {spaces to opening paren}
                # This allows us to reject matching the rightmost commas
                # in the above two lines, since they are for different
                # function calls.  This encoding is done in
                # 'sub send_lines_to_vertical_aligner'.

                # Pick off actual token.
                # Everything up to the first digit is the actual token.
                my $alignment_token = $new_tok;
                if ( $alignment_token =~ /^([^\d]+)/ ) { $alignment_token = $1 }

                # see if the decorated tokens match
                my $tokens_match = $new_tok eq $old_tok

                  # Exception for matching terminal : of ternary statement..
                  # consider containers prefixed by ? and : a match
                  || ( $new_tok =~ /^,\d*\+\:/ && $old_tok =~ /^,\d*\+\?/ );

                # No match if the alignment tokens differ...
                if ( !$tokens_match ) {

                    # ...Unless this is a side comment
                    if (
                        $j == $jlimit

                        # and there is either at least one alignment token
                        # or this is a single item following a list.  This
                        # latter rule is required for 'December' to join
                        # the following list:
                        # my (@months) = (
                        #     '',       'January',   'February', 'March',
                        #     'April',  'May',       'June',     'July',
                        #     'August', 'September', 'October',  'November',
                        #     'December'
                        # );
                        # If it doesn't then the -lp formatting will fail.
                        && ( $j > 0 || $old_tok =~ /^,/ )
                      )
                    {
                        $marginal_match = 1
                          if ( $marginal_match == 0
                            && $maximum_line_index == 0 );
                        last;
                    }

                    goto NO_MATCH;
                }

                # Calculate amount of padding required to fit this in.
                # $pad is the number of spaces by which we must increase
                # the current field to squeeze in this field.
                my $pad =
                  length( $$rfields[$j] ) - $old_line->current_field_width($j);
                if ( $j == 0 ) { $pad += $leading_space_count; }

                # remember max pads to limit marginal cases
                if ( $alignment_token ne '#' ) {
                    if ( $pad > $max_pad ) { $max_pad = $pad }
                    if ( $pad < $min_pad ) { $min_pad = $pad }
                }
                if ( $is_good_alignment{$alignment_token} ) {
                    $saw_good_alignment = 1;
                }

                # If patterns don't match, we have to be careful...
                if ( $$old_rpatterns[$j] ne $$rpatterns[$j] ) {

                    # flag this as a marginal match since patterns differ
                    $marginal_match = 1
                      if ( $marginal_match == 0 && $maximum_line_index == 0 );

                    # We have to be very careful about aligning commas
                    # when the pattern's don't match, because it can be
                    # worse to create an alignment where none is needed
                    # than to omit one.  Here's an example where the ','s
                    # are not in named continers.  The first line below
                    # should not match the next two:
                    #   ( $a, $b ) = ( $b, $r );
                    #   ( $x1, $x2 ) = ( $x2 - $q * $x1, $x1 );
                    #   ( $y1, $y2 ) = ( $y2 - $q * $y1, $y1 );
                    if ( $alignment_token eq ',' ) {

                       # do not align commas unless they are in named containers
                        goto NO_MATCH unless ( $new_tok =~ /[A-Za-z]/ );
                    }

                    # do not align parens unless patterns match;
                    # large ugly spaces can occur in math expressions.
                    elsif ( $alignment_token eq '(' ) {

                        # But we can allow a match if the parens don't
                        # require any padding.
                        if ( $pad != 0 ) { goto NO_MATCH }
                    }

                    # Handle an '=' alignment with different patterns to
                    # the left.
                    elsif ( $alignment_token eq '=' ) {

                        # It is best to be a little restrictive when
                        # aligning '=' tokens.  Here is an example of
                        # two lines that we will not align:
                        #       my $variable=6;
                        #       $bb=4;
                        # The problem is that one is a 'my' declaration,
                        # and the other isn't, so they're not very similar.
                        # We will filter these out by comparing the first
                        # letter of the pattern.  This is crude, but works
                        # well enough.
                        if (
                            substr( $$old_rpatterns[$j], 0, 1 ) ne
                            substr( $$rpatterns[$j],     0, 1 ) )
                        {
                            goto NO_MATCH;
                        }

                        # If we pass that test, we'll call it a marginal match.
                        # Here is an example of a marginal match:
                        #       $done{$$op} = 1;
                        #       $op         = compile_bblock($op);
                        # The left tokens are both identifiers, but
                        # one accesses a hash and the other doesn't.
                        # We'll let this be a tentative match and undo
                        # it later if we don't find more than 2 lines
                        # in the group.
                        elsif ( $maximum_line_index == 0 ) {
                            $marginal_match =
                              2;    # =2 prevents being undone below
                        }
                    }
                }

                # Don't let line with fewer fields increase column widths
                # ( align3.t )
                if ( $maximum_field_index > $jmax ) {

                    # Exception: suspend this rule to allow last lines to join
                    if ( $pad > 0 ) { goto NO_MATCH; }
                }
            } ## end for my $j ( 0 .. $jlimit)

            # Turn off the "marginal match" flag in some cases...
            # A "marginal match" occurs when the alignment tokens agree
            # but there are differences in the other tokens (patterns).
            # If we leave the marginal match flag set, then the rule is that we
            # will align only if there are more than two lines in the group.
            # We will turn of the flag if we almost have a match
            # and either we have seen a good alignment token or we
            # just need a small pad (2 spaces) to fit.  These rules are
            # the result of experimentation.  Tokens which misaligned by just
            # one or two characters are annoying.  On the other hand,
            # large gaps to less important alignment tokens are also annoying.
            if (   $marginal_match == 1
                && $jmax == $maximum_field_index
                && ( $saw_good_alignment || ( $max_pad < 3 && $min_pad > -3 ) )
              )
            {
                $marginal_match = 0;
            }
            ##print "marginal=$marginal_match saw=$saw_good_alignment jmax=$jmax max=$maximum_field_index maxpad=$max_pad minpad=$min_pad\n";
        }

        # We have a match (even if marginal).
        # If the current line has fewer fields than the current group
        # but otherwise matches, copy the remaining group fields to
        # make it a perfect match.
        if ( $maximum_field_index > $jmax ) {
            my $comment = $$rfields[$jmax];
            for $jmax ( $jlimit .. $maximum_field_index ) {
                $$rtokens[$jmax]     = $$old_rtokens[$jmax];
                $$rfields[ ++$jmax ] = '';
                $$rpatterns[$jmax]   = $$old_rpatterns[$jmax];
            }
            $$rfields[$jmax] = $comment;
            $new_line->set_jmax($jmax);
        }
        return;

      NO_MATCH:
        ##print "BUBBA: no match jmax=$jmax  max=$maximum_field_index $group_list_type lines=$maximum_line_index token=$$old_rtokens[0]\n";
        my_flush();
        return;
    }
}

sub check_fit {

    return unless ( $maximum_line_index >= 0 );
    my $new_line = shift;
    my $old_line = shift;

    my $jmax                    = $new_line->get_jmax();
    my $leading_space_count     = $new_line->get_leading_space_count();
    my $is_hanging_side_comment = $new_line->get_is_hanging_side_comment();
    my $rtokens                 = $new_line->get_rtokens();
    my $rfields                 = $new_line->get_rfields();
    my $rpatterns               = $new_line->get_rpatterns();

    my $group_list_type = $group_lines[0]->get_list_type();

    my $padding_so_far    = 0;
    my $padding_available = $old_line->get_available_space_on_right();

    # save current columns in case this doesn't work
    save_alignment_columns();

    my ( $j, $pad, $eight );
    my $maximum_field_index = $old_line->get_jmax();
    for $j ( 0 .. $jmax ) {

        $pad = length( $$rfields[$j] ) - $old_line->current_field_width($j);

        if ( $j == 0 ) {
            $pad += $leading_space_count;
        }

        # remember largest gap of the group, excluding gap to side comment
        if (   $pad < 0
            && $group_maximum_gap < -$pad
            && $j > 0
            && $j < $jmax - 1 )
        {
            $group_maximum_gap = -$pad;
        }

        next if $pad < 0;

        ## This patch helps sometimes, but it doesn't check to see if
        ## the line is too long even without the side comment.  It needs
        ## to be reworked.
        ##don't let a long token with no trailing side comment push
        ##side comments out, or end a group.  (sidecmt1.t)
        ##next if ($j==$jmax-1 && length($$rfields[$jmax])==0);

        # This line will need space; lets see if we want to accept it..
        if (

            # not if this won't fit
            ( $pad > $padding_available )

            # previously, there were upper bounds placed on padding here
            # (maximum_whitespace_columns), but they were not really helpful

          )
        {

            # revert to starting state then flush; things didn't work out
            restore_alignment_columns();
            my_flush();
            last;
        }

        # patch to avoid excessive gaps in previous lines,
        # due to a line of fewer fields.
        #   return join( ".",
        #       $self->{"dfi"},  $self->{"aa"}, $self->rsvd,     $self->{"rd"},
        #       $self->{"area"}, $self->{"id"}, $self->{"sel"} );
        next if ( $jmax < $maximum_field_index && $j == $jmax - 1 );

        # looks ok, squeeze this field in
        $old_line->increase_field_width( $j, $pad );
        $padding_available -= $pad;

        # remember largest gap of the group, excluding gap to side comment
        if ( $pad > $group_maximum_gap && $j > 0 && $j < $jmax - 1 ) {
            $group_maximum_gap = $pad;
        }
    }
}

sub add_to_group {

    # The current line either starts a new alignment group or is
    # accepted into the current alignment group.
    my $new_line = shift;
    $group_lines[ ++$maximum_line_index ] = $new_line;

    # initialize field lengths if starting new group
    if ( $maximum_line_index == 0 ) {

        my $jmax    = $new_line->get_jmax();
        my $rfields = $new_line->get_rfields();
        my $rtokens = $new_line->get_rtokens();
        my $j;
        my $col = $new_line->get_leading_space_count();

        for $j ( 0 .. $jmax ) {
            $col += length( $$rfields[$j] );

            # create initial alignments for the new group
            my $token = "";
            if ( $j < $jmax ) { $token = $$rtokens[$j] }
            my $alignment = make_alignment( $col, $token );
            $new_line->set_alignment( $j, $alignment );
        }

        $maximum_jmax_seen = $jmax;
        $minimum_jmax_seen = $jmax;
    }

    # use previous alignments otherwise
    else {
        my @new_alignments =
          $group_lines[ $maximum_line_index - 1 ]->get_alignments();
        $new_line->set_alignments(@new_alignments);
    }

    # remember group jmax extremes for next call to valign_input
    $previous_minimum_jmax_seen = $minimum_jmax_seen;
    $previous_maximum_jmax_seen = $maximum_jmax_seen;
}

sub dump_array {

    # debug routine to dump array contents
    local $" = ')(';
    print STDOUT "(@_)\n";
}

# flush() sends the current Perl::Tidy::VerticalAligner group down the
# pipeline to Perl::Tidy::FileWriter.

# This is the external flush, which also empties the buffer and cache
sub flush {

    # the buffer must be emptied first, then any cached text
    dump_valign_buffer();

    if ( $maximum_line_index < 0 ) {
        if ($cached_line_type) {
            $seqno_string = $cached_seqno_string;
            valign_output_step_C( $cached_line_text,
                $cached_line_leading_space_count,
                $last_level_written );
            $cached_line_type    = 0;
            $cached_line_text    = "";
            $cached_seqno_string = "";
        }
    }
    else {
        my_flush();
    }
}

sub reduce_valign_buffer_indentation {

    my ($diff) = @_;
    if ( $valign_buffer_filling && $diff ) {
        my $max_valign_buffer = @valign_buffer;
        for ( my $i = 0 ; $i < $max_valign_buffer ; $i++ ) {
            my ( $line, $leading_space_count, $level ) =
              @{ $valign_buffer[$i] };
            my $ws = substr( $line, 0, $diff );
            if ( ( length($ws) == $diff ) && $ws =~ /^\s+$/ ) {
                $line = substr( $line, $diff );
            }
            if ( $leading_space_count >= $diff ) {
                $leading_space_count -= $diff;
                $level = level_change( $leading_space_count, $diff, $level );
            }
            $valign_buffer[$i] = [ $line, $leading_space_count, $level ];
        }
    }
}

sub level_change {

    # compute decrease in level when we remove $diff spaces from the
    # leading spaces
    my ( $leading_space_count, $diff, $level ) = @_;
    if ($rOpts_indent_columns) {
        my $olev =
          int( ( $leading_space_count + $diff ) / $rOpts_indent_columns );
        my $nlev = int( $leading_space_count / $rOpts_indent_columns );
        $level -= ( $olev - $nlev );
        if ( $level < 0 ) { $level = 0 }
    }
    return $level;
}

sub dump_valign_buffer {
    if (@valign_buffer) {
        foreach (@valign_buffer) {
            valign_output_step_D( @{$_} );
        }
        @valign_buffer = ();
    }
    $valign_buffer_filling = "";
}

# This is the internal flush, which leaves the cache intact
sub my_flush {

    return if ( $maximum_line_index < 0 );

    # handle a group of comment lines
    if ( $group_type eq 'COMMENT' ) {

        VALIGN_DEBUG_FLAG_APPEND0 && do {
            my ( $a, $b, $c ) = caller();
            print STDOUT
"APPEND0: Flush called from $a $b $c for COMMENT group: lines=$maximum_line_index \n";

        };
        my $leading_space_count = $comment_leading_space_count;
        my $leading_string      = get_leading_string($leading_space_count);

        # zero leading space count if any lines are too long
        my $max_excess = 0;
        for my $i ( 0 .. $maximum_line_index ) {
            my $str = $group_lines[$i];
            my $excess =
              length($str) +
              $leading_space_count -
              maximum_line_length_for_level($group_level);
            if ( $excess > $max_excess ) {
                $max_excess = $excess;
            }
        }

        if ( $max_excess > 0 ) {
            $leading_space_count -= $max_excess;
            if ( $leading_space_count < 0 ) { $leading_space_count = 0 }
            $last_outdented_line_at =
              $file_writer_object->get_output_line_number();
            unless ($outdented_line_count) {
                $first_outdented_line_at = $last_outdented_line_at;
            }
            $outdented_line_count += ( $maximum_line_index + 1 );
        }

        # write the group of lines
        my $outdent_long_lines = 0;
        for my $i ( 0 .. $maximum_line_index ) {
            valign_output_step_B( $leading_space_count, $group_lines[$i], 0,
                $outdent_long_lines, "", $group_level );
        }
    }

    # handle a group of code lines
    else {

        VALIGN_DEBUG_FLAG_APPEND0 && do {
            my $group_list_type = $group_lines[0]->get_list_type();
            my ( $a, $b, $c ) = caller();
            my $maximum_field_index = $group_lines[0]->get_jmax();
            print STDOUT
"APPEND0: Flush called from $a $b $c fields=$maximum_field_index list=$group_list_type lines=$maximum_line_index extra=$extra_indent_ok\n";

        };

        # some small groups are best left unaligned
        my $do_not_align = decide_if_aligned();

        # optimize side comment location
        $do_not_align = adjust_side_comment($do_not_align);

        # recover spaces for -lp option if possible
        my $extra_leading_spaces = get_extra_leading_spaces();

        # all lines of this group have the same basic leading spacing
        my $group_leader_length = $group_lines[0]->get_leading_space_count();

        # add extra leading spaces if helpful
        my $min_ci_gap = improve_continuation_indentation( $do_not_align,
            $group_leader_length );

        # loop to output all lines
        for my $i ( 0 .. $maximum_line_index ) {
            my $line = $group_lines[$i];
            valign_output_step_A( $line, $min_ci_gap, $do_not_align,
                $group_leader_length, $extra_leading_spaces );
        }
    }
    initialize_for_new_group();
}

sub decide_if_aligned {

    # Do not try to align two lines which are not really similar
    return unless $maximum_line_index == 1;
    return if ($is_matching_terminal_line);

    my $group_list_type = $group_lines[0]->get_list_type();

    my $do_not_align = (

        # always align lists
        !$group_list_type

          && (

            # don't align if it was just a marginal match
            $marginal_match

            # don't align two lines with big gap
            || $group_maximum_gap > 12

            # or lines with differing number of alignment tokens
            # TODO: this could be improved.  It occasionally rejects
            # good matches.
            || $previous_maximum_jmax_seen != $previous_minimum_jmax_seen
          )
    );

    # But try to convert them into a simple comment group if the first line
    # a has side comment
    my $rfields             = $group_lines[0]->get_rfields();
    my $maximum_field_index = $group_lines[0]->get_jmax();
    if (   $do_not_align
        && ( $maximum_line_index > 0 )
        && ( length( $$rfields[$maximum_field_index] ) > 0 ) )
    {
        combine_fields();
        $do_not_align = 0;
    }
    return $do_not_align;
}

sub adjust_side_comment {

    my $do_not_align = shift;

    # let's see if we can move the side comment field out a little
    # to improve readability (the last field is always a side comment field)
    my $have_side_comment       = 0;
    my $first_side_comment_line = -1;
    my $maximum_field_index     = $group_lines[0]->get_jmax();
    for my $i ( 0 .. $maximum_line_index ) {
        my $line = $group_lines[$i];

        if ( length( $line->get_rfields()->[$maximum_field_index] ) ) {
            $have_side_comment       = 1;
            $first_side_comment_line = $i;
            last;
        }
    }

    my $kmax = $maximum_field_index + 1;

    if ($have_side_comment) {

        my $line = $group_lines[0];

        # the maximum space without exceeding the line length:
        my $avail = $line->get_available_space_on_right();

        # try to use the previous comment column
        my $side_comment_column = $line->get_column( $kmax - 2 );
        my $move                = $last_comment_column - $side_comment_column;

##        my $sc_line0 = $side_comment_history[0]->[0];
##        my $sc_col0  = $side_comment_history[0]->[1];
##        my $sc_line1 = $side_comment_history[1]->[0];
##        my $sc_col1  = $side_comment_history[1]->[1];
##        my $sc_line2 = $side_comment_history[2]->[0];
##        my $sc_col2  = $side_comment_history[2]->[1];
##
##        # FUTURE UPDATES:
##        # Be sure to ignore 'do not align' and  '} # end comments'
##        # Find first $move > 0 and $move <= $avail as follows:
##        # 1. try sc_col1 if sc_col1 == sc_col0 && (line-sc_line0) < 12
##        # 2. try sc_col2 if (line-sc_line2) < 12
##        # 3. try min possible space, plus up to 8,
##        # 4. try min possible space

        if ( $kmax > 0 && !$do_not_align ) {

            # but if this doesn't work, give up and use the minimum space
            if ( $move > $avail ) {
                $move = $rOpts_minimum_space_to_comment - 1;
            }

            # but we want some minimum space to the comment
            my $min_move = $rOpts_minimum_space_to_comment - 1;
            if (   $move >= 0
                && $last_side_comment_length > 0
                && ( $first_side_comment_line == 0 )
                && $group_level == $last_level_written )
            {
                $min_move = 0;
            }

            if ( $move < $min_move ) {
                $move = $min_move;
            }

            # prevously, an upper bound was placed on $move here,
            # (maximum_space_to_comment), but it was not helpful

            # don't exceed the available space
            if ( $move > $avail ) { $move = $avail }

            # we can only increase space, never decrease
            if ( $move > 0 ) {
                $line->increase_field_width( $maximum_field_index - 1, $move );
            }

            # remember this column for the next group
            $last_comment_column = $line->get_column( $kmax - 2 );
        }
        else {

            # try to at least line up the existing side comment location
            if ( $kmax > 0 && $move > 0 && $move < $avail ) {
                $line->increase_field_width( $maximum_field_index - 1, $move );
                $do_not_align = 0;
            }

            # reset side comment column if we can't align
            else {
                forget_side_comment();
            }
        }
    }
    return $do_not_align;
}

sub improve_continuation_indentation {
    my ( $do_not_align, $group_leader_length ) = @_;

    # See if we can increase the continuation indentation
    # to move all continuation lines closer to the next field
    # (unless it is a comment).
    #
    # '$min_ci_gap'is the extra indentation that we may need to introduce.
    # We will only introduce this to fields which already have some ci.
    # Without this variable, we would occasionally get something like this
    # (Complex.pm):
    #
    # use overload '+' => \&plus,
    #   '-'            => \&minus,
    #   '*'            => \&multiply,
    #   ...
    #   'tan'          => \&tan,
    #   'atan2'        => \&atan2,
    #
    # Whereas with this variable, we can shift variables over to get this:
    #
    # use overload '+' => \&plus,
    #          '-'     => \&minus,
    #          '*'     => \&multiply,
    #          ...
    #          'tan'   => \&tan,
    #          'atan2' => \&atan2,

    ## Deactivated####################
    # The trouble with this patch is that it may, for example,
    # move in some 'or's  or ':'s, and leave some out, so that the
    # left edge alignment suffers.
    return 0;
    ###########################################

    my $maximum_field_index = $group_lines[0]->get_jmax();

    my $min_ci_gap = maximum_line_length_for_level($group_level);
    if ( $maximum_field_index > 1 && !$do_not_align ) {

        for my $i ( 0 .. $maximum_line_index ) {
            my $line                = $group_lines[$i];
            my $leading_space_count = $line->get_leading_space_count();
            my $rfields             = $line->get_rfields();

            my $gap =
              $line->get_column(0) -
              $leading_space_count -
              length( $$rfields[0] );

            if ( $leading_space_count > $group_leader_length ) {
                if ( $gap < $min_ci_gap ) { $min_ci_gap = $gap }
            }
        }

        if ( $min_ci_gap >= maximum_line_length_for_level($group_level) ) {
            $min_ci_gap = 0;
        }
    }
    else {
        $min_ci_gap = 0;
    }
    return $min_ci_gap;
}

sub valign_output_step_A {

    ###############################################################
    # This is Step A in writing vertically aligned lines.
    # The line is prepared according to the alignments which have
    # been found and shipped to the next step.
    ###############################################################

    my ( $line, $min_ci_gap, $do_not_align, $group_leader_length,
        $extra_leading_spaces )
      = @_;
    my $rfields                   = $line->get_rfields();
    my $leading_space_count       = $line->get_leading_space_count();
    my $outdent_long_lines        = $line->get_outdent_long_lines();
    my $maximum_field_index       = $line->get_jmax();
    my $rvertical_tightness_flags = $line->get_rvertical_tightness_flags();

    # add any extra spaces
    if ( $leading_space_count > $group_leader_length ) {
        $leading_space_count += $min_ci_gap;
    }

    my $str = $$rfields[0];

    # loop to concatenate all fields of this line and needed padding
    my $total_pad_count = 0;
    my ( $j, $pad );
    for $j ( 1 .. $maximum_field_index ) {

        # skip zero-length side comments
        last
          if ( ( $j == $maximum_field_index )
            && ( !defined( $$rfields[$j] ) || ( length( $$rfields[$j] ) == 0 ) )
          );

        # compute spaces of padding before this field
        my $col = $line->get_column( $j - 1 );
        $pad = $col - ( length($str) + $leading_space_count );

        if ($do_not_align) {
            $pad =
              ( $j < $maximum_field_index )
              ? 0
              : $rOpts_minimum_space_to_comment - 1;
        }

        # if the -fpsc flag is set, move the side comment to the selected
        # column if and only if it is possible, ignoring constraints on
        # line length and minimum space to comment
        if ( $rOpts_fixed_position_side_comment && $j == $maximum_field_index )
        {
            my $newpad = $pad + $rOpts_fixed_position_side_comment - $col - 1;
            if ( $newpad >= 0 ) { $pad = $newpad; }
        }

        # accumulate the padding
        if ( $pad > 0 ) { $total_pad_count += $pad; }

        # add this field
        if ( !defined $$rfields[$j] ) {
            write_diagnostics("UNDEFined field at j=$j\n");
        }

        # only add padding when we have a finite field;
        # this avoids extra terminal spaces if we have empty fields
        if ( length( $$rfields[$j] ) > 0 ) {
            $str .= ' ' x $total_pad_count;
            $total_pad_count = 0;
            $str .= $$rfields[$j];
        }
        else {
            $total_pad_count = 0;
        }

        # update side comment history buffer
        if ( $j == $maximum_field_index ) {
            my $lineno = $file_writer_object->get_output_line_number();
            shift @side_comment_history;
            push @side_comment_history, [ $lineno, $col ];
        }
    }

    my $side_comment_length = ( length( $$rfields[$maximum_field_index] ) );

    # ship this line off
    valign_output_step_B( $leading_space_count + $extra_leading_spaces,
        $str, $side_comment_length, $outdent_long_lines,
        $rvertical_tightness_flags, $group_level );
}

sub get_extra_leading_spaces {

    #----------------------------------------------------------
    # Define any extra indentation space (for the -lp option).
    # Here is why:
    # If a list has side comments, sub scan_list must dump the
    # list before it sees everything.  When this happens, it sets
    # the indentation to the standard scheme, but notes how
    # many spaces it would have liked to use.  We may be able
    # to recover that space here in the event that that all of the
    # lines of a list are back together again.
    #----------------------------------------------------------

    my $extra_leading_spaces = 0;
    if ($extra_indent_ok) {
        my $object = $group_lines[0]->get_indentation();
        if ( ref($object) ) {
            my $extra_indentation_spaces_wanted =
              get_RECOVERABLE_SPACES($object);

            # all indentation objects must be the same
            my $i;
            for $i ( 1 .. $maximum_line_index ) {
                if ( $object != $group_lines[$i]->get_indentation() ) {
                    $extra_indentation_spaces_wanted = 0;
                    last;
                }
            }

            if ($extra_indentation_spaces_wanted) {

                # the maximum space without exceeding the line length:
                my $avail = $group_lines[0]->get_available_space_on_right();
                $extra_leading_spaces =
                  ( $avail > $extra_indentation_spaces_wanted )
                  ? $extra_indentation_spaces_wanted
                  : $avail;

                # update the indentation object because with -icp the terminal
                # ');' will use the same adjustment.
                $object->permanently_decrease_AVAILABLE_SPACES(
                    -$extra_leading_spaces );
            }
        }
    }
    return $extra_leading_spaces;
}

sub combine_fields {

    # combine all fields except for the comment field  ( sidecmt.t )
    # Uses global variables:
    #  @group_lines
    #  $maximum_line_index
    my ( $j, $k );
    my $maximum_field_index = $group_lines[0]->get_jmax();
    for ( $j = 0 ; $j <= $maximum_line_index ; $j++ ) {
        my $line    = $group_lines[$j];
        my $rfields = $line->get_rfields();
        foreach ( 1 .. $maximum_field_index - 1 ) {
            $$rfields[0] .= $$rfields[$_];
        }
        $$rfields[1] = $$rfields[$maximum_field_index];

        $line->set_jmax(1);
        $line->set_column( 0, 0 );
        $line->set_column( 1, 0 );

    }
    $maximum_field_index = 1;

    for $j ( 0 .. $maximum_line_index ) {
        my $line    = $group_lines[$j];
        my $rfields = $line->get_rfields();
        for $k ( 0 .. $maximum_field_index ) {
            my $pad = length( $$rfields[$k] ) - $line->current_field_width($k);
            if ( $k == 0 ) {
                $pad += $group_lines[$j]->get_leading_space_count();
            }

            if ( $pad > 0 ) { $line->increase_field_width( $k, $pad ) }

        }
    }
}

sub get_output_line_number {

    # the output line number reported to a caller is the number of items
    # written plus the number of items in the buffer
    my $self = shift;
    1 + $maximum_line_index + $file_writer_object->get_output_line_number();
}

sub valign_output_step_B {

    ###############################################################
    # This is Step B in writing vertically aligned lines.
    # Vertical tightness is applied according to preset flags.
    # In particular this routine handles stacking of opening
    # and closing tokens.
    ###############################################################

    my ( $leading_space_count, $str, $side_comment_length, $outdent_long_lines,
        $rvertical_tightness_flags, $level )
      = @_;

    # handle outdenting of long lines:
    if ($outdent_long_lines) {
        my $excess =
          length($str) -
          $side_comment_length +
          $leading_space_count -
          maximum_line_length_for_level($level);
        if ( $excess > 0 ) {
            $leading_space_count = 0;
            $last_outdented_line_at =
              $file_writer_object->get_output_line_number();

            unless ($outdented_line_count) {
                $first_outdented_line_at = $last_outdented_line_at;
            }
            $outdented_line_count++;
        }
    }

    # Make preliminary leading whitespace.  It could get changed
    # later by entabbing, so we have to keep track of any changes
    # to the leading_space_count from here on.
    my $leading_string =
      $leading_space_count > 0 ? ( ' ' x $leading_space_count ) : "";

    # Unpack any recombination data; it was packed by
    # sub send_lines_to_vertical_aligner. Contents:
    #
    #   [0] type: 1=opening non-block    2=closing non-block
    #             3=opening block brace  4=closing block brace
    #   [1] flag: if opening: 1=no multiple steps, 2=multiple steps ok
    #             if closing: spaces of padding to use
    #   [2] sequence number of container
    #   [3] valid flag: do not append if this flag is false
    #
    my ( $open_or_close, $tightness_flag, $seqno, $valid, $seqno_beg,
        $seqno_end );
    if ($rvertical_tightness_flags) {
        (
            $open_or_close, $tightness_flag, $seqno, $valid, $seqno_beg,
            $seqno_end
        ) = @{$rvertical_tightness_flags};
    }

    $seqno_string = $seqno_end;

    # handle any cached line ..
    # either append this line to it or write it out
    if ( length($cached_line_text) ) {

        # Dump an invalid cached line
        if ( !$cached_line_valid ) {
            valign_output_step_C( $cached_line_text,
                $cached_line_leading_space_count,
                $last_level_written );
        }

        # Handle cached line ending in OPENING tokens
        elsif ( $cached_line_type == 1 || $cached_line_type == 3 ) {

            my $gap = $leading_space_count - length($cached_line_text);

            # handle option of just one tight opening per line:
            if ( $cached_line_flag == 1 ) {
                if ( defined($open_or_close) && $open_or_close == 1 ) {
                    $gap = -1;
                }
            }

            if ( $gap >= 0 && defined($seqno_beg) ) {
                $leading_string      = $cached_line_text . ' ' x $gap;
                $leading_space_count = $cached_line_leading_space_count;
                $seqno_string        = $cached_seqno_string . ':' . $seqno_beg;
                $level               = $last_level_written;
            }
            else {
                valign_output_step_C( $cached_line_text,
                    $cached_line_leading_space_count,
                    $last_level_written );
            }
        }

        # Handle cached line ending in CLOSING tokens
        else {
            my $test_line = $cached_line_text . ' ' x $cached_line_flag . $str;
            if (

                # The new line must start with container
                $seqno_beg

                # The container combination must be okay..
                && (

                    # okay to combine like types
                    ( $open_or_close == $cached_line_type )

                    # closing block brace may append to non-block
                    || ( $cached_line_type == 2 && $open_or_close == 4 )

                    # something like ');'
                    || ( !$open_or_close && $cached_line_type == 2 )

                )

                # The combined line must fit
                && (
                    length($test_line) <=
                    maximum_line_length_for_level($last_level_written) )
              )
            {

                $seqno_string = $cached_seqno_string . ':' . $seqno_beg;

                # Patch to outdent closing tokens ending # in ');'
                # If we are joining a line like ');' to a previous stacked
                # set of closing tokens, then decide if we may outdent the
                # combined stack to the indentation of the ');'.  Since we
                # should not normally outdent any of the other tokens more than
                # the indentation of the lines that contained them, we will
                # only do this if all of the corresponding opening
                # tokens were on the same line.  This can happen with
                # -sot and -sct.  For example, it is ok here:
                #   __PACKAGE__->load_components( qw(
                #         PK::Auto
                #         Core
                #   ));
                #
                #   But, for example, we do not outdent in this example because
                #   that would put the closing sub brace out farther than the
                #   opening sub brace:
                #
                #   perltidy -sot -sct
                #   $c->Tk::bind(
                #       '<Control-f>' => sub {
                #           my ($c) = @_;
                #           my $e = $c->XEvent;
                #           itemsUnderArea $c;
                #       } );
                #
                if ( $str =~ /^\);/ && $cached_line_text =~ /^[\)\}\]\s]*$/ ) {

                    # The way to tell this is if the stacked sequence numbers
                    # of this output line are the reverse of the stacked
                    # sequence numbers of the previous non-blank line of
                    # sequence numbers.  So we can join if the previous
                    # nonblank string of tokens is the mirror image.  For
                    # example if stack )}] is 13:8:6 then we are looking for a
                    # leading stack like [{( which is 6:8:13 We only need to
                    # check the two ends, because the intermediate tokens must
                    # fall in order.  Note on speed: having to split on colons
                    # and eliminate multiple colons might appear to be slow,
                    # but it's not an issue because we almost never come
                    # through here.  In a typical file we don't.
                    $seqno_string =~ s/^:+//;
                    $last_nonblank_seqno_string =~ s/^:+//;
                    $seqno_string =~ s/:+/:/g;
                    $last_nonblank_seqno_string =~ s/:+/:/g;

                    # how many spaces can we outdent?
                    my $diff =
                      $cached_line_leading_space_count - $leading_space_count;
                    if (   $diff > 0
                        && length($seqno_string)
                        && length($last_nonblank_seqno_string) ==
                        length($seqno_string) )
                    {
                        my @seqno_last =
                          ( split ':', $last_nonblank_seqno_string );
                        my @seqno_now = ( split ':', $seqno_string );
                        if (   $seqno_now[-1] == $seqno_last[0]
                            && $seqno_now[0] == $seqno_last[-1] )
                        {

                            # OK to outdent ..
                            # for absolute safety, be sure we only remove
                            # whitespace
                            my $ws = substr( $test_line, 0, $diff );
                            if ( ( length($ws) == $diff ) && $ws =~ /^\s+$/ ) {

                                $test_line = substr( $test_line, $diff );
                                $cached_line_leading_space_count -= $diff;
                                $last_level_written =
                                  level_change(
                                    $cached_line_leading_space_count,
                                    $diff, $last_level_written );
                                reduce_valign_buffer_indentation($diff);
                            }

                            # shouldn't happen, but not critical:
                            ##else {
                            ## ERROR transferring indentation here
                            ##}
                        }
                    }
                }

                $str                 = $test_line;
                $leading_string      = "";
                $leading_space_count = $cached_line_leading_space_count;
                $level               = $last_level_written;
            }
            else {
                valign_output_step_C( $cached_line_text,
                    $cached_line_leading_space_count,
                    $last_level_written );
            }
        }
    }
    $cached_line_type = 0;
    $cached_line_text = "";

    # make the line to be written
    my $line = $leading_string . $str;

    # write or cache this line
    if ( !$open_or_close || $side_comment_length > 0 ) {
        valign_output_step_C( $line, $leading_space_count, $level );
    }
    else {
        $cached_line_text                = $line;
        $cached_line_type                = $open_or_close;
        $cached_line_flag                = $tightness_flag;
        $cached_seqno                    = $seqno;
        $cached_line_valid               = $valid;
        $cached_line_leading_space_count = $leading_space_count;
        $cached_seqno_string             = $seqno_string;
    }

    $last_level_written       = $level;
    $last_side_comment_length = $side_comment_length;
    $extra_indent_ok          = 0;
}

sub valign_output_step_C {

    ###############################################################
    # This is Step C in writing vertically aligned lines.
    # Lines are either stored in a buffer or passed along to the next step.
    # The reason for storing lines is that we may later want to reduce their
    # indentation when -sot and -sct are both used.
    ###############################################################
    my @args = @_;

    # Dump any saved lines if we see a line with an unbalanced opening or
    # closing token.
    dump_valign_buffer() if ( $seqno_string && $valign_buffer_filling );

    # Either store or write this line
    if ($valign_buffer_filling) {
        push @valign_buffer, [@args];
    }
    else {
        valign_output_step_D(@args);
    }

    # For lines starting or ending with opening or closing tokens..
    if ($seqno_string) {
        $last_nonblank_seqno_string = $seqno_string;

        # Start storing lines when we see a line with multiple stacked opening
        # tokens.
        if ( $args[0] =~ /[\{\(\[]\s*[\{\(\[]$/ ) {
            $valign_buffer_filling = $seqno_string;
        }
    }
}

sub valign_output_step_D {

    ###############################################################
    # This is Step D in writing vertically aligned lines.
    # Write one vertically aligned line of code to the output object.
    ###############################################################

    my ( $line, $leading_space_count, $level ) = @_;

    # The line is currently correct if there is no tabbing (recommended!)
    # We may have to lop off some leading spaces and replace with tabs.
    if ( $leading_space_count > 0 ) {

        # Nothing to do if no tabs
        if ( !( $rOpts_tabs || $rOpts_entab_leading_whitespace )
            || $rOpts_indent_columns <= 0 )
        {

            # nothing to do
        }

        # Handle entab option
        elsif ($rOpts_entab_leading_whitespace) {
            my $space_count =
              $leading_space_count % $rOpts_entab_leading_whitespace;
            my $tab_count =
              int( $leading_space_count / $rOpts_entab_leading_whitespace );
            my $leading_string = "\t" x $tab_count . ' ' x $space_count;
            if ( $line =~ /^\s{$leading_space_count,$leading_space_count}/ ) {
                substr( $line, 0, $leading_space_count ) = $leading_string;
            }
            else {

                # shouldn't happen - program error counting whitespace
                # - skip entabbing
                VALIGN_DEBUG_FLAG_TABS
                  && warning(
"Error entabbing in valign_output_step_D: expected count=$leading_space_count\n"
                  );
            }
        }

        # Handle option of one tab per level
        else {
            my $leading_string = ( "\t" x $level );
            my $space_count =
              $leading_space_count - $level * $rOpts_indent_columns;

            # shouldn't happen:
            if ( $space_count < 0 ) {

                # But it could be an outdented comment
                if ( $line !~ /^\s*#/ ) {
                    VALIGN_DEBUG_FLAG_TABS
                      && warning(
"Error entabbing in valign_output_step_D: for level=$group_level count=$leading_space_count\n"
                      );
                }
                $leading_string = ( ' ' x $leading_space_count );
            }
            else {
                $leading_string .= ( ' ' x $space_count );
            }
            if ( $line =~ /^\s{$leading_space_count,$leading_space_count}/ ) {
                substr( $line, 0, $leading_space_count ) = $leading_string;
            }
            else {

                # shouldn't happen - program error counting whitespace
                # we'll skip entabbing
                VALIGN_DEBUG_FLAG_TABS
                  && warning(
"Error entabbing in valign_output_step_D: expected count=$leading_space_count\n"
                  );
            }
        }
    }
    $file_writer_object->write_code_line( $line . "\n" );
}

{    # begin get_leading_string

    my @leading_string_cache;

    sub get_leading_string {

        # define the leading whitespace string for this line..
        my $leading_whitespace_count = shift;

        # Handle case of zero whitespace, which includes multi-line quotes
        # (which may have a finite level; this prevents tab problems)
        if ( $leading_whitespace_count <= 0 ) {
            return "";
        }

        # look for previous result
        elsif ( $leading_string_cache[$leading_whitespace_count] ) {
            return $leading_string_cache[$leading_whitespace_count];
        }

        # must compute a string for this number of spaces
        my $leading_string;

        # Handle simple case of no tabs
        if ( !( $rOpts_tabs || $rOpts_entab_leading_whitespace )
            || $rOpts_indent_columns <= 0 )
        {
            $leading_string = ( ' ' x $leading_whitespace_count );
        }

        # Handle entab option
        elsif ($rOpts_entab_leading_whitespace) {
            my $space_count =
              $leading_whitespace_count % $rOpts_entab_leading_whitespace;
            my $tab_count = int(
                $leading_whitespace_count / $rOpts_entab_leading_whitespace );
            $leading_string = "\t" x $tab_count . ' ' x $space_count;
        }

        # Handle option of one tab per level
        else {
            $leading_string = ( "\t" x $group_level );
            my $space_count =
              $leading_whitespace_count - $group_level * $rOpts_indent_columns;

            # shouldn't happen:
            if ( $space_count < 0 ) {
                VALIGN_DEBUG_FLAG_TABS
                  && warning(
"Error in get_leading_string: for level=$group_level count=$leading_whitespace_count\n"
                  );

                # -- skip entabbing
                $leading_string = ( ' ' x $leading_whitespace_count );
            }
            else {
                $leading_string .= ( ' ' x $space_count );
            }
        }
        $leading_string_cache[$leading_whitespace_count] = $leading_string;
        return $leading_string;
    }
}    # end get_leading_string

sub report_anything_unusual {
    my $self = shift;
    if ( $outdented_line_count > 0 ) {
        write_logfile_entry(
            "$outdented_line_count long lines were outdented:\n");
        write_logfile_entry(
            "  First at output line $first_outdented_line_at\n");

        if ( $outdented_line_count > 1 ) {
            write_logfile_entry(
                "   Last at output line $last_outdented_line_at\n");
        }
        write_logfile_entry(
            "  use -noll to prevent outdenting, -l=n to increase line length\n"
        );
        write_logfile_entry("\n");
    }
}

#####################################################################
#
# the Perl::Tidy::FileWriter class writes the output file
#
#####################################################################

package Perl::Tidy::FileWriter;

# Maximum number of little messages; probably need not be changed.
use constant MAX_NAG_MESSAGES => 6;

sub write_logfile_entry {
    my $self          = shift;
    my $logger_object = $self->{_logger_object};
    if ($logger_object) {
        $logger_object->write_logfile_entry(@_);
    }
}

sub new {
    my $class = shift;
    my ( $line_sink_object, $rOpts, $logger_object ) = @_;

    bless {
        _line_sink_object           => $line_sink_object,
        _logger_object              => $logger_object,
        _rOpts                      => $rOpts,
        _output_line_number         => 1,
        _consecutive_blank_lines    => 0,
        _consecutive_nonblank_lines => 0,
        _first_line_length_error    => 0,
        _max_line_length_error      => 0,
        _last_line_length_error     => 0,
        _first_line_length_error_at => 0,
        _max_line_length_error_at   => 0,
        _last_line_length_error_at  => 0,
        _line_length_error_count    => 0,
        _max_output_line_length     => 0,
        _max_output_line_length_at  => 0,
    }, $class;
}

sub tee_on {
    my $self = shift;
    $self->{_line_sink_object}->tee_on();
}

sub tee_off {
    my $self = shift;
    $self->{_line_sink_object}->tee_off();
}

sub get_output_line_number {
    my $self = shift;
    return $self->{_output_line_number};
}

sub decrement_output_line_number {
    my $self = shift;
    $self->{_output_line_number}--;
}

sub get_consecutive_nonblank_lines {
    my $self = shift;
    return $self->{_consecutive_nonblank_lines};
}

sub reset_consecutive_blank_lines {
    my $self = shift;
    $self->{_consecutive_blank_lines} = 0;
}

sub want_blank_line {
    my $self = shift;
    unless ( $self->{_consecutive_blank_lines} ) {
        $self->write_blank_code_line();
    }
}

sub require_blank_code_lines {

    # write out the requested number of blanks regardless of the value of -mbl
    # unless -mbl=0.  This allows extra blank lines to be written for subs and
    # packages even with the default -mbl=1
    my $self   = shift;
    my $count  = shift;
    my $need   = $count - $self->{_consecutive_blank_lines};
    my $rOpts  = $self->{_rOpts};
    my $forced = $rOpts->{'maximum-consecutive-blank-lines'} > 0;
    for ( my $i = 0 ; $i < $need ; $i++ ) {
        $self->write_blank_code_line($forced);
    }
}

sub write_blank_code_line {
    my $self   = shift;
    my $forced = shift;
    my $rOpts  = $self->{_rOpts};
    return
      if (!$forced
        && $self->{_consecutive_blank_lines} >=
        $rOpts->{'maximum-consecutive-blank-lines'} );
    $self->{_consecutive_blank_lines}++;
    $self->{_consecutive_nonblank_lines} = 0;
    $self->write_line("\n");
}

sub write_code_line {
    my $self = shift;
    my $a    = shift;

    if ( $a =~ /^\s*$/ ) {
        my $rOpts = $self->{_rOpts};
        return
          if ( $self->{_consecutive_blank_lines} >=
            $rOpts->{'maximum-consecutive-blank-lines'} );
        $self->{_consecutive_blank_lines}++;
        $self->{_consecutive_nonblank_lines} = 0;
    }
    else {
        $self->{_consecutive_blank_lines} = 0;
        $self->{_consecutive_nonblank_lines}++;
    }
    $self->write_line($a);
}

sub write_line {
    my $self = shift;
    my $a    = shift;

    # TODO: go through and see if the test is necessary here
    if ( $a =~ /\n$/ ) { $self->{_output_line_number}++; }

    $self->{_line_sink_object}->write_line($a);

    # This calculation of excess line length ignores any internal tabs
    my $rOpts  = $self->{_rOpts};
    my $exceed = length($a) - $rOpts->{'maximum-line-length'} - 1;
    if ( $a =~ /^\t+/g ) {
        $exceed += pos($a) * ( $rOpts->{'indent-columns'} - 1 );
    }

    # Note that we just incremented output line number to future value
    # so we must subtract 1 for current line number
    if ( length($a) > 1 + $self->{_max_output_line_length} ) {
        $self->{_max_output_line_length}    = length($a) - 1;
        $self->{_max_output_line_length_at} = $self->{_output_line_number} - 1;
    }

    if ( $exceed > 0 ) {
        my $output_line_number = $self->{_output_line_number};
        $self->{_last_line_length_error}    = $exceed;
        $self->{_last_line_length_error_at} = $output_line_number - 1;
        if ( $self->{_line_length_error_count} == 0 ) {
            $self->{_first_line_length_error}    = $exceed;
            $self->{_first_line_length_error_at} = $output_line_number - 1;
        }

        if (
            $self->{_last_line_length_error} > $self->{_max_line_length_error} )
        {
            $self->{_max_line_length_error}    = $exceed;
            $self->{_max_line_length_error_at} = $output_line_number - 1;
        }

        if ( $self->{_line_length_error_count} < MAX_NAG_MESSAGES ) {
            $self->write_logfile_entry(
                "Line length exceeded by $exceed characters\n");
        }
        $self->{_line_length_error_count}++;
    }

}

sub report_line_length_errors {
    my $self                    = shift;
    my $rOpts                   = $self->{_rOpts};
    my $line_length_error_count = $self->{_line_length_error_count};
    if ( $line_length_error_count == 0 ) {
        $self->write_logfile_entry(
            "No lines exceeded $rOpts->{'maximum-line-length'} characters\n");
        my $max_output_line_length    = $self->{_max_output_line_length};
        my $max_output_line_length_at = $self->{_max_output_line_length_at};
        $self->write_logfile_entry(
"  Maximum output line length was $max_output_line_length at line $max_output_line_length_at\n"
        );

    }
    else {

        my $word = ( $line_length_error_count > 1 ) ? "s" : "";
        $self->write_logfile_entry(
"$line_length_error_count output line$word exceeded $rOpts->{'maximum-line-length'} characters:\n"
        );

        $word = ( $line_length_error_count > 1 ) ? "First" : "";
        my $first_line_length_error    = $self->{_first_line_length_error};
        my $first_line_length_error_at = $self->{_first_line_length_error_at};
        $self->write_logfile_entry(
" $word at line $first_line_length_error_at by $first_line_length_error characters\n"
        );

        if ( $line_length_error_count > 1 ) {
            my $max_line_length_error     = $self->{_max_line_length_error};
            my $max_line_length_error_at  = $self->{_max_line_length_error_at};
            my $last_line_length_error    = $self->{_last_line_length_error};
            my $last_line_length_error_at = $self->{_last_line_length_error_at};
            $self->write_logfile_entry(
" Maximum at line $max_line_length_error_at by $max_line_length_error characters\n"
            );
            $self->write_logfile_entry(
" Last at line $last_line_length_error_at by $last_line_length_error characters\n"
            );
        }
    }
}

#####################################################################
#
# The Perl::Tidy::Debugger class shows line tokenization
#
#####################################################################

package Perl::Tidy::Debugger;

sub new {

    my ( $class, $filename ) = @_;

    bless {
        _debug_file        => $filename,
        _debug_file_opened => 0,
        _fh                => undef,
    }, $class;
}

sub really_open_debug_file {

    my $self       = shift;
    my $debug_file = $self->{_debug_file};
    my $fh;
    unless ( $fh = IO::File->new("> $debug_file") ) {
        Perl::Tidy::Warn("can't open $debug_file: $!\n");
    }
    $self->{_debug_file_opened} = 1;
    $self->{_fh}                = $fh;
    print $fh
      "Use -dump-token-types (-dtt) to get a list of token type codes\n";
}

sub close_debug_file {

    my $self = shift;
    my $fh   = $self->{_fh};
    if ( $self->{_debug_file_opened} ) {

        eval { $self->{_fh}->close() };
    }
}

sub write_debug_entry {

    # This is a debug dump routine which may be modified as necessary
    # to dump tokens on a line-by-line basis.  The output will be written
    # to the .DEBUG file when the -D flag is entered.
    my $self           = shift;
    my $line_of_tokens = shift;

    my $input_line        = $line_of_tokens->{_line_text};
    my $rtoken_type       = $line_of_tokens->{_rtoken_type};
    my $rtokens           = $line_of_tokens->{_rtokens};
    my $rlevels           = $line_of_tokens->{_rlevels};
    my $rslevels          = $line_of_tokens->{_rslevels};
    my $rblock_type       = $line_of_tokens->{_rblock_type};
    my $input_line_number = $line_of_tokens->{_line_number};
    my $line_type         = $line_of_tokens->{_line_type};

    my ( $j, $num );

    my $token_str              = "$input_line_number: ";
    my $reconstructed_original = "$input_line_number: ";
    my $block_str              = "$input_line_number: ";

    #$token_str .= "$line_type: ";
    #$reconstructed_original .= "$line_type: ";

    my $pattern   = "";
    my @next_char = ( '"', '"' );
    my $i_next    = 0;
    unless ( $self->{_debug_file_opened} ) { $self->really_open_debug_file() }
    my $fh = $self->{_fh};

    for ( $j = 0 ; $j < @$rtoken_type ; $j++ ) {

        # testing patterns
        if ( $$rtoken_type[$j] eq 'k' ) {
            $pattern .= $$rtokens[$j];
        }
        else {
            $pattern .= $$rtoken_type[$j];
        }
        $reconstructed_original .= $$rtokens[$j];
        $block_str .= "($$rblock_type[$j])";
        $num = length( $$rtokens[$j] );
        my $type_str = $$rtoken_type[$j];

        # be sure there are no blank tokens (shouldn't happen)
        # This can only happen if a programming error has been made
        # because all valid tokens are non-blank
        if ( $type_str eq ' ' ) {
            print $fh "BLANK TOKEN on the next line\n";
            $type_str = $next_char[$i_next];
            $i_next   = 1 - $i_next;
        }

        if ( length($type_str) == 1 ) {
            $type_str = $type_str x $num;
        }
        $token_str .= $type_str;
    }

    # Write what you want here ...
    # print $fh "$input_line\n";
    # print $fh "$pattern\n";
    print $fh "$reconstructed_original\n";
    print $fh "$token_str\n";

    #print $fh "$block_str\n";
}

#####################################################################
#
# The Perl::Tidy::LineBuffer class supplies a 'get_line()'
# method for returning the next line to be parsed, as well as a
# 'peek_ahead()' method
#
# The input parameter is an object with a 'get_line()' method
# which returns the next line to be parsed
#
#####################################################################

package Perl::Tidy::LineBuffer;

sub new {

    my $class              = shift;
    my $line_source_object = shift;

    return bless {
        _line_source_object => $line_source_object,
        _rlookahead_buffer  => [],
    }, $class;
}

sub peek_ahead {
    my $self               = shift;
    my $buffer_index       = shift;
    my $line               = undef;
    my $line_source_object = $self->{_line_source_object};
    my $rlookahead_buffer  = $self->{_rlookahead_buffer};
    if ( $buffer_index < scalar(@$rlookahead_buffer) ) {
        $line = $$rlookahead_buffer[$buffer_index];
    }
    else {
        $line = $line_source_object->get_line();
        push( @$rlookahead_buffer, $line );
    }
    return $line;
}

sub get_line {
    my $self               = shift;
    my $line               = undef;
    my $line_source_object = $self->{_line_source_object};
    my $rlookahead_buffer  = $self->{_rlookahead_buffer};

    if ( scalar(@$rlookahead_buffer) ) {
        $line = shift @$rlookahead_buffer;
    }
    else {
        $line = $line_source_object->get_line();
    }
    return $line;
}

########################################################################
#
# the Perl::Tidy::Tokenizer package is essentially a filter which
# reads lines of perl source code from a source object and provides
# corresponding tokenized lines through its get_line() method.  Lines
# flow from the source_object to the caller like this:
#
# source_object --> LineBuffer_object --> Tokenizer -->  calling routine
#   get_line()         get_line()           get_line()     line_of_tokens
#
# The source object can be any object with a get_line() method which
# supplies one line (a character string) perl call.
# The LineBuffer object is created by the Tokenizer.
# The Tokenizer returns a reference to a data structure 'line_of_tokens'
# containing one tokenized line for each call to its get_line() method.
#
# WARNING: This is not a real class yet.  Only one tokenizer my be used.
#
########################################################################

package Perl::Tidy::Tokenizer;

BEGIN {

    # Caution: these debug flags produce a lot of output
    # They should all be 0 except when debugging small scripts

    use constant TOKENIZER_DEBUG_FLAG_EXPECT   => 0;
    use constant TOKENIZER_DEBUG_FLAG_NSCAN    => 0;
    use constant TOKENIZER_DEBUG_FLAG_QUOTE    => 0;
    use constant TOKENIZER_DEBUG_FLAG_SCAN_ID  => 0;
    use constant TOKENIZER_DEBUG_FLAG_TOKENIZE => 0;

    my $debug_warning = sub {
        print STDOUT "TOKENIZER_DEBUGGING with key $_[0]\n";
    };

    TOKENIZER_DEBUG_FLAG_EXPECT   && $debug_warning->('EXPECT');
    TOKENIZER_DEBUG_FLAG_NSCAN    && $debug_warning->('NSCAN');
    TOKENIZER_DEBUG_FLAG_QUOTE    && $debug_warning->('QUOTE');
    TOKENIZER_DEBUG_FLAG_SCAN_ID  && $debug_warning->('SCAN_ID');
    TOKENIZER_DEBUG_FLAG_TOKENIZE && $debug_warning->('TOKENIZE');

}

use Carp;

# PACKAGE VARIABLES for for processing an entire FILE.
use vars qw{
  $tokenizer_self

  $last_nonblank_token
  $last_nonblank_type
  $last_nonblank_block_type
  $statement_type
  $in_attribute_list
  $current_package
  $context

  %is_constant
  %is_user_function
  %user_function_prototype
  %is_block_function
  %is_block_list_function
  %saw_function_definition

  $brace_depth
  $paren_depth
  $square_bracket_depth

  @current_depth
  @total_depth
  $total_depth
  @nesting_sequence_number
  @current_sequence_number
  @paren_type
  @paren_semicolon_count
  @paren_structural_type
  @brace_type
  @brace_structural_type
  @brace_context
  @brace_package
  @square_bracket_type
  @square_bracket_structural_type
  @depth_array
  @nested_ternary_flag
  @nested_statement_type
  @starting_line_of_current_depth
};

# GLOBAL CONSTANTS for routines in this package
use vars qw{
  %is_indirect_object_taker
  %is_block_operator
  %expecting_operator_token
  %expecting_operator_types
  %expecting_term_types
  %expecting_term_token
  %is_digraph
  %is_file_test_operator
  %is_trigraph
  %is_valid_token_type
  %is_keyword
  %is_code_block_token
  %really_want_term
  @opening_brace_names
  @closing_brace_names
  %is_keyword_taking_list
  %is_q_qq_qw_qx_qr_s_y_tr_m
};

# possible values of operator_expected()
use constant TERM     => -1;
use constant UNKNOWN  => 0;
use constant OPERATOR => 1;

# possible values of context
use constant SCALAR_CONTEXT  => -1;
use constant UNKNOWN_CONTEXT => 0;
use constant LIST_CONTEXT    => 1;

# Maximum number of little messages; probably need not be changed.
use constant MAX_NAG_MESSAGES => 6;

{

    # methods to count instances
    my $_count = 0;
    sub get_count        { $_count; }
    sub _increment_count { ++$_count }
    sub _decrement_count { --$_count }
}

sub DESTROY {
    $_[0]->_decrement_count();
}

sub new {

    my $class = shift;

    # Note: 'tabs' and 'indent_columns' are temporary and should be
    # removed asap
    my %defaults = (
        source_object        => undef,
        debugger_object      => undef,
        diagnostics_object   => undef,
        logger_object        => undef,
        starting_level       => undef,
        indent_columns       => 4,
        tabsize              => 8,
        look_for_hash_bang   => 0,
        trim_qw              => 1,
        look_for_autoloader  => 1,
        look_for_selfloader  => 1,
        starting_line_number => 1,
    );
    my %args = ( %defaults, @_ );

    # we are given an object with a get_line() method to supply source lines
    my $source_object = $args{source_object};

    # we create another object with a get_line() and peek_ahead() method
    my $line_buffer_object = Perl::Tidy::LineBuffer->new($source_object);

    # Tokenizer state data is as follows:
    # _rhere_target_list    reference to list of here-doc targets
    # _here_doc_target      the target string for a here document
    # _here_quote_character the type of here-doc quoting (" ' ` or none)
    #                       to determine if interpolation is done
    # _quote_target         character we seek if chasing a quote
    # _line_start_quote     line where we started looking for a long quote
    # _in_here_doc          flag indicating if we are in a here-doc
    # _in_pod               flag set if we are in pod documentation
    # _in_error             flag set if we saw severe error (binary in script)
    # _in_data              flag set if we are in __DATA__ section
    # _in_end               flag set if we are in __END__ section
    # _in_format            flag set if we are in a format description
    # _in_attribute_list    flag telling if we are looking for attributes
    # _in_quote             flag telling if we are chasing a quote
    # _starting_level       indentation level of first line
    # _line_buffer_object   object with get_line() method to supply source code
    # _diagnostics_object   place to write debugging information
    # _unexpected_error_count  error count used to limit output
    # _lower_case_labels_at  line numbers where lower case labels seen
    $tokenizer_self = {
        _rhere_target_list                  => [],
        _in_here_doc                        => 0,
        _here_doc_target                    => "",
        _here_quote_character               => "",
        _in_data                            => 0,
        _in_end                             => 0,
        _in_format                          => 0,
        _in_error                           => 0,
        _in_pod                             => 0,
        _in_attribute_list                  => 0,
        _in_quote                           => 0,
        _quote_target                       => "",
        _line_start_quote                   => -1,
        _starting_level                     => $args{starting_level},
        _know_starting_level                => defined( $args{starting_level} ),
        _tabsize                            => $args{tabsize},
        _indent_columns                     => $args{indent_columns},
        _look_for_hash_bang                 => $args{look_for_hash_bang},
        _trim_qw                            => $args{trim_qw},
        _continuation_indentation           => $args{continuation_indentation},
        _outdent_labels                     => $args{outdent_labels},
        _last_line_number                   => $args{starting_line_number} - 1,
        _saw_perl_dash_P                    => 0,
        _saw_perl_dash_w                    => 0,
        _saw_use_strict                     => 0,
        _saw_v_string                       => 0,
        _look_for_autoloader                => $args{look_for_autoloader},
        _look_for_selfloader                => $args{look_for_selfloader},
        _saw_autoloader                     => 0,
        _saw_selfloader                     => 0,
        _saw_hash_bang                      => 0,
        _saw_end                            => 0,
        _saw_data                           => 0,
        _saw_negative_indentation           => 0,
        _started_tokenizing                 => 0,
        _line_buffer_object                 => $line_buffer_object,
        _debugger_object                    => $args{debugger_object},
        _diagnostics_object                 => $args{diagnostics_object},
        _logger_object                      => $args{logger_object},
        _unexpected_error_count             => 0,
        _started_looking_for_here_target_at => 0,
        _nearly_matched_here_target_at      => undef,
        _line_text                          => "",
        _rlower_case_labels_at              => undef,
    };

    prepare_for_a_new_file();
    find_starting_indentation_level();

    bless $tokenizer_self, $class;

    # This is not a full class yet, so die if an attempt is made to
    # create more than one object.

    if ( _increment_count() > 1 ) {
        confess
"Attempt to create more than 1 object in $class, which is not a true class yet\n";
    }

    return $tokenizer_self;

}

# interface to Perl::Tidy::Logger routines
sub warning {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->warning(@_);
    }
}

sub complain {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->complain(@_);
    }
}

sub write_logfile_entry {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->write_logfile_entry(@_);
    }
}

sub interrupt_logfile {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->interrupt_logfile();
    }
}

sub resume_logfile {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->resume_logfile();
    }
}

sub increment_brace_error {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->increment_brace_error();
    }
}

sub report_definite_bug {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->report_definite_bug();
    }
}

sub brace_warning {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->brace_warning(@_);
    }
}

sub get_saw_brace_error {
    my $logger_object = $tokenizer_self->{_logger_object};
    if ($logger_object) {
        $logger_object->get_saw_brace_error();
    }
    else {
        0;
    }
}

# interface to Perl::Tidy::Diagnostics routines
sub write_diagnostics {
    if ( $tokenizer_self->{_diagnostics_object} ) {
        $tokenizer_self->{_diagnostics_object}->write_diagnostics(@_);
    }
}

sub report_tokenization_errors {

    my $self = shift;

    my $level = get_indentation_level();
    if ( $level != $tokenizer_self->{_starting_level} ) {
        warning("final indentation level: $level\n");
    }

    check_final_nesting_depths();

    if ( $tokenizer_self->{_look_for_hash_bang}
        && !$tokenizer_self->{_saw_hash_bang} )
    {
        warning(
            "hit EOF without seeing hash-bang line; maybe don't need -x?\n");
    }

    if ( $tokenizer_self->{_in_format} ) {
        warning("hit EOF while in format description\n");
    }

    if ( $tokenizer_self->{_in_pod} ) {

        # Just write log entry if this is after __END__ or __DATA__
        # because this happens to often, and it is not likely to be
        # a parsing error.
        if ( $tokenizer_self->{_saw_data} || $tokenizer_self->{_saw_end} ) {
            write_logfile_entry(
"hit eof while in pod documentation (no =cut seen)\n\tthis can cause trouble with some pod utilities\n"
            );
        }

        else {
            complain(
"hit eof while in pod documentation (no =cut seen)\n\tthis can cause trouble with some pod utilities\n"
            );
        }

    }

    if ( $tokenizer_self->{_in_here_doc} ) {
        my $here_doc_target = $tokenizer_self->{_here_doc_target};
        my $started_looking_for_here_target_at =
          $tokenizer_self->{_started_looking_for_here_target_at};
        if ($here_doc_target) {
            warning(
"hit EOF in here document starting at line $started_looking_for_here_target_at with target: $here_doc_target\n"
            );
        }
        else {
            warning(
"hit EOF in here document starting at line $started_looking_for_here_target_at with empty target string\n"
            );
        }
        my $nearly_matched_here_target_at =
          $tokenizer_self->{_nearly_matched_here_target_at};
        if ($nearly_matched_here_target_at) {
            warning(
"NOTE: almost matched at input line $nearly_matched_here_target_at except for whitespace\n"
            );
        }
    }

    if ( $tokenizer_self->{_in_quote} ) {
        my $line_start_quote = $tokenizer_self->{_line_start_quote};
        my $quote_target     = $tokenizer_self->{_quote_target};
        my $what =
          ( $tokenizer_self->{_in_attribute_list} )
          ? "attribute list"
          : "quote/pattern";
        warning(
"hit EOF seeking end of $what starting at line $line_start_quote ending in $quote_target\n"
        );
    }

    unless ( $tokenizer_self->{_saw_perl_dash_w} ) {
        if ( $] < 5.006 ) {
            write_logfile_entry("Suggest including '-w parameter'\n");
        }
        else {
            write_logfile_entry("Suggest including 'use warnings;'\n");
        }
    }

    if ( $tokenizer_self->{_saw_perl_dash_P} ) {
        write_logfile_entry("Use of -P parameter for defines is discouraged\n");
    }

    unless ( $tokenizer_self->{_saw_use_strict} ) {
        write_logfile_entry("Suggest including 'use strict;'\n");
    }

    # it is suggested that lables have at least one upper case character
    # for legibility and to avoid code breakage as new keywords are introduced
    if ( $tokenizer_self->{_rlower_case_labels_at} ) {
        my @lower_case_labels_at =
          @{ $tokenizer_self->{_rlower_case_labels_at} };
        write_logfile_entry(
            "Suggest using upper case characters in label(s)\n");
        local $" = ')(';
        write_logfile_entry("  defined at line(s): (@lower_case_labels_at)\n");
    }
}

sub report_v_string {

    # warn if this version can't handle v-strings
    my $tok = shift;
    unless ( $tokenizer_self->{_saw_v_string} ) {
        $tokenizer_self->{_saw_v_string} = $tokenizer_self->{_last_line_number};
    }
    if ( $] < 5.006 ) {
        warning(
"Found v-string '$tok' but v-strings are not implemented in your version of perl; see Camel 3 book ch 2\n"
        );
    }
}

sub get_input_line_number {
    return $tokenizer_self->{_last_line_number};
}

# returns the next tokenized line
sub get_line {

    my $self = shift;

    # USES GLOBAL VARIABLES: $tokenizer_self, $brace_depth,
    # $square_bracket_depth, $paren_depth

    my $input_line = $tokenizer_self->{_line_buffer_object}->get_line();
    $tokenizer_self->{_line_text} = $input_line;

    return undef unless ($input_line);

    my $input_line_number = ++$tokenizer_self->{_last_line_number};

    # Find and remove what characters terminate this line, including any
    # control r
    my $input_line_separator = "";
    if ( chomp($input_line) ) { $input_line_separator = $/ }

    # TODO: what other characters should be included here?
    if ( $input_line =~ s/((\r|\035|\032)+)$// ) {
        $input_line_separator = $2 . $input_line_separator;
    }

    # for backwards compatibility we keep the line text terminated with
    # a newline character
    $input_line .= "\n";
    $tokenizer_self->{_line_text} = $input_line;    # update

    # create a data structure describing this line which will be
    # returned to the caller.

    # _line_type codes are:
    #   SYSTEM         - system-specific code before hash-bang line
    #   CODE           - line of perl code (including comments)
    #   POD_START      - line starting pod, such as '=head'
    #   POD            - pod documentation text
    #   POD_END        - last line of pod section, '=cut'
    #   HERE           - text of here-document
    #   HERE_END       - last line of here-doc (target word)
    #   FORMAT         - format section
    #   FORMAT_END     - last line of format section, '.'
    #   DATA_START     - __DATA__ line
    #   DATA           - unidentified text following __DATA__
    #   END_START      - __END__ line
    #   END            - unidentified text following __END__
    #   ERROR          - we are in big trouble, probably not a perl script

    # Other variables:
    #   _curly_brace_depth     - depth of curly braces at start of line
    #   _square_bracket_depth  - depth of square brackets at start of line
    #   _paren_depth           - depth of parens at start of line
    #   _starting_in_quote     - this line continues a multi-line quote
    #                            (so don't trim leading blanks!)
    #   _ending_in_quote       - this line ends in a multi-line quote
    #                            (so don't trim trailing blanks!)
    my $line_of_tokens = {
        _line_type                 => 'EOF',
        _line_text                 => $input_line,
        _line_number               => $input_line_number,
        _rtoken_type               => undef,
        _rtokens                   => undef,
        _rlevels                   => undef,
        _rslevels                  => undef,
        _rblock_type               => undef,
        _rcontainer_type           => undef,
        _rcontainer_environment    => undef,
        _rtype_sequence            => undef,
        _rnesting_tokens           => undef,
        _rci_levels                => undef,
        _rnesting_blocks           => undef,
        _guessed_indentation_level => 0,
        _starting_in_quote    => 0,                    # to be set by subroutine
        _ending_in_quote      => 0,
        _curly_brace_depth    => $brace_depth,
        _square_bracket_depth => $square_bracket_depth,
        _paren_depth          => $paren_depth,
        _quote_character      => '',
    };

    # must print line unchanged if we are in a here document
    if ( $tokenizer_self->{_in_here_doc} ) {

        $line_of_tokens->{_line_type} = 'HERE';
        my $here_doc_target      = $tokenizer_self->{_here_doc_target};
        my $here_quote_character = $tokenizer_self->{_here_quote_character};
        my $candidate_target     = $input_line;
        chomp $candidate_target;
        if ( $candidate_target eq $here_doc_target ) {
            $tokenizer_self->{_nearly_matched_here_target_at} = undef;
            $line_of_tokens->{_line_type}                     = 'HERE_END';
            write_logfile_entry("Exiting HERE document $here_doc_target\n");

            my $rhere_target_list = $tokenizer_self->{_rhere_target_list};
            if (@$rhere_target_list) {    # there can be multiple here targets
                ( $here_doc_target, $here_quote_character ) =
                  @{ shift @$rhere_target_list };
                $tokenizer_self->{_here_doc_target} = $here_doc_target;
                $tokenizer_self->{_here_quote_character} =
                  $here_quote_character;
                write_logfile_entry(
                    "Entering HERE document $here_doc_target\n");
                $tokenizer_self->{_nearly_matched_here_target_at} = undef;
                $tokenizer_self->{_started_looking_for_here_target_at} =
                  $input_line_number;
            }
            else {
                $tokenizer_self->{_in_here_doc}          = 0;
                $tokenizer_self->{_here_doc_target}      = "";
                $tokenizer_self->{_here_quote_character} = "";
            }
        }

        # check for error of extra whitespace
        # note for PERL6: leading whitespace is allowed
        else {
            $candidate_target =~ s/\s*$//;
            $candidate_target =~ s/^\s*//;
            if ( $candidate_target eq $here_doc_target ) {
                $tokenizer_self->{_nearly_matched_here_target_at} =
                  $input_line_number;
            }
        }
        return $line_of_tokens;
    }

    # must print line unchanged if we are in a format section
    elsif ( $tokenizer_self->{_in_format} ) {

        if ( $input_line =~ /^\.[\s#]*$/ ) {
            write_logfile_entry("Exiting format section\n");
            $tokenizer_self->{_in_format} = 0;
            $line_of_tokens->{_line_type} = 'FORMAT_END';
        }
        else {
            $line_of_tokens->{_line_type} = 'FORMAT';
        }
        return $line_of_tokens;
    }

    # must print line unchanged if we are in pod documentation
    elsif ( $tokenizer_self->{_in_pod} ) {

        $line_of_tokens->{_line_type} = 'POD';
        if ( $input_line =~ /^=cut/ ) {
            $line_of_tokens->{_line_type} = 'POD_END';
            write_logfile_entry("Exiting POD section\n");
            $tokenizer_self->{_in_pod} = 0;
        }
        if ( $input_line =~ /^\#\!.*perl\b/ ) {
            warning(
                "Hash-bang in pod can cause older versions of perl to fail! \n"
            );
        }

        return $line_of_tokens;
    }

    # must print line unchanged if we have seen a severe error (i.e., we
    # are seeing illegal tokens and connot continue.  Syntax errors do
    # not pass this route).  Calling routine can decide what to do, but
    # the default can be to just pass all lines as if they were after __END__
    elsif ( $tokenizer_self->{_in_error} ) {
        $line_of_tokens->{_line_type} = 'ERROR';
        return $line_of_tokens;
    }

    # print line unchanged if we are __DATA__ section
    elsif ( $tokenizer_self->{_in_data} ) {

        # ...but look for POD
        # Note that the _in_data and _in_end flags remain set
        # so that we return to that state after seeing the
        # end of a pod section
        if ( $input_line =~ /^=(?!cut)/ ) {
            $line_of_tokens->{_line_type} = 'POD_START';
            write_logfile_entry("Entering POD section\n");
            $tokenizer_self->{_in_pod} = 1;
            return $line_of_tokens;
        }
        else {
            $line_of_tokens->{_line_type} = 'DATA';
            return $line_of_tokens;
        }
    }

    # print line unchanged if we are in __END__ section
    elsif ( $tokenizer_self->{_in_end} ) {

        # ...but look for POD
        # Note that the _in_data and _in_end flags remain set
        # so that we return to that state after seeing the
        # end of a pod section
        if ( $input_line =~ /^=(?!cut)/ ) {
            $line_of_tokens->{_line_type} = 'POD_START';
            write_logfile_entry("Entering POD section\n");
            $tokenizer_self->{_in_pod} = 1;
            return $line_of_tokens;
        }
        else {
            $line_of_tokens->{_line_type} = 'END';
            return $line_of_tokens;
        }
    }

    # check for a hash-bang line if we haven't seen one
    if ( !$tokenizer_self->{_saw_hash_bang} ) {
        if ( $input_line =~ /^\#\!.*perl\b/ ) {
            $tokenizer_self->{_saw_hash_bang} = $input_line_number;

            # check for -w and -P flags
            if ( $input_line =~ /^\#\!.*perl\s.*-.*P/ ) {
                $tokenizer_self->{_saw_perl_dash_P} = 1;
            }

            if ( $input_line =~ /^\#\!.*perl\s.*-.*w/ ) {
                $tokenizer_self->{_saw_perl_dash_w} = 1;
            }

            if (   ( $input_line_number > 1 )
                && ( !$tokenizer_self->{_look_for_hash_bang} ) )
            {

                # this is helpful for VMS systems; we may have accidentally
                # tokenized some DCL commands
                if ( $tokenizer_self->{_started_tokenizing} ) {
                    warning(
"There seems to be a hash-bang after line 1; do you need to run with -x ?\n"
                    );
                }
                else {
                    complain("Useless hash-bang after line 1\n");
                }
            }

            # Report the leading hash-bang as a system line
            # This will prevent -dac from deleting it
            else {
                $line_of_tokens->{_line_type} = 'SYSTEM';
                return $line_of_tokens;
            }
        }
    }

    # wait for a hash-bang before parsing if the user invoked us with -x
    if ( $tokenizer_self->{_look_for_hash_bang}
        && !$tokenizer_self->{_saw_hash_bang} )
    {
        $line_of_tokens->{_line_type} = 'SYSTEM';
        return $line_of_tokens;
    }

    # a first line of the form ': #' will be marked as SYSTEM
    # since lines of this form may be used by tcsh
    if ( $input_line_number == 1 && $input_line =~ /^\s*\:\s*\#/ ) {
        $line_of_tokens->{_line_type} = 'SYSTEM';
        return $line_of_tokens;
    }

    # now we know that it is ok to tokenize the line...
    # the line tokenizer will modify any of these private variables:
    #        _rhere_target_list
    #        _in_data
    #        _in_end
    #        _in_format
    #        _in_error
    #        _in_pod
    #        _in_quote
    my $ending_in_quote_last = $tokenizer_self->{_in_quote};
    tokenize_this_line($line_of_tokens);

    # Now finish defining the return structure and return it
    $line_of_tokens->{_ending_in_quote} = $tokenizer_self->{_in_quote};

    # handle severe error (binary data in script)
    if ( $tokenizer_self->{_in_error} ) {
        $tokenizer_self->{_in_quote} = 0;    # to avoid any more messages
        warning("Giving up after error\n");
        $line_of_tokens->{_line_type} = 'ERROR';
        reset_indentation_level(0);          # avoid error messages
        return $line_of_tokens;
    }

    # handle start of pod documentation
    if ( $tokenizer_self->{_in_pod} ) {

        # This gets tricky..above a __DATA__ or __END__ section, perl
        # accepts '=cut' as the start of pod section. But afterwards,
        # only pod utilities see it and they may ignore an =cut without
        # leading =head.  In any case, this isn't good.
        if ( $input_line =~ /^=cut\b/ ) {
            if ( $tokenizer_self->{_saw_data} || $tokenizer_self->{_saw_end} ) {
                complain("=cut while not in pod ignored\n");
                $tokenizer_self->{_in_pod}    = 0;
                $line_of_tokens->{_line_type} = 'POD_END';
            }
            else {
                $line_of_tokens->{_line_type} = 'POD_START';
                complain(
"=cut starts a pod section .. this can fool pod utilities.\n"
                );
                write_logfile_entry("Entering POD section\n");
            }
        }

        else {
            $line_of_tokens->{_line_type} = 'POD_START';
            write_logfile_entry("Entering POD section\n");
        }

        return $line_of_tokens;
    }

    # update indentation levels for log messages
    if ( $input_line !~ /^\s*$/ ) {
        my $rlevels = $line_of_tokens->{_rlevels};
        $line_of_tokens->{_guessed_indentation_level} =
          guess_old_indentation_level($input_line);
    }

    # see if this line contains here doc targets
    my $rhere_target_list = $tokenizer_self->{_rhere_target_list};
    if (@$rhere_target_list) {

        my ( $here_doc_target, $here_quote_character ) =
          @{ shift @$rhere_target_list };
        $tokenizer_self->{_in_here_doc}          = 1;
        $tokenizer_self->{_here_doc_target}      = $here_doc_target;
        $tokenizer_self->{_here_quote_character} = $here_quote_character;
        write_logfile_entry("Entering HERE document $here_doc_target\n");
        $tokenizer_self->{_started_looking_for_here_target_at} =
          $input_line_number;
    }

    # NOTE: __END__ and __DATA__ statements are written unformatted
    # because they can theoretically contain additional characters
    # which are not tokenized (and cannot be read with <DATA> either!).
    if ( $tokenizer_self->{_in_data} ) {
        $line_of_tokens->{_line_type} = 'DATA_START';
        write_logfile_entry("Starting __DATA__ section\n");
        $tokenizer_self->{_saw_data} = 1;

        # keep parsing after __DATA__ if use SelfLoader was seen
        if ( $tokenizer_self->{_saw_selfloader} ) {
            $tokenizer_self->{_in_data} = 0;
            write_logfile_entry(
                "SelfLoader seen, continuing; -nlsl deactivates\n");
        }

        return $line_of_tokens;
    }

    elsif ( $tokenizer_self->{_in_end} ) {
        $line_of_tokens->{_line_type} = 'END_START';
        write_logfile_entry("Starting __END__ section\n");
        $tokenizer_self->{_saw_end} = 1;

        # keep parsing after __END__ if use AutoLoader was seen
        if ( $tokenizer_self->{_saw_autoloader} ) {
            $tokenizer_self->{_in_end} = 0;
            write_logfile_entry(
                "AutoLoader seen, continuing; -nlal deactivates\n");
        }
        return $line_of_tokens;
    }

    # now, finally, we know that this line is type 'CODE'
    $line_of_tokens->{_line_type} = 'CODE';

    # remember if we have seen any real code
    if (  !$tokenizer_self->{_started_tokenizing}
        && $input_line !~ /^\s*$/
        && $input_line !~ /^\s*#/ )
    {
        $tokenizer_self->{_started_tokenizing} = 1;
    }

    if ( $tokenizer_self->{_debugger_object} ) {
        $tokenizer_self->{_debugger_object}->write_debug_entry($line_of_tokens);
    }

    # Note: if keyword 'format' occurs in this line code, it is still CODE
    # (keyword 'format' need not start a line)
    if ( $tokenizer_self->{_in_format} ) {
        write_logfile_entry("Entering format section\n");
    }

    if ( $tokenizer_self->{_in_quote}
        and ( $tokenizer_self->{_line_start_quote} < 0 ) )
    {

        #if ( ( my $quote_target = get_quote_target() ) !~ /^\s*$/ ) {
        if (
            ( my $quote_target = $tokenizer_self->{_quote_target} ) !~ /^\s*$/ )
        {
            $tokenizer_self->{_line_start_quote} = $input_line_number;
            write_logfile_entry(
                "Start multi-line quote or pattern ending in $quote_target\n");
        }
    }
    elsif ( ( $tokenizer_self->{_line_start_quote} >= 0 )
        and !$tokenizer_self->{_in_quote} )
    {
        $tokenizer_self->{_line_start_quote} = -1;
        write_logfile_entry("End of multi-line quote or pattern\n");
    }

    # we are returning a line of CODE
    return $line_of_tokens;
}

sub find_starting_indentation_level {

    # We need to find the indentation level of the first line of the
    # script being formatted.  Often it will be zero for an entire file,
    # but if we are formatting a local block of code (within an editor for
    # example) it may not be zero.  The user may specify this with the
    # -sil=n parameter but normally doesn't so we have to guess.
    #
    # USES GLOBAL VARIABLES: $tokenizer_self
    my $starting_level = 0;

    # use value if given as parameter
    if ( $tokenizer_self->{_know_starting_level} ) {
        $starting_level = $tokenizer_self->{_starting_level};
    }

    # if we know there is a hash_bang line, the level must be zero
    elsif ( $tokenizer_self->{_look_for_hash_bang} ) {
        $tokenizer_self->{_know_starting_level} = 1;
    }

    # otherwise figure it out from the input file
    else {
        my $line;
        my $i = 0;

        # keep looking at lines until we find a hash bang or piece of code
        my $msg = "";
        while ( $line =
            $tokenizer_self->{_line_buffer_object}->peek_ahead( $i++ ) )
        {

            # if first line is #! then assume starting level is zero
            if ( $i == 1 && $line =~ /^\#\!/ ) {
                $starting_level = 0;
                last;
            }
            next if ( $line =~ /^\s*#/ );    # skip past comments
            next if ( $line =~ /^\s*$/ );    # skip past blank lines
            $starting_level = guess_old_indentation_level($line);
            last;
        }
        $msg = "Line $i implies starting-indentation-level = $starting_level\n";
        write_logfile_entry("$msg");
    }
    $tokenizer_self->{_starting_level} = $starting_level;
    reset_indentation_level($starting_level);
}

sub guess_old_indentation_level {
    my ($line) = @_;

    # Guess the indentation level of an input line.
    #
    # For the first line of code this result will define the starting
    # indentation level.  It will mainly be non-zero when perltidy is applied
    # within an editor to a local block of code.
    #
    # This is an impossible task in general because we can't know what tabs
    # meant for the old script and how many spaces were used for one
    # indentation level in the given input script.  For example it may have
    # been previously formatted with -i=7 -et=3.  But we can at least try to
    # make sure that perltidy guesses correctly if it is applied repeatedly to
    # a block of code within an editor, so that the block stays at the same
    # level when perltidy is applied repeatedly.
    #
    # USES GLOBAL VARIABLES: $tokenizer_self
    my $level = 0;

    # find leading tabs, spaces, and any statement label
    my $spaces = 0;
    if ( $line =~ /^(\t+)?(\s+)?(\w+:[^:])?/ ) {

        # If there are leading tabs, we use the tab scheme for this run, if
        # any, so that the code will remain stable when editing.
        if ($1) { $spaces += length($1) * $tokenizer_self->{_tabsize} }

        if ($2) { $spaces += length($2) }

        # correct for outdented labels
        if ( $3 && $tokenizer_self->{'_outdent_labels'} ) {
            $spaces += $tokenizer_self->{_continuation_indentation};
        }
    }

    # compute indentation using the value of -i for this run.
    # If -i=0 is used for this run (which is possible) it doesn't matter
    # what we do here but we'll guess that the old run used 4 spaces per level.
    my $indent_columns = $tokenizer_self->{_indent_columns};
    $indent_columns = 4 if ( !$indent_columns );
    $level = int( $spaces / $indent_columns );
    return ($level);
}

# This is a currently unused debug routine
sub dump_functions {

    my $fh = *STDOUT;
    my ( $pkg, $sub );
    foreach $pkg ( keys %is_user_function ) {
        print $fh "\nnon-constant subs in package $pkg\n";

        foreach $sub ( keys %{ $is_user_function{$pkg} } ) {
            my $msg = "";
            if ( $is_block_list_function{$pkg}{$sub} ) {
                $msg = 'block_list';
            }

            if ( $is_block_function{$pkg}{$sub} ) {
                $msg = 'block';
            }
            print $fh "$sub $msg\n";
        }
    }

    foreach $pkg ( keys %is_constant ) {
        print $fh "\nconstants and constant subs in package $pkg\n";

        foreach $sub ( keys %{ $is_constant{$pkg} } ) {
            print $fh "$sub\n";
        }
    }
}

sub ones_count {

    # count number of 1's in a string of 1's and 0's
    # example: ones_count("010101010101") gives 6
    return ( my $cis = $_[0] ) =~ tr/1/0/;
}

sub prepare_for_a_new_file {

    # previous tokens needed to determine what to expect next
    $last_nonblank_token      = ';';    # the only possible starting state which
    $last_nonblank_type       = ';';    # will make a leading brace a code block
    $last_nonblank_block_type = '';

    # scalars for remembering statement types across multiple lines
    $statement_type    = '';            # '' or 'use' or 'sub..' or 'case..'
    $in_attribute_list = 0;

    # scalars for remembering where we are in the file
    $current_package = "main";
    $context         = UNKNOWN_CONTEXT;

    # hashes used to remember function information
    %is_constant             = ();      # user-defined constants
    %is_user_function        = ();      # user-defined functions
    %user_function_prototype = ();      # their prototypes
    %is_block_function       = ();
    %is_block_list_function  = ();
    %saw_function_definition = ();

    # variables used to track depths of various containers
    # and report nesting errors
    $paren_depth          = 0;
    $brace_depth          = 0;
    $square_bracket_depth = 0;
    @current_depth[ 0 .. $#closing_brace_names ] =
      (0) x scalar @closing_brace_names;
    $total_depth = 0;
    @total_depth = ();
    @nesting_sequence_number[ 0 .. $#closing_brace_names ] =
      ( 0 .. $#closing_brace_names );
    @current_sequence_number             = ();
    $paren_type[$paren_depth]            = '';
    $paren_semicolon_count[$paren_depth] = 0;
    $paren_structural_type[$brace_depth] = '';
    $brace_type[$brace_depth] = ';';    # identify opening brace as code block
    $brace_structural_type[$brace_depth]                   = '';
    $brace_context[$brace_depth]                           = UNKNOWN_CONTEXT;
    $brace_package[$paren_depth]                           = $current_package;
    $square_bracket_type[$square_bracket_depth]            = '';
    $square_bracket_structural_type[$square_bracket_depth] = '';

    initialize_tokenizer_state();
}

{                                       # begin tokenize_this_line

    use constant BRACE          => 0;
    use constant SQUARE_BRACKET => 1;
    use constant PAREN          => 2;
    use constant QUESTION_COLON => 3;

    # TV1: scalars for processing one LINE.
    # Re-initialized on each entry to sub tokenize_this_line.
    my (
        $block_type,        $container_type,    $expecting,
        $i,                 $i_tok,             $input_line,
        $input_line_number, $last_nonblank_i,   $max_token_index,
        $next_tok,          $next_type,         $peeked_ahead,
        $prototype,         $rhere_target_list, $rtoken_map,
        $rtoken_type,       $rtokens,           $tok,
        $type,              $type_sequence,     $indent_flag,
    );

    # TV2: refs to ARRAYS for processing one LINE
    # Re-initialized on each call.
    my $routput_token_list     = [];    # stack of output token indexes
    my $routput_token_type     = [];    # token types
    my $routput_block_type     = [];    # types of code block
    my $routput_container_type = [];    # paren types, such as if, elsif, ..
    my $routput_type_sequence  = [];    # nesting sequential number
    my $routput_indent_flag    = [];    #

    # TV3: SCALARS for quote variables.  These are initialized with a
    # subroutine call and continually updated as lines are processed.
    my ( $in_quote, $quote_type, $quote_character, $quote_pos, $quote_depth,
        $quoted_string_1, $quoted_string_2, $allowed_quote_modifiers, );

    # TV4: SCALARS for multi-line identifiers and
    # statements. These are initialized with a subroutine call
    # and continually updated as lines are processed.
    my ( $id_scan_state, $identifier, $want_paren, $indented_if_level );

    # TV5: SCALARS for tracking indentation level.
    # Initialized once and continually updated as lines are
    # processed.
    my (
        $nesting_token_string,      $nesting_type_string,
        $nesting_block_string,      $nesting_block_flag,
        $nesting_list_string,       $nesting_list_flag,
        $ci_string_in_tokenizer,    $continuation_string_in_tokenizer,
        $in_statement_continuation, $level_in_tokenizer,
        $slevel_in_tokenizer,       $rslevel_stack,
    );

    # TV6: SCALARS for remembering several previous
    # tokens. Initialized once and continually updated as
    # lines are processed.
    my (
        $last_nonblank_container_type,     $last_nonblank_type_sequence,
        $last_last_nonblank_token,         $last_last_nonblank_type,
        $last_last_nonblank_block_type,    $last_last_nonblank_container_type,
        $last_last_nonblank_type_sequence, $last_nonblank_prototype,
    );

    # ----------------------------------------------------------------
    # beginning of tokenizer variable access and manipulation routines
    # ----------------------------------------------------------------

    sub initialize_tokenizer_state {

        # TV1: initialized on each call
        # TV2: initialized on each call
        # TV3:
        $in_quote                = 0;
        $quote_type              = 'Q';
        $quote_character         = "";
        $quote_pos               = 0;
        $quote_depth             = 0;
        $quoted_string_1         = "";
        $quoted_string_2         = "";
        $allowed_quote_modifiers = "";

        # TV4:
        $id_scan_state     = '';
        $identifier        = '';
        $want_paren        = "";
        $indented_if_level = 0;

        # TV5:
        $nesting_token_string             = "";
        $nesting_type_string              = "";
        $nesting_block_string             = '1';    # initially in a block
        $nesting_block_flag               = 1;
        $nesting_list_string              = '0';    # initially not in a list
        $nesting_list_flag                = 0;      # initially not in a list
        $ci_string_in_tokenizer           = "";
        $continuation_string_in_tokenizer = "0";
        $in_statement_continuation        = 0;
        $level_in_tokenizer               = 0;
        $slevel_in_tokenizer              = 0;
        $rslevel_stack                    = [];

        # TV6:
        $last_nonblank_container_type      = '';
        $last_nonblank_type_sequence       = '';
        $last_last_nonblank_token          = ';';
        $last_last_nonblank_type           = ';';
        $last_last_nonblank_block_type     = '';
        $last_last_nonblank_container_type = '';
        $last_last_nonblank_type_sequence  = '';
        $last_nonblank_prototype           = "";
    }

    sub save_tokenizer_state {

        my $rTV1 = [
            $block_type,        $container_type,    $expecting,
            $i,                 $i_tok,             $input_line,
            $input_line_number, $last_nonblank_i,   $max_token_index,
            $next_tok,          $next_type,         $peeked_ahead,
            $prototype,         $rhere_target_list, $rtoken_map,
            $rtoken_type,       $rtokens,           $tok,
            $type,              $type_sequence,     $indent_flag,
        ];

        my $rTV2 = [
            $routput_token_list,    $routput_token_type,
            $routput_block_type,    $routput_container_type,
            $routput_type_sequence, $routput_indent_flag,
        ];

        my $rTV3 = [
            $in_quote,        $quote_type,
            $quote_character, $quote_pos,
            $quote_depth,     $quoted_string_1,
            $quoted_string_2, $allowed_quote_modifiers,
        ];

        my $rTV4 =
          [ $id_scan_state, $identifier, $want_paren, $indented_if_level ];

        my $rTV5 = [
            $nesting_token_string,      $nesting_type_string,
            $nesting_block_string,      $nesting_block_flag,
            $nesting_list_string,       $nesting_list_flag,
            $ci_string_in_tokenizer,    $continuation_string_in_tokenizer,
            $in_statement_continuation, $level_in_tokenizer,
            $slevel_in_tokenizer,       $rslevel_stack,
        ];

        my $rTV6 = [
            $last_nonblank_container_type,
            $last_nonblank_type_sequence,
            $last_last_nonblank_token,
            $last_last_nonblank_type,
            $last_last_nonblank_block_type,
            $last_last_nonblank_container_type,
            $last_last_nonblank_type_sequence,
            $last_nonblank_prototype,
        ];
        return [ $rTV1, $rTV2, $rTV3, $rTV4, $rTV5, $rTV6 ];
    }

    sub restore_tokenizer_state {
        my ($rstate) = @_;
        my ( $rTV1, $rTV2, $rTV3, $rTV4, $rTV5, $rTV6 ) = @{$rstate};
        (
            $block_type,        $container_type,    $expecting,
            $i,                 $i_tok,             $input_line,
            $input_line_number, $last_nonblank_i,   $max_token_index,
            $next_tok,          $next_type,         $peeked_ahead,
            $prototype,         $rhere_target_list, $rtoken_map,
            $rtoken_type,       $rtokens,           $tok,
            $type,              $type_sequence,     $indent_flag,
        ) = @{$rTV1};

        (
            $routput_token_list,    $routput_token_type,
            $routput_block_type,    $routput_container_type,
            $routput_type_sequence, $routput_type_sequence,
        ) = @{$rTV2};

        (
            $in_quote, $quote_type, $quote_character, $quote_pos, $quote_depth,
            $quoted_string_1, $quoted_string_2, $allowed_quote_modifiers,
        ) = @{$rTV3};

        ( $id_scan_state, $identifier, $want_paren, $indented_if_level ) =
          @{$rTV4};

        (
            $nesting_token_string,      $nesting_type_string,
            $nesting_block_string,      $nesting_block_flag,
            $nesting_list_string,       $nesting_list_flag,
            $ci_string_in_tokenizer,    $continuation_string_in_tokenizer,
            $in_statement_continuation, $level_in_tokenizer,
            $slevel_in_tokenizer,       $rslevel_stack,
        ) = @{$rTV5};

        (
            $last_nonblank_container_type,
            $last_nonblank_type_sequence,
            $last_last_nonblank_token,
            $last_last_nonblank_type,
            $last_last_nonblank_block_type,
            $last_last_nonblank_container_type,
            $last_last_nonblank_type_sequence,
            $last_nonblank_prototype,
        ) = @{$rTV6};
    }

    sub get_indentation_level {

        # patch to avoid reporting error if indented if is not terminated
        if ($indented_if_level) { return $level_in_tokenizer - 1 }
        return $level_in_tokenizer;
    }

    sub reset_indentation_level {
        $level_in_tokenizer  = $_[0];
        $slevel_in_tokenizer = $_[0];
        push @{$rslevel_stack}, $slevel_in_tokenizer;
    }

    sub peeked_ahead {
        $peeked_ahead = defined( $_[0] ) ? $_[0] : $peeked_ahead;
    }

    # ------------------------------------------------------------
    # end of tokenizer variable access and manipulation routines
    # ------------------------------------------------------------

    # ------------------------------------------------------------
    # beginning of various scanner interface routines
    # ------------------------------------------------------------
    sub scan_replacement_text {

        # check for here-docs in replacement text invoked by
        # a substitution operator with executable modifier 'e'.
        #
        # given:
        #  $replacement_text
        # return:
        #  $rht = reference to any here-doc targets
        my ($replacement_text) = @_;

        # quick check
        return undef unless ( $replacement_text =~ /<</ );

        write_logfile_entry("scanning replacement text for here-doc targets\n");

        # save the logger object for error messages
        my $logger_object = $tokenizer_self->{_logger_object};

        # localize all package variables
        local (
            $tokenizer_self,                 $last_nonblank_token,
            $last_nonblank_type,             $last_nonblank_block_type,
            $statement_type,                 $in_attribute_list,
            $current_package,                $context,
            %is_constant,                    %is_user_function,
            %user_function_prototype,        %is_block_function,
            %is_block_list_function,         %saw_function_definition,
            $brace_depth,                    $paren_depth,
            $square_bracket_depth,           @current_depth,
            @total_depth,                    $total_depth,
            @nesting_sequence_number,        @current_sequence_number,
            @paren_type,                     @paren_semicolon_count,
            @paren_structural_type,          @brace_type,
            @brace_structural_type,          @brace_context,
            @brace_package,                  @square_bracket_type,
            @square_bracket_structural_type, @depth_array,
            @starting_line_of_current_depth, @nested_ternary_flag,
            @nested_statement_type,
        );

        # save all lexical variables
        my $rstate = save_tokenizer_state();
        _decrement_count();    # avoid error check for multiple tokenizers

        # make a new tokenizer
        my $rOpts = {};
        my $rpending_logfile_message;
        my $source_object =
          Perl::Tidy::LineSource->new( \$replacement_text, $rOpts,
            $rpending_logfile_message );
        my $tokenizer = Perl::Tidy::Tokenizer->new(
            source_object        => $source_object,
            logger_object        => $logger_object,
            starting_line_number => $input_line_number,
        );

        # scan the replacement text
        1 while ( $tokenizer->get_line() );

        # remove any here doc targets
        my $rht = undef;
        if ( $tokenizer_self->{_in_here_doc} ) {
            $rht = [];
            push @{$rht},
              [
                $tokenizer_self->{_here_doc_target},
                $tokenizer_self->{_here_quote_character}
              ];
            if ( $tokenizer_self->{_rhere_target_list} ) {
                push @{$rht}, @{ $tokenizer_self->{_rhere_target_list} };
                $tokenizer_self->{_rhere_target_list} = undef;
            }
            $tokenizer_self->{_in_here_doc} = undef;
        }

        # now its safe to report errors
        $tokenizer->report_tokenization_errors();

        # restore all tokenizer lexical variables
        restore_tokenizer_state($rstate);

        # return the here doc targets
        return $rht;
    }

    sub scan_bare_identifier {
        ( $i, $tok, $type, $prototype ) =
          scan_bare_identifier_do( $input_line, $i, $tok, $type, $prototype,
            $rtoken_map, $max_token_index );
    }

    sub scan_identifier {
        ( $i, $tok, $type, $id_scan_state, $identifier ) =
          scan_identifier_do( $i, $id_scan_state, $identifier, $rtokens,
            $max_token_index, $expecting );
    }

    sub scan_id {
        ( $i, $tok, $type, $id_scan_state ) =
          scan_id_do( $input_line, $i, $tok, $rtokens, $rtoken_map,
            $id_scan_state, $max_token_index );
    }

    sub scan_number {
        my $number;
        ( $i, $type, $number ) =
          scan_number_do( $input_line, $i, $rtoken_map, $type,
            $max_token_index );
        return $number;
    }

    # a sub to warn if token found where term expected
    sub error_if_expecting_TERM {
        if ( $expecting == TERM ) {
            if ( $really_want_term{$last_nonblank_type} ) {
                unexpected( $tok, "term", $i_tok, $last_nonblank_i, $rtoken_map,
                    $rtoken_type, $input_line );
                1;
            }
        }
    }

    # a sub to warn if token found where operator expected
    sub error_if_expecting_OPERATOR {
        if ( $expecting == OPERATOR ) {
            my $thing = defined $_[0] ? $_[0] : $tok;
            unexpected( $thing, "operator", $i_tok, $last_nonblank_i,
                $rtoken_map, $rtoken_type, $input_line );
            if ( $i_tok == 0 ) {
                interrupt_logfile();
                warning("Missing ';' above?\n");
                resume_logfile();
            }
            1;
        }
    }

    # ------------------------------------------------------------
    # end scanner interfaces
    # ------------------------------------------------------------

    my %is_for_foreach;
    @_ = qw(for foreach);
    @is_for_foreach{@_} = (1) x scalar(@_);

    my %is_my_our;
    @_ = qw(my our);
    @is_my_our{@_} = (1) x scalar(@_);

    # These keywords may introduce blocks after parenthesized expressions,
    # in the form:
    # keyword ( .... ) { BLOCK }
    # patch for SWITCH/CASE: added 'switch' 'case' 'given' 'when'
    my %is_blocktype_with_paren;
    @_ = qw(if elsif unless while until for foreach switch case given when);
    @is_blocktype_with_paren{@_} = (1) x scalar(@_);

    # ------------------------------------------------------------
    # begin hash of code for handling most token types
    # ------------------------------------------------------------
    my $tokenization_code = {

        # no special code for these types yet, but syntax checks
        # could be added

##      '!'   => undef,
##      '!='  => undef,
##      '!~'  => undef,
##      '%='  => undef,
##      '&&=' => undef,
##      '&='  => undef,
##      '+='  => undef,
##      '-='  => undef,
##      '..'  => undef,
##      '..'  => undef,
##      '...' => undef,
##      '.='  => undef,
##      '<<=' => undef,
##      '<='  => undef,
##      '<=>' => undef,
##      '<>'  => undef,
##      '='   => undef,
##      '=='  => undef,
##      '=~'  => undef,
##      '>='  => undef,
##      '>>'  => undef,
##      '>>=' => undef,
##      '\\'  => undef,
##      '^='  => undef,
##      '|='  => undef,
##      '||=' => undef,
##      '//=' => undef,
##      '~'   => undef,
##      '~~'  => undef,
##      '!~~'  => undef,

        '>' => sub {
            error_if_expecting_TERM()
              if ( $expecting == TERM );
        },
        '|' => sub {
            error_if_expecting_TERM()
              if ( $expecting == TERM );
        },
        '$' => sub {

            # start looking for a scalar
            error_if_expecting_OPERATOR("Scalar")
              if ( $expecting == OPERATOR );
            scan_identifier();

            if ( $identifier eq '$^W' ) {
                $tokenizer_self->{_saw_perl_dash_w} = 1;
            }

            # Check for indentifier in indirect object slot
            # (vorboard.pl, sort.t).  Something like:
            #   /^(print|printf|sort|exec|system)$/
            if (
                $is_indirect_object_taker{$last_nonblank_token}

                || ( ( $last_nonblank_token eq '(' )
                    && $is_indirect_object_taker{ $paren_type[$paren_depth] } )
                || ( $last_nonblank_type =~ /^[Uw]$/ )    # possible object
              )
            {
                $type = 'Z';
            }
        },
        '(' => sub {

            ++$paren_depth;
            $paren_semicolon_count[$paren_depth] = 0;
            if ($want_paren) {
                $container_type = $want_paren;
                $want_paren     = "";
            }
            else {
                $container_type = $last_nonblank_token;

                # We can check for a syntax error here of unexpected '(',
                # but this is going to get messy...
                if (
                    $expecting == OPERATOR

                    # be sure this is not a method call of the form
                    # &method(...), $method->(..), &{method}(...),
                    # $ref[2](list) is ok & short for $ref[2]->(list)
                    # NOTE: at present, braces in something like &{ xxx }
                    # are not marked as a block, we might have a method call
                    && $last_nonblank_token !~ /^([\]\}\&]|\-\>)/

                  )
                {

                    # ref: camel 3 p 703.
                    if ( $last_last_nonblank_token eq 'do' ) {
                        complain(
"do SUBROUTINE is deprecated; consider & or -> notation\n"
                        );
                    }
                    else {

                        # if this is an empty list, (), then it is not an
                        # error; for example, we might have a constant pi and
                        # invoke it with pi() or just pi;
                        my ( $next_nonblank_token, $i_next ) =
                          find_next_nonblank_token( $i, $rtokens,
                            $max_token_index );
                        if ( $next_nonblank_token ne ')' ) {
                            my $hint;
                            error_if_expecting_OPERATOR('(');

                            if ( $last_nonblank_type eq 'C' ) {
                                $hint =
                                  "$last_nonblank_token has a void prototype\n";
                            }
                            elsif ( $last_nonblank_type eq 'i' ) {
                                if (   $i_tok > 0
                                    && $last_nonblank_token =~ /^\$/ )
                                {
                                    $hint =
"Do you mean '$last_nonblank_token->(' ?\n";
                                }
                            }
                            if ($hint) {
                                interrupt_logfile();
                                warning($hint);
                                resume_logfile();
                            }
                        } ## end if ( $next_nonblank_token...
                    } ## end else [ if ( $last_last_nonblank_token...
                } ## end if ( $expecting == OPERATOR...
            }
            $paren_type[$paren_depth] = $container_type;
            ( $type_sequence, $indent_flag ) =
              increase_nesting_depth( PAREN, $$rtoken_map[$i_tok] );

            # propagate types down through nested parens
            # for example: the second paren in 'if ((' would be structural
            # since the first is.

            if ( $last_nonblank_token eq '(' ) {
                $type = $last_nonblank_type;
            }

            #     We exclude parens as structural after a ',' because it
            #     causes subtle problems with continuation indentation for
            #     something like this, where the first 'or' will not get
            #     indented.
            #
            #         assert(
            #             __LINE__,
            #             ( not defined $check )
            #               or ref $check
            #               or $check eq "new"
            #               or $check eq "old",
            #         );
            #
            #     Likewise, we exclude parens where a statement can start
            #     because of problems with continuation indentation, like
            #     these:
            #
            #         ($firstline =~ /^#\!.*perl/)
            #         and (print $File::Find::name, "\n")
            #           and (return 1);
            #
            #         (ref($usage_fref) =~ /CODE/)
            #         ? &$usage_fref
            #           : (&blast_usage, &blast_params, &blast_general_params);

            else {
                $type = '{';
            }

            if ( $last_nonblank_type eq ')' ) {
                warning(
                    "Syntax error? found token '$last_nonblank_type' then '('\n"
                );
            }
            $paren_structural_type[$paren_depth] = $type;

        },
        ')' => sub {
            ( $type_sequence, $indent_flag ) =
              decrease_nesting_depth( PAREN, $$rtoken_map[$i_tok] );

            if ( $paren_structural_type[$paren_depth] eq '{' ) {
                $type = '}';
            }

            $container_type = $paren_type[$paren_depth];

            #    /^(for|foreach)$/
            if ( $is_for_foreach{ $paren_type[$paren_depth] } ) {
                my $num_sc = $paren_semicolon_count[$paren_depth];
                if ( $num_sc > 0 && $num_sc != 2 ) {
                    warning("Expected 2 ';' in 'for(;;)' but saw $num_sc\n");
                }
            }

            if ( $paren_depth > 0 ) { $paren_depth-- }
        },
        ',' => sub {
            if ( $last_nonblank_type eq ',' ) {
                complain("Repeated ','s \n");
            }

            # patch for operator_expected: note if we are in the list (use.t)
            if ( $statement_type eq 'use' ) { $statement_type = '_use' }
##                FIXME: need to move this elsewhere, perhaps check after a '('
##                elsif ($last_nonblank_token eq '(') {
##                    warning("Leading ','s illegal in some versions of perl\n");
##                }
        },
        ';' => sub {
            $context        = UNKNOWN_CONTEXT;
            $statement_type = '';

            #    /^(for|foreach)$/
            if ( $is_for_foreach{ $paren_type[$paren_depth] } )
            {    # mark ; in for loop

                # Be careful: we do not want a semicolon such as the
                # following to be included:
                #
                #    for (sort {strcoll($a,$b);} keys %investments) {

                if (   $brace_depth == $depth_array[PAREN][BRACE][$paren_depth]
                    && $square_bracket_depth ==
                    $depth_array[PAREN][SQUARE_BRACKET][$paren_depth] )
                {

                    $type = 'f';
                    $paren_semicolon_count[$paren_depth]++;
                }
            }

        },
        '"' => sub {
            error_if_expecting_OPERATOR("String")
              if ( $expecting == OPERATOR );
            $in_quote                = 1;
            $type                    = 'Q';
            $allowed_quote_modifiers = "";
        },
        "'" => sub {
            error_if_expecting_OPERATOR("String")
              if ( $expecting == OPERATOR );
            $in_quote                = 1;
            $type                    = 'Q';
            $allowed_quote_modifiers = "";
        },
        '`' => sub {
            error_if_expecting_OPERATOR("String")
              if ( $expecting == OPERATOR );
            $in_quote                = 1;
            $type                    = 'Q';
            $allowed_quote_modifiers = "";
        },
        '/' => sub {
            my $is_pattern;

            if ( $expecting == UNKNOWN ) {    # indeterminte, must guess..
                my $msg;
                ( $is_pattern, $msg ) =
                  guess_if_pattern_or_division( $i, $rtokens, $rtoken_map,
                    $max_token_index );

                if ($msg) {
                    write_diagnostics("DIVIDE:$msg\n");
                    write_logfile_entry($msg);
                }
            }
            else { $is_pattern = ( $expecting == TERM ) }

            if ($is_pattern) {
                $in_quote                = 1;
                $type                    = 'Q';
                $allowed_quote_modifiers = '[msixpodualgc]';
            }
            else {    # not a pattern; check for a /= token

                if ( $$rtokens[ $i + 1 ] eq '=' ) {    # form token /=
                    $i++;
                    $tok  = '/=';
                    $type = $tok;
                }

              #DEBUG - collecting info on what tokens follow a divide
              # for development of guessing algorithm
              #if ( numerator_expected( $i, $rtokens, $max_token_index ) < 0 ) {
              #    #write_diagnostics( "DIVIDE? $input_line\n" );
              #}
            }
        },
        '{' => sub {

            # if we just saw a ')', we will label this block with
            # its type.  We need to do this to allow sub
            # code_block_type to determine if this brace starts a
            # code block or anonymous hash.  (The type of a paren
            # pair is the preceding token, such as 'if', 'else',
            # etc).
            $container_type = "";

            # ATTRS: for a '{' following an attribute list, reset
            # things to look like we just saw the sub name
            if ( $statement_type =~ /^sub/ ) {
                $last_nonblank_token = $statement_type;
                $last_nonblank_type  = 'i';
                $statement_type      = "";
            }

            # patch for SWITCH/CASE: hide these keywords from an immediately
            # following opening brace
            elsif ( ( $statement_type eq 'case' || $statement_type eq 'when' )
                && $statement_type eq $last_nonblank_token )
            {
                $last_nonblank_token = ";";
            }

            elsif ( $last_nonblank_token eq ')' ) {
                $last_nonblank_token = $paren_type[ $paren_depth + 1 ];

                # defensive move in case of a nesting error (pbug.t)
                # in which this ')' had no previous '('
                # this nesting error will have been caught
                if ( !defined($last_nonblank_token) ) {
                    $last_nonblank_token = 'if';
                }

                # check for syntax error here;
                unless ( $is_blocktype_with_paren{$last_nonblank_token} ) {
                    my $list = join( ' ', sort keys %is_blocktype_with_paren );
                    warning(
                        "syntax error at ') {', didn't see one of: $list\n");
                }
            }

            # patch for paren-less for/foreach glitch, part 2.
            # see note below under 'qw'
            elsif ($last_nonblank_token eq 'qw'
                && $is_for_foreach{$want_paren} )
            {
                $last_nonblank_token = $want_paren;
                if ( $last_last_nonblank_token eq $want_paren ) {
                    warning(
"syntax error at '$want_paren .. {' -- missing \$ loop variable\n"
                    );

                }
                $want_paren = "";
            }

            # now identify which of the three possible types of
            # curly braces we have: hash index container, anonymous
            # hash reference, or code block.

            # non-structural (hash index) curly brace pair
            # get marked 'L' and 'R'
            if ( is_non_structural_brace() ) {
                $type = 'L';

                # patch for SWITCH/CASE:
                # allow paren-less identifier after 'when'
                # if the brace is preceded by a space
                if (   $statement_type eq 'when'
                    && $last_nonblank_type eq 'i'
                    && $last_last_nonblank_type eq 'k'
                    && ( $i_tok == 0 || $rtoken_type->[ $i_tok - 1 ] eq 'b' ) )
                {
                    $type       = '{';
                    $block_type = $statement_type;
                }
            }

            # code and anonymous hash have the same type, '{', but are
            # distinguished by 'block_type',
            # which will be blank for an anonymous hash
            else {

                $block_type = code_block_type( $i_tok, $rtokens, $rtoken_type,
                    $max_token_index );

                # remember a preceding smartmatch operator
                if ( $last_nonblank_type eq '~~' ) {
                    $block_type = $last_nonblank_type;
                }

                # patch to promote bareword type to function taking block
                if (   $block_type
                    && $last_nonblank_type eq 'w'
                    && $last_nonblank_i >= 0 )
                {
                    if ( $routput_token_type->[$last_nonblank_i] eq 'w' ) {
                        $routput_token_type->[$last_nonblank_i] = 'G';
                    }
                }

                # patch for SWITCH/CASE: if we find a stray opening block brace
                # where we might accept a 'case' or 'when' block, then take it
                if (   $statement_type eq 'case'
                    || $statement_type eq 'when' )
                {
                    if ( !$block_type || $block_type eq '}' ) {
                        $block_type = $statement_type;
                    }
                }
            }
            $brace_type[ ++$brace_depth ]        = $block_type;
            $brace_package[$brace_depth]         = $current_package;
            $brace_structural_type[$brace_depth] = $type;
            $brace_context[$brace_depth]         = $context;
            ( $type_sequence, $indent_flag ) =
              increase_nesting_depth( BRACE, $$rtoken_map[$i_tok] );
        },
        '}' => sub {
            $block_type = $brace_type[$brace_depth];
            if ($block_type) { $statement_type = '' }
            if ( defined( $brace_package[$brace_depth] ) ) {
                $current_package = $brace_package[$brace_depth];
            }

            # can happen on brace error (caught elsewhere)
            else {
            }
            ( $type_sequence, $indent_flag ) =
              decrease_nesting_depth( BRACE, $$rtoken_map[$i_tok] );

            if ( $brace_structural_type[$brace_depth] eq 'L' ) {
                $type = 'R';
            }

            # propagate type information for 'do' and 'eval' blocks, and also
            # for smartmatch operator.  This is necessary to enable us to know
            # if an operator or term is expected next.
            if ( $is_block_operator{$block_type} || $block_type eq '~~' ) {
                $tok = $block_type;
            }

            $context = $brace_context[$brace_depth];
            if ( $brace_depth > 0 ) { $brace_depth--; }
        },
        '&' => sub {    # maybe sub call? start looking

            # We have to check for sub call unless we are sure we
            # are expecting an operator.  This example from s2p
            # got mistaken as a q operator in an early version:
            #   print BODY &q(<<'EOT');
            if ( $expecting != OPERATOR ) {

                # But only look for a sub call if we are expecting a term or
                # if there is no existing space after the &.
                # For example we probably don't want & as sub call here:
                #    Fcntl::S_IRUSR & $mode;
                if ( $expecting == TERM || $next_type ne 'b' ) {
                    scan_identifier();
                }
            }
            else {
            }
        },
        '<' => sub {    # angle operator or less than?

            if ( $expecting != OPERATOR ) {
                ( $i, $type ) =
                  find_angle_operator_termination( $input_line, $i, $rtoken_map,
                    $expecting, $max_token_index );

                if ( $type eq '<' && $expecting == TERM ) {
                    error_if_expecting_TERM();
                    interrupt_logfile();
                    warning("Unterminated <> operator?\n");
                    resume_logfile();
                }
            }
            else {
            }
        },
        '?' => sub {    # ?: conditional or starting pattern?

            my $is_pattern;

            if ( $expecting == UNKNOWN ) {

                my $msg;
                ( $is_pattern, $msg ) =
                  guess_if_pattern_or_conditional( $i, $rtokens, $rtoken_map,
                    $max_token_index );

                if ($msg) { write_logfile_entry($msg) }
            }
            else { $is_pattern = ( $expecting == TERM ) }

            if ($is_pattern) {
                $in_quote                = 1;
                $type                    = 'Q';
                $allowed_quote_modifiers = '[msixpodualgc]';
            }
            else {
                ( $type_sequence, $indent_flag ) =
                  increase_nesting_depth( QUESTION_COLON,
                    $$rtoken_map[$i_tok] );
            }
        },
        '*' => sub {    # typeglob, or multiply?

            if ( $expecting == TERM ) {
                scan_identifier();
            }
            else {

                if ( $$rtokens[ $i + 1 ] eq '=' ) {
                    $tok  = '*=';
                    $type = $tok;
                    $i++;
                }
                elsif ( $$rtokens[ $i + 1 ] eq '*' ) {
                    $tok  = '**';
                    $type = $tok;
                    $i++;
                    if ( $$rtokens[ $i + 1 ] eq '=' ) {
                        $tok  = '**=';
                        $type = $tok;
                        $i++;
                    }
                }
            }
        },
        '.' => sub {    # what kind of . ?

            if ( $expecting != OPERATOR ) {
                scan_number();
                if ( $type eq '.' ) {
                    error_if_expecting_TERM()
                      if ( $expecting == TERM );
                }
            }
            else {
            }
        },
        ':' => sub {

            # if this is the first nonblank character, call it a label
            # since perl seems to just swallow it
            if ( $input_line_number == 1 && $last_nonblank_i == -1 ) {
                $type = 'J';
            }

            # ATTRS: check for a ':' which introduces an attribute list
            # (this might eventually get its own token type)
            elsif ( $statement_type =~ /^sub/ ) {
                $type              = 'A';
                $in_attribute_list = 1;
            }

            # check for scalar attribute, such as
            # my $foo : shared = 1;
            elsif ($is_my_our{$statement_type}
                && $current_depth[QUESTION_COLON] == 0 )
            {
                $type              = 'A';
                $in_attribute_list = 1;
            }

            # otherwise, it should be part of a ?/: operator
            else {
                ( $type_sequence, $indent_flag ) =
                  decrease_nesting_depth( QUESTION_COLON,
                    $$rtoken_map[$i_tok] );
                if ( $last_nonblank_token eq '?' ) {
                    warning("Syntax error near ? :\n");
                }
            }
        },
        '+' => sub {    # what kind of plus?

            if ( $expecting == TERM ) {
                my $number = scan_number();

                # unary plus is safest assumption if not a number
                if ( !defined($number) ) { $type = 'p'; }
            }
            elsif ( $expecting == OPERATOR ) {
            }
            else {
                if ( $next_type eq 'w' ) { $type = 'p' }
            }
        },
        '@' => sub {

            error_if_expecting_OPERATOR("Array")
              if ( $expecting == OPERATOR );
            scan_identifier();
        },
        '%' => sub {    # hash or modulo?

            # first guess is hash if no following blank
            if ( $expecting == UNKNOWN ) {
                if ( $next_type ne 'b' ) { $expecting = TERM }
            }
            if ( $expecting == TERM ) {
                scan_identifier();
            }
        },
        '[' => sub {
            $square_bracket_type[ ++$square_bracket_depth ] =
              $last_nonblank_token;
            ( $type_sequence, $indent_flag ) =
              increase_nesting_depth( SQUARE_BRACKET, $$rtoken_map[$i_tok] );

            # It may seem odd, but structural square brackets have
            # type '{' and '}'.  This simplifies the indentation logic.
            if ( !is_non_structural_brace() ) {
                $type = '{';
            }
            $square_bracket_structural_type[$square_bracket_depth] = $type;
        },
        ']' => sub {
            ( $type_sequence, $indent_flag ) =
              decrease_nesting_depth( SQUARE_BRACKET, $$rtoken_map[$i_tok] );

            if ( $square_bracket_structural_type[$square_bracket_depth] eq '{' )
            {
                $type = '}';
            }

            # propagate type information for smartmatch operator.  This is
            # necessary to enable us to know if an operator or term is expected
            # next.
            if ( $square_bracket_type[$square_bracket_depth] eq '~~' ) {
                $tok = $square_bracket_type[$square_bracket_depth];
            }

            if ( $square_bracket_depth > 0 ) { $square_bracket_depth--; }
        },
        '-' => sub {    # what kind of minus?

            if ( ( $expecting != OPERATOR )
                && $is_file_test_operator{$next_tok} )
            {
                my ( $next_nonblank_token, $i_next ) =
                  find_next_nonblank_token( $i + 1, $rtokens,
                    $max_token_index );

                # check for a quoted word like "-w=>xx";
                # it is sufficient to just check for a following '='
                if ( $next_nonblank_token eq '=' ) {
                    $type = 'm';
                }
                else {
                    $i++;
                    $tok .= $next_tok;
                    $type = 'F';
                }
            }
            elsif ( $expecting == TERM ) {
                my $number = scan_number();

                # maybe part of bareword token? unary is safest
                if ( !defined($number) ) { $type = 'm'; }

            }
            elsif ( $expecting == OPERATOR ) {
            }
            else {

                if ( $next_type eq 'w' ) {
                    $type = 'm';
                }
            }
        },

        '^' => sub {

            # check for special variables like ${^WARNING_BITS}
            if ( $expecting == TERM ) {

                # FIXME: this should work but will not catch errors
                # because we also have to be sure that previous token is
                # a type character ($,@,%).
                if ( $last_nonblank_token eq '{'
                    && ( $next_tok =~ /^[A-Za-z_]/ ) )
                {

                    if ( $next_tok eq 'W' ) {
                        $tokenizer_self->{_saw_perl_dash_w} = 1;
                    }
                    $tok  = $tok . $next_tok;
                    $i    = $i + 1;
                    $type = 'w';
                }

                else {
                    unless ( error_if_expecting_TERM() ) {

                        # Something like this is valid but strange:
                        # undef ^I;
                        complain("The '^' seems unusual here\n");
                    }
                }
            }
        },

        '::' => sub {    # probably a sub call
            scan_bare_identifier();
        },
        '<<' => sub {    # maybe a here-doc?
            return
              unless ( $i < $max_token_index )
              ;          # here-doc not possible if end of line

            if ( $expecting != OPERATOR ) {
                my ( $found_target, $here_doc_target, $here_quote_character,
                    $saw_error );
                (
                    $found_target, $here_doc_target, $here_quote_character, $i,
                    $saw_error
                  )
                  = find_here_doc( $expecting, $i, $rtokens, $rtoken_map,
                    $max_token_index );

                if ($found_target) {
                    push @{$rhere_target_list},
                      [ $here_doc_target, $here_quote_character ];
                    $type = 'h';
                    if ( length($here_doc_target) > 80 ) {
                        my $truncated = substr( $here_doc_target, 0, 80 );
                        complain("Long here-target: '$truncated' ...\n");
                    }
                    elsif ( $here_doc_target !~ /^[A-Z_]\w+$/ ) {
                        complain(
                            "Unconventional here-target: '$here_doc_target'\n"
                        );
                    }
                }
                elsif ( $expecting == TERM ) {
                    unless ($saw_error) {

                        # shouldn't happen..
                        warning("Program bug; didn't find here doc target\n");
                        report_definite_bug();
                    }
                }
            }
            else {
            }
        },
        '->' => sub {

            # if -> points to a bare word, we must scan for an identifier,
            # otherwise something like ->y would look like the y operator
            scan_identifier();
        },

        # type = 'pp' for pre-increment, '++' for post-increment
        '++' => sub {
            if ( $expecting == TERM ) { $type = 'pp' }
            elsif ( $expecting == UNKNOWN ) {
                my ( $next_nonblank_token, $i_next ) =
                  find_next_nonblank_token( $i, $rtokens, $max_token_index );
                if ( $next_nonblank_token eq '$' ) { $type = 'pp' }
            }
        },

        '=>' => sub {
            if ( $last_nonblank_type eq $tok ) {
                complain("Repeated '=>'s \n");
            }

            # patch for operator_expected: note if we are in the list (use.t)
            # TODO: make version numbers a new token type
            if ( $statement_type eq 'use' ) { $statement_type = '_use' }
        },

        # type = 'mm' for pre-decrement, '--' for post-decrement
        '--' => sub {

            if ( $expecting == TERM ) { $type = 'mm' }
            elsif ( $expecting == UNKNOWN ) {
                my ( $next_nonblank_token, $i_next ) =
                  find_next_nonblank_token( $i, $rtokens, $max_token_index );
                if ( $next_nonblank_token eq '$' ) { $type = 'mm' }
            }
        },

        '&&' => sub {
            error_if_expecting_TERM()
              if ( $expecting == TERM );
        },

        '||' => sub {
            error_if_expecting_TERM()
              if ( $expecting == TERM );
        },

        '//' => sub {
            error_if_expecting_TERM()
              if ( $expecting == TERM );
        },
    };

    # ------------------------------------------------------------
    # end hash of code for handling individual token types
    # ------------------------------------------------------------

    my %matching_start_token = ( '}' => '{', ']' => '[', ')' => '(' );

    # These block types terminate statements and do not need a trailing
    # semicolon
    # patched for SWITCH/CASE/
    my %is_zero_continuation_block_type;
    @_ = qw( } { BEGIN END CHECK INIT AUTOLOAD DESTROY UNITCHECK continue ;
      if elsif else unless while until for foreach switch case given when);
    @is_zero_continuation_block_type{@_} = (1) x scalar(@_);

    my %is_not_zero_continuation_block_type;
    @_ = qw(sort grep map do eval);
    @is_not_zero_continuation_block_type{@_} = (1) x scalar(@_);

    my %is_logical_container;
    @_ = qw(if elsif unless while and or err not && !  || for foreach);
    @is_logical_container{@_} = (1) x scalar(@_);

    my %is_binary_type;
    @_ = qw(|| &&);
    @is_binary_type{@_} = (1) x scalar(@_);

    my %is_binary_keyword;
    @_ = qw(and or err eq ne cmp);
    @is_binary_keyword{@_} = (1) x scalar(@_);

    # 'L' is token for opening { at hash key
    my %is_opening_type;
    @_ = qw" L { ( [ ";
    @is_opening_type{@_} = (1) x scalar(@_);

    # 'R' is token for closing } at hash key
    my %is_closing_type;
    @_ = qw" R } ) ] ";
    @is_closing_type{@_} = (1) x scalar(@_);

    my %is_redo_last_next_goto;
    @_ = qw(redo last next goto);
    @is_redo_last_next_goto{@_} = (1) x scalar(@_);

    my %is_use_require;
    @_ = qw(use require);
    @is_use_require{@_} = (1) x scalar(@_);

    my %is_sub_package;
    @_ = qw(sub package);
    @is_sub_package{@_} = (1) x scalar(@_);

    # This hash holds the hash key in $tokenizer_self for these keywords:
    my %is_format_END_DATA = (
        'format'   => '_in_format',
        '__END__'  => '_in_end',
        '__DATA__' => '_in_data',
    );

    # ref: camel 3 p 147,
    # but perl may accept undocumented flags
    # perl 5.10 adds 'p' (preserve)
    # Perl version 5.16, http://perldoc.perl.org/perlop.html,  has these:
    # /PATTERN/msixpodualgc or m?PATTERN?msixpodualgc
    # s/PATTERN/REPLACEMENT/msixpodualgcer
    # y/SEARCHLIST/REPLACEMENTLIST/cdsr
    # tr/SEARCHLIST/REPLACEMENTLIST/cdsr
    # qr/STRING/msixpodual
    my %quote_modifiers = (
        's'  => '[msixpodualgcer]',
        'y'  => '[cdsr]',
        'tr' => '[cdsr]',
        'm'  => '[msixpodualgc]',
        'qr' => '[msixpodual]',
        'q'  => "",
        'qq' => "",
        'qw' => "",
        'qx' => "",
    );

    # table showing how many quoted things to look for after quote operator..
    # s, y, tr have 2 (pattern and replacement)
    # others have 1 (pattern only)
    my %quote_items = (
        's'  => 2,
        'y'  => 2,
        'tr' => 2,
        'm'  => 1,
        'qr' => 1,
        'q'  => 1,
        'qq' => 1,
        'qw' => 1,
        'qx' => 1,
    );

    sub tokenize_this_line {

  # This routine breaks a line of perl code into tokens which are of use in
  # indentation and reformatting.  One of my goals has been to define tokens
  # such that a newline may be inserted between any pair of tokens without
  # changing or invalidating the program. This version comes close to this,
  # although there are necessarily a few exceptions which must be caught by
  # the formatter.  Many of these involve the treatment of bare words.
  #
  # The tokens and their types are returned in arrays.  See previous
  # routine for their names.
  #
  # See also the array "valid_token_types" in the BEGIN section for an
  # up-to-date list.
  #
  # To simplify things, token types are either a single character, or they
  # are identical to the tokens themselves.
  #
  # As a debugging aid, the -D flag creates a file containing a side-by-side
  # comparison of the input string and its tokenization for each line of a file.
  # This is an invaluable debugging aid.
  #
  # In addition to tokens, and some associated quantities, the tokenizer
  # also returns flags indication any special line types.  These include
  # quotes, here_docs, formats.
  #
  # -----------------------------------------------------------------------
  #
  # How to add NEW_TOKENS:
  #
  # New token types will undoubtedly be needed in the future both to keep up
  # with changes in perl and to help adapt the tokenizer to other applications.
  #
  # Here are some notes on the minimal steps.  I wrote these notes while
  # adding the 'v' token type for v-strings, which are things like version
  # numbers 5.6.0, and ip addresses, and will use that as an example.  ( You
  # can use your editor to search for the string "NEW_TOKENS" to find the
  # appropriate sections to change):
  #
  # *. Try to talk somebody else into doing it!  If not, ..
  #
  # *. Make a backup of your current version in case things don't work out!
  #
  # *. Think of a new, unused character for the token type, and add to
  # the array @valid_token_types in the BEGIN section of this package.
  # For example, I used 'v' for v-strings.
  #
  # *. Implement coding to recognize the $type of the token in this routine.
  # This is the hardest part, and is best done by imitating or modifying
  # some of the existing coding.  For example, to recognize v-strings, I
  # patched 'sub scan_bare_identifier' to recognize v-strings beginning with
  # 'v' and 'sub scan_number' to recognize v-strings without the leading 'v'.
  #
  # *. Update sub operator_expected.  This update is critically important but
  # the coding is trivial.  Look at the comments in that routine for help.
  # For v-strings, which should behave like numbers, I just added 'v' to the
  # regex used to handle numbers and strings (types 'n' and 'Q').
  #
  # *. Implement a 'bond strength' rule in sub set_bond_strengths in
  # Perl::Tidy::Formatter for breaking lines around this token type.  You can
  # skip this step and take the default at first, then adjust later to get
  # desired results.  For adding type 'v', I looked at sub bond_strength and
  # saw that number type 'n' was using default strengths, so I didn't do
  # anything.  I may tune it up someday if I don't like the way line
  # breaks with v-strings look.
  #
  # *. Implement a 'whitespace' rule in sub set_white_space_flag in
  # Perl::Tidy::Formatter.  For adding type 'v', I looked at this routine
  # and saw that type 'n' used spaces on both sides, so I just added 'v'
  # to the array @spaces_both_sides.
  #
  # *. Update HtmlWriter package so that users can colorize the token as
  # desired.  This is quite easy; see comments identified by 'NEW_TOKENS' in
  # that package.  For v-strings, I initially chose to use a default color
  # equal to the default for numbers, but it might be nice to change that
  # eventually.
  #
  # *. Update comments in Perl::Tidy::Tokenizer::dump_token_types.
  #
  # *. Run lots and lots of debug tests.  Start with special files designed
  # to test the new token type.  Run with the -D flag to create a .DEBUG
  # file which shows the tokenization.  When these work ok, test as many old
  # scripts as possible.  Start with all of the '.t' files in the 'test'
  # directory of the distribution file.  Compare .tdy output with previous
  # version and updated version to see the differences.  Then include as
  # many more files as possible. My own technique has been to collect a huge
  # number of perl scripts (thousands!) into one directory and run perltidy
  # *, then run diff between the output of the previous version and the
  # current version.
  #
  # *. For another example, search for the smartmatch operator '~~'
  # with your editor to see where updates were made for it.
  #
  # -----------------------------------------------------------------------

        my $line_of_tokens = shift;
        my ($untrimmed_input_line) = $line_of_tokens->{_line_text};

        # patch while coding change is underway
        # make callers private data to allow access
        # $tokenizer_self = $caller_tokenizer_self;

        # extract line number for use in error messages
        $input_line_number = $line_of_tokens->{_line_number};

        # reinitialize for multi-line quote
        $line_of_tokens->{_starting_in_quote} = $in_quote && $quote_type eq 'Q';

        # check for pod documentation
        if ( ( $untrimmed_input_line =~ /^=[A-Za-z_]/ ) ) {

            # must not be in multi-line quote
            # and must not be in an eqn
            if ( !$in_quote and ( operator_expected( 'b', '=', 'b' ) == TERM ) )
            {
                $tokenizer_self->{_in_pod} = 1;
                return;
            }
        }

        $input_line = $untrimmed_input_line;

        chomp $input_line;

        # trim start of this line unless we are continuing a quoted line
        # do not trim end because we might end in a quote (test: deken4.pl)
        # Perl::Tidy::Formatter will delete needless trailing blanks
        unless ( $in_quote && ( $quote_type eq 'Q' ) ) {
            $input_line =~ s/^\s*//;    # trim left end
        }

        # update the copy of the line for use in error messages
        # This must be exactly what we give the pre_tokenizer
        $tokenizer_self->{_line_text} = $input_line;

        # re-initialize for the main loop
        $routput_token_list     = [];    # stack of output token indexes
        $routput_token_type     = [];    # token types
        $routput_block_type     = [];    # types of code block
        $routput_container_type = [];    # paren types, such as if, elsif, ..
        $routput_type_sequence  = [];    # nesting sequential number

        $rhere_target_list = [];

        $tok             = $last_nonblank_token;
        $type            = $last_nonblank_type;
        $prototype       = $last_nonblank_prototype;
        $last_nonblank_i = -1;
        $block_type      = $last_nonblank_block_type;
        $container_type  = $last_nonblank_container_type;
        $type_sequence   = $last_nonblank_type_sequence;
        $indent_flag     = 0;
        $peeked_ahead    = 0;

        # tokenization is done in two stages..
        # stage 1 is a very simple pre-tokenization
        my $max_tokens_wanted = 0; # this signals pre_tokenize to get all tokens

        # a little optimization for a full-line comment
        if ( !$in_quote && ( $input_line =~ /^#/ ) ) {
            $max_tokens_wanted = 1    # no use tokenizing a comment
        }

        # start by breaking the line into pre-tokens
        ( $rtokens, $rtoken_map, $rtoken_type ) =
          pre_tokenize( $input_line, $max_tokens_wanted );

        $max_token_index = scalar(@$rtokens) - 1;
        push( @$rtokens,    ' ', ' ', ' ' ); # extra whitespace simplifies logic
        push( @$rtoken_map, 0,   0,   0 );   # shouldn't be referenced
        push( @$rtoken_type, 'b', 'b', 'b' );

        # initialize for main loop
        for $i ( 0 .. $max_token_index + 3 ) {
            $routput_token_type->[$i]     = "";
            $routput_block_type->[$i]     = "";
            $routput_container_type->[$i] = "";
            $routput_type_sequence->[$i]  = "";
            $routput_indent_flag->[$i]    = 0;
        }
        $i     = -1;
        $i_tok = -1;

        # ------------------------------------------------------------
        # begin main tokenization loop
        # ------------------------------------------------------------

        # we are looking at each pre-token of one line and combining them
        # into tokens
        while ( ++$i <= $max_token_index ) {

            if ($in_quote) {    # continue looking for end of a quote
                $type = $quote_type;

                unless ( @{$routput_token_list} )
                {               # initialize if continuation line
                    push( @{$routput_token_list}, $i );
                    $routput_token_type->[$i] = $type;

                }
                $tok = $quote_character unless ( $quote_character =~ /^\s*$/ );

                # scan for the end of the quote or pattern
                (
                    $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
                    $quoted_string_1, $quoted_string_2
                  )
                  = do_quote(
                    $i,               $in_quote,    $quote_character,
                    $quote_pos,       $quote_depth, $quoted_string_1,
                    $quoted_string_2, $rtokens,     $rtoken_map,
                    $max_token_index
                  );

                # all done if we didn't find it
                last if ($in_quote);

                # save pattern and replacement text for rescanning
                my $qs1 = $quoted_string_1;
                my $qs2 = $quoted_string_2;

                # re-initialize for next search
                $quote_character = '';
                $quote_pos       = 0;
                $quote_type      = 'Q';
                $quoted_string_1 = "";
                $quoted_string_2 = "";
                last if ( ++$i > $max_token_index );

                # look for any modifiers
                if ($allowed_quote_modifiers) {

                    # check for exact quote modifiers
                    if ( $$rtokens[$i] =~ /^[A-Za-z_]/ ) {
                        my $str = $$rtokens[$i];
                        my $saw_modifier_e;
                        while ( $str =~ /\G$allowed_quote_modifiers/gc ) {
                            my $pos = pos($str);
                            my $char = substr( $str, $pos - 1, 1 );
                            $saw_modifier_e ||= ( $char eq 'e' );
                        }

                        # For an 'e' quote modifier we must scan the replacement
                        # text for here-doc targets.
                        if ($saw_modifier_e) {

                            my $rht = scan_replacement_text($qs1);

                            # Change type from 'Q' to 'h' for quotes with
                            # here-doc targets so that the formatter (see sub
                            # print_line_of_tokens) will not make any line
                            # breaks after this point.
                            if ($rht) {
                                push @{$rhere_target_list}, @{$rht};
                                $type = 'h';
                                if ( $i_tok < 0 ) {
                                    my $ilast = $routput_token_list->[-1];
                                    $routput_token_type->[$ilast] = $type;
                                }
                            }
                        }

                        if ( defined( pos($str) ) ) {

                            # matched
                            if ( pos($str) == length($str) ) {
                                last if ( ++$i > $max_token_index );
                            }

                            # Looks like a joined quote modifier
                            # and keyword, maybe something like
                            # s/xxx/yyy/gefor @k=...
                            # Example is "galgen.pl".  Would have to split
                            # the word and insert a new token in the
                            # pre-token list.  This is so rare that I haven't
                            # done it.  Will just issue a warning citation.

                            # This error might also be triggered if my quote
                            # modifier characters are incomplete
                            else {
                                warning(<<EOM);

Partial match to quote modifier $allowed_quote_modifiers at word: '$str'
Please put a space between quote modifiers and trailing keywords.
EOM

                           # print "token $$rtokens[$i]\n";
                           # my $num = length($str) - pos($str);
                           # $$rtokens[$i]=substr($$rtokens[$i],pos($str),$num);
                           # print "continuing with new token $$rtokens[$i]\n";

                                # skipping past this token does least damage
                                last if ( ++$i > $max_token_index );
                            }
                        }
                        else {

                            # example file: rokicki4.pl
                            # This error might also be triggered if my quote
                            # modifier characters are incomplete
                            write_logfile_entry(
"Note: found word $str at quote modifier location\n"
                            );
                        }
                    }

                    # re-initialize
                    $allowed_quote_modifiers = "";
                }
            }

            unless ( $tok =~ /^\s*$/ || $tok eq 'CORE::' ) {

                # try to catch some common errors
                if ( ( $type eq 'n' ) && ( $tok ne '0' ) ) {

                    if ( $last_nonblank_token eq 'eq' ) {
                        complain("Should 'eq' be '==' here ?\n");
                    }
                    elsif ( $last_nonblank_token eq 'ne' ) {
                        complain("Should 'ne' be '!=' here ?\n");
                    }
                }

                $last_last_nonblank_token      = $last_nonblank_token;
                $last_last_nonblank_type       = $last_nonblank_type;
                $last_last_nonblank_block_type = $last_nonblank_block_type;
                $last_last_nonblank_container_type =
                  $last_nonblank_container_type;
                $last_last_nonblank_type_sequence =
                  $last_nonblank_type_sequence;
                $last_nonblank_token          = $tok;
                $last_nonblank_type           = $type;
                $last_nonblank_prototype      = $prototype;
                $last_nonblank_block_type     = $block_type;
                $last_nonblank_container_type = $container_type;
                $last_nonblank_type_sequence  = $type_sequence;
                $last_nonblank_i              = $i_tok;
            }

            # store previous token type
            if ( $i_tok >= 0 ) {
                $routput_token_type->[$i_tok]     = $type;
                $routput_block_type->[$i_tok]     = $block_type;
                $routput_container_type->[$i_tok] = $container_type;
                $routput_type_sequence->[$i_tok]  = $type_sequence;
                $routput_indent_flag->[$i_tok]    = $indent_flag;
            }
            my $pre_tok  = $$rtokens[$i];        # get the next pre-token
            my $pre_type = $$rtoken_type[$i];    # and type
            $tok  = $pre_tok;
            $type = $pre_type;                   # to be modified as necessary
            $block_type = "";    # blank for all tokens except code block braces
            $container_type = "";    # blank for all tokens except some parens
            $type_sequence  = "";    # blank for all tokens except ?/:
            $indent_flag    = 0;
            $prototype = "";    # blank for all tokens except user defined subs
            $i_tok     = $i;

            # this pre-token will start an output token
            push( @{$routput_token_list}, $i_tok );

            # continue gathering identifier if necessary
            # but do not start on blanks and comments
            if ( $id_scan_state && $pre_type !~ /[b#]/ ) {

                if ( $id_scan_state =~ /^(sub|package)/ ) {
                    scan_id();
                }
                else {
                    scan_identifier();
                }

                last if ($id_scan_state);
                next if ( ( $i > 0 ) || $type );

                # didn't find any token; start over
                $type = $pre_type;
                $tok  = $pre_tok;
            }

            # handle whitespace tokens..
            next if ( $type eq 'b' );
            my $prev_tok  = $i > 0 ? $$rtokens[ $i - 1 ]     : ' ';
            my $prev_type = $i > 0 ? $$rtoken_type[ $i - 1 ] : 'b';

            # Build larger tokens where possible, since we are not in a quote.
            #
            # First try to assemble digraphs.  The following tokens are
            # excluded and handled specially:
            # '/=' is excluded because the / might start a pattern.
            # 'x=' is excluded since it might be $x=, with $ on previous line
            # '**' and *= might be typeglobs of punctuation variables
            # I have allowed tokens starting with <, such as <=,
            # because I don't think these could be valid angle operators.
            # test file: storrs4.pl
            my $test_tok   = $tok . $$rtokens[ $i + 1 ];
            my $combine_ok = $is_digraph{$test_tok};

            # check for special cases which cannot be combined
            if ($combine_ok) {

                # '//' must be defined_or operator if an operator is expected.
                # TODO: Code for other ambiguous digraphs (/=, x=, **, *=)
                # could be migrated here for clarity
                if ( $test_tok eq '//' ) {
                    my $next_type = $$rtokens[ $i + 1 ];
                    my $expecting =
                      operator_expected( $prev_type, $tok, $next_type );
                    $combine_ok = 0 unless ( $expecting == OPERATOR );
                }
            }

            if (
                $combine_ok
                && ( $test_tok ne '/=' )    # might be pattern
                && ( $test_tok ne 'x=' )    # might be $x
                && ( $test_tok ne '**' )    # typeglob?
                && ( $test_tok ne '*=' )    # typeglob?
              )
            {
                $tok = $test_tok;
                $i++;

                # Now try to assemble trigraphs.  Note that all possible
                # perl trigraphs can be constructed by appending a character
                # to a digraph.
                $test_tok = $tok . $$rtokens[ $i + 1 ];

                if ( $is_trigraph{$test_tok} ) {
                    $tok = $test_tok;
                    $i++;
                }
            }

            $type      = $tok;
            $next_tok  = $$rtokens[ $i + 1 ];
            $next_type = $$rtoken_type[ $i + 1 ];

            TOKENIZER_DEBUG_FLAG_TOKENIZE && do {
                local $" = ')(';
                my @debug_list = (
                    $last_nonblank_token,      $tok,
                    $next_tok,                 $brace_depth,
                    $brace_type[$brace_depth], $paren_depth,
                    $paren_type[$paren_depth]
                );
                print STDOUT "TOKENIZE:(@debug_list)\n";
            };

            # turn off attribute list on first non-blank, non-bareword
            if ( $pre_type ne 'w' ) { $in_attribute_list = 0 }

            ###############################################################
            # We have the next token, $tok.
            # Now we have to examine this token and decide what it is
            # and define its $type
            #
            # section 1: bare words
            ###############################################################

            if ( $pre_type eq 'w' ) {
                $expecting = operator_expected( $prev_type, $tok, $next_type );
                my ( $next_nonblank_token, $i_next ) =
                  find_next_nonblank_token( $i, $rtokens, $max_token_index );

                # ATTRS: handle sub and variable attributes
                if ($in_attribute_list) {

                    # treat bare word followed by open paren like qw(
                    if ( $next_nonblank_token eq '(' ) {
                        $in_quote                = $quote_items{'q'};
                        $allowed_quote_modifiers = $quote_modifiers{'q'};
                        $type                    = 'q';
                        $quote_type              = 'q';
                        next;
                    }

                    # handle bareword not followed by open paren
                    else {
                        $type = 'w';
                        next;
                    }
                }

                # quote a word followed by => operator
                if ( $next_nonblank_token eq '=' ) {

                    if ( $$rtokens[ $i_next + 1 ] eq '>' ) {
                        if ( $is_constant{$current_package}{$tok} ) {
                            $type = 'C';
                        }
                        elsif ( $is_user_function{$current_package}{$tok} ) {
                            $type = 'U';
                            $prototype =
                              $user_function_prototype{$current_package}{$tok};
                        }
                        elsif ( $tok =~ /^v\d+$/ ) {
                            $type = 'v';
                            report_v_string($tok);
                        }
                        else { $type = 'w' }

                        next;
                    }
                }

     # quote a bare word within braces..like xxx->{s}; note that we
     # must be sure this is not a structural brace, to avoid
     # mistaking {s} in the following for a quoted bare word:
     #     for(@[){s}bla}BLA}
     # Also treat q in something like var{-q} as a bare word, not qoute operator
                if (
                    $next_nonblank_token eq '}'
                    && (
                        $last_nonblank_type eq 'L'
                        || (   $last_nonblank_type eq 'm'
                            && $last_last_nonblank_type eq 'L' )
                    )
                  )
                {
                    $type = 'w';
                    next;
                }

                # a bare word immediately followed by :: is not a keyword;
                # use $tok_kw when testing for keywords to avoid a mistake
                my $tok_kw = $tok;
                if ( $$rtokens[ $i + 1 ] eq ':' && $$rtokens[ $i + 2 ] eq ':' )
                {
                    $tok_kw .= '::';
                }

                # handle operator x (now we know it isn't $x=)
                if ( ( $tok =~ /^x\d*$/ ) && ( $expecting == OPERATOR ) ) {
                    if ( $tok eq 'x' ) {

                        if ( $$rtokens[ $i + 1 ] eq '=' ) {    # x=
                            $tok  = 'x=';
                            $type = $tok;
                            $i++;
                        }
                        else {
                            $type = 'x';
                        }
                    }

                    # FIXME: Patch: mark something like x4 as an integer for now
                    # It gets fixed downstream.  This is easier than
                    # splitting the pretoken.
                    else {
                        $type = 'n';
                    }
                }
                elsif ( $tok_kw eq 'CORE::' ) {
                    $type = $tok = $tok_kw;
                    $i += 2;
                }
                elsif ( ( $tok eq 'strict' )
                    and ( $last_nonblank_token eq 'use' ) )
                {
                    $tokenizer_self->{_saw_use_strict} = 1;
                    scan_bare_identifier();
                }

                elsif ( ( $tok eq 'warnings' )
                    and ( $last_nonblank_token eq 'use' ) )
                {
                    $tokenizer_self->{_saw_perl_dash_w} = 1;

                    # scan as identifier, so that we pick up something like:
                    # use warnings::register
                    scan_bare_identifier();
                }

                elsif (
                       $tok eq 'AutoLoader'
                    && $tokenizer_self->{_look_for_autoloader}
                    && (
                        $last_nonblank_token eq 'use'

                        # these regexes are from AutoSplit.pm, which we want
                        # to mimic
                        || $input_line =~ /^\s*(use|require)\s+AutoLoader\b/
                        || $input_line =~ /\bISA\s*=.*\bAutoLoader\b/
                    )
                  )
                {
                    write_logfile_entry("AutoLoader seen, -nlal deactivates\n");
                    $tokenizer_self->{_saw_autoloader}      = 1;
                    $tokenizer_self->{_look_for_autoloader} = 0;
                    scan_bare_identifier();
                }

                elsif (
                       $tok eq 'SelfLoader'
                    && $tokenizer_self->{_look_for_selfloader}
                    && (   $last_nonblank_token eq 'use'
                        || $input_line =~ /^\s*(use|require)\s+SelfLoader\b/
                        || $input_line =~ /\bISA\s*=.*\bSelfLoader\b/ )
                  )
                {
                    write_logfile_entry("SelfLoader seen, -nlsl deactivates\n");
                    $tokenizer_self->{_saw_selfloader}      = 1;
                    $tokenizer_self->{_look_for_selfloader} = 0;
                    scan_bare_identifier();
                }

                elsif ( ( $tok eq 'constant' )
                    and ( $last_nonblank_token eq 'use' ) )
                {
                    scan_bare_identifier();
                    my ( $next_nonblank_token, $i_next ) =
                      find_next_nonblank_token( $i, $rtokens,
                        $max_token_index );

                    if ($next_nonblank_token) {

                        if ( $is_keyword{$next_nonblank_token} ) {

                            # Assume qw is used as a quote and okay, as in:
                            #  use constant qw{ DEBUG 0 };
                            # Not worth trying to parse for just a warning

                            # NOTE: This warning is deactivated because recent
                            # versions of perl do not complain here, but
                            # the coding is retained for reference.
                            if ( 0 && $next_nonblank_token ne 'qw' ) {
                                warning(
"Attempting to define constant '$next_nonblank_token' which is a perl keyword\n"
                                );
                            }
                        }

                        # FIXME: could check for error in which next token is
                        # not a word (number, punctuation, ..)
                        else {
                            $is_constant{$current_package}{$next_nonblank_token}
                              = 1;
                        }
                    }
                }

                # various quote operators
                elsif ( $is_q_qq_qw_qx_qr_s_y_tr_m{$tok} ) {
                    if ( $expecting == OPERATOR ) {

                        # patch for paren-less for/foreach glitch, part 1
                        # perl will accept this construct as valid:
                        #
                        #    foreach my $key qw\Uno Due Tres Quadro\ {
                        #        print "Set $key\n";
                        #    }
                        unless ( $tok eq 'qw' && $is_for_foreach{$want_paren} )
                        {
                            error_if_expecting_OPERATOR();
                        }
                    }
                    $in_quote                = $quote_items{$tok};
                    $allowed_quote_modifiers = $quote_modifiers{$tok};

                   # All quote types are 'Q' except possibly qw quotes.
                   # qw quotes are special in that they may generally be trimmed
                   # of leading and trailing whitespace.  So they are given a
                   # separate type, 'q', unless requested otherwise.
                    $type =
                      ( $tok eq 'qw' && $tokenizer_self->{_trim_qw} )
                      ? 'q'
                      : 'Q';
                    $quote_type = $type;
                }

                # check for a statement label
                elsif (
                       ( $next_nonblank_token eq ':' )
                    && ( $$rtokens[ $i_next + 1 ] ne ':' )
                    && ( $i_next <= $max_token_index )      # colon on same line
                    && label_ok()
                  )
                {
                    if ( $tok !~ /[A-Z]/ ) {
                        push @{ $tokenizer_self->{_rlower_case_labels_at} },
                          $input_line_number;
                    }
                    $type = 'J';
                    $tok .= ':';
                    $i = $i_next;
                    next;
                }

                #      'sub' || 'package'
                elsif ( $is_sub_package{$tok_kw} ) {
                    error_if_expecting_OPERATOR()
                      if ( $expecting == OPERATOR );
                    scan_id();
                }

                # Note on token types for format, __DATA__, __END__:
                # It simplifies things to give these type ';', so that when we
                # start rescanning we will be expecting a token of type TERM.
                # We will switch to type 'k' before outputting the tokens.
                elsif ( $is_format_END_DATA{$tok_kw} ) {
                    $type = ';';    # make tokenizer look for TERM next
                    $tokenizer_self->{ $is_format_END_DATA{$tok_kw} } = 1;
                    last;
                }

                elsif ( $is_keyword{$tok_kw} ) {
                    $type = 'k';

                    # Since for and foreach may not be followed immediately
                    # by an opening paren, we have to remember which keyword
                    # is associated with the next '('
                    if ( $is_for_foreach{$tok} ) {
                        if ( new_statement_ok() ) {
                            $want_paren = $tok;
                        }
                    }

                    # recognize 'use' statements, which are special
                    elsif ( $is_use_require{$tok} ) {
                        $statement_type = $tok;
                        error_if_expecting_OPERATOR()
                          if ( $expecting == OPERATOR );
                    }

                    # remember my and our to check for trailing ": shared"
                    elsif ( $is_my_our{$tok} ) {
                        $statement_type = $tok;
                    }

                    # Check for misplaced 'elsif' and 'else', but allow isolated
                    # else or elsif blocks to be formatted.  This is indicated
                    # by a last noblank token of ';'
                    elsif ( $tok eq 'elsif' ) {
                        if (   $last_nonblank_token ne ';'
                            && $last_nonblank_block_type !~
                            /^(if|elsif|unless)$/ )
                        {
                            warning(
"expecting '$tok' to follow one of 'if|elsif|unless'\n"
                            );
                        }
                    }
                    elsif ( $tok eq 'else' ) {

                        # patched for SWITCH/CASE
                        if (   $last_nonblank_token ne ';'
                            && $last_nonblank_block_type !~
                            /^(if|elsif|unless|case|when)$/ )
                        {
                            warning(
"expecting '$tok' to follow one of 'if|elsif|unless|case|when'\n"
                            );
                        }
                    }
                    elsif ( $tok eq 'continue' ) {
                        if (   $last_nonblank_token ne ';'
                            && $last_nonblank_block_type !~
                            /(^(\{|\}|;|while|until|for|foreach)|:$)/ )
                        {

                            # note: ';' '{' and '}' in list above
                            # because continues can follow bare blocks;
                            # ':' is labeled block
                            #
                            ############################################
                            # NOTE: This check has been deactivated because
                            # continue has an alternative usage for given/when
                            # blocks in perl 5.10
                            ## warning("'$tok' should follow a block\n");
                            ############################################
                        }
                    }

                    # patch for SWITCH/CASE if 'case' and 'when are
                    # treated as keywords.
                    elsif ( $tok eq 'when' || $tok eq 'case' ) {
                        $statement_type = $tok;    # next '{' is block
                    }

                    #
                    # indent trailing if/unless/while/until
                    # outdenting will be handled by later indentation loop
## DEACTIVATED: unfortunately this can cause some unwanted indentation like:
##$opt_o = 1
##  if !(
##             $opt_b
##          || $opt_c
##          || $opt_d
##          || $opt_f
##          || $opt_i
##          || $opt_l
##          || $opt_o
##          || $opt_x
##  );
##                    if (   $tok =~ /^(if|unless|while|until)$/
##                        && $next_nonblank_token ne '(' )
##                    {
##                        $indent_flag = 1;
##                    }
                }

                # check for inline label following
                #         /^(redo|last|next|goto)$/
                elsif (( $last_nonblank_type eq 'k' )
                    && ( $is_redo_last_next_goto{$last_nonblank_token} ) )
                {
                    $type = 'j';
                    next;
                }

                # something else --
                else {

                    scan_bare_identifier();
                    if ( $type eq 'w' ) {

                        if ( $expecting == OPERATOR ) {

                            # don't complain about possible indirect object
                            # notation.
                            # For example:
                            #   package main;
                            #   sub new($) { ... }
                            #   $b = new A::;  # calls A::new
                            #   $c = new A;    # same thing but suspicious
                            # This will call A::new but we have a 'new' in
                            # main:: which looks like a constant.
                            #
                            if ( $last_nonblank_type eq 'C' ) {
                                if ( $tok !~ /::$/ ) {
                                    complain(<<EOM);
Expecting operator after '$last_nonblank_token' but found bare word '$tok'
       Maybe indirectet object notation?
EOM
                                }
                            }
                            else {
                                error_if_expecting_OPERATOR("bareword");
                            }
                        }

                        # mark bare words immediately followed by a paren as
                        # functions
                        $next_tok = $$rtokens[ $i + 1 ];
                        if ( $next_tok eq '(' ) {
                            $type = 'U';
                        }

                        # underscore after file test operator is file handle
                        if ( $tok eq '_' && $last_nonblank_type eq 'F' ) {
                            $type = 'Z';
                        }

                        # patch for SWITCH/CASE if 'case' and 'when are
                        # not treated as keywords:
                        if (
                            (
                                   $tok eq 'case'
                                && $brace_type[$brace_depth] eq 'switch'
                            )
                            || (   $tok eq 'when'
                                && $brace_type[$brace_depth] eq 'given' )
                          )
                        {
                            $statement_type = $tok;    # next '{' is block
                            $type = 'k';    # for keyword syntax coloring
                        }

                        # patch for SWITCH/CASE if switch and given not keywords
                        # Switch is not a perl 5 keyword, but we will gamble
                        # and mark switch followed by paren as a keyword.  This
                        # is only necessary to get html syntax coloring nice,
                        # and does not commit this as being a switch/case.
                        if ( $next_nonblank_token eq '('
                            && ( $tok eq 'switch' || $tok eq 'given' ) )
                        {
                            $type = 'k';    # for keyword syntax coloring
                        }
                    }
                }
            }

            ###############################################################
            # section 2: strings of digits
            ###############################################################
            elsif ( $pre_type eq 'd' ) {
                $expecting = operator_expected( $prev_type, $tok, $next_type );
                error_if_expecting_OPERATOR("Number")
                  if ( $expecting == OPERATOR );
                my $number = scan_number();
                if ( !defined($number) ) {

                    # shouldn't happen - we should always get a number
                    warning("non-number beginning with digit--program bug\n");
                    report_definite_bug();
                }
            }

            ###############################################################
            # section 3: all other tokens
            ###############################################################

            else {
                last if ( $tok eq '#' );
                my $code = $tokenization_code->{$tok};
                if ($code) {
                    $expecting =
                      operator_expected( $prev_type, $tok, $next_type );
                    $code->();
                    redo if $in_quote;
                }
            }
        }

        # -----------------------------
        # end of main tokenization loop
        # -----------------------------

        if ( $i_tok >= 0 ) {
            $routput_token_type->[$i_tok]     = $type;
            $routput_block_type->[$i_tok]     = $block_type;
            $routput_container_type->[$i_tok] = $container_type;
            $routput_type_sequence->[$i_tok]  = $type_sequence;
            $routput_indent_flag->[$i_tok]    = $indent_flag;
        }

        unless ( ( $type eq 'b' ) || ( $type eq '#' ) ) {
            $last_last_nonblank_token          = $last_nonblank_token;
            $last_last_nonblank_type           = $last_nonblank_type;
            $last_last_nonblank_block_type     = $last_nonblank_block_type;
            $last_last_nonblank_container_type = $last_nonblank_container_type;
            $last_last_nonblank_type_sequence  = $last_nonblank_type_sequence;
            $last_nonblank_token               = $tok;
            $last_nonblank_type                = $type;
            $last_nonblank_block_type          = $block_type;
            $last_nonblank_container_type      = $container_type;
            $last_nonblank_type_sequence       = $type_sequence;
            $last_nonblank_prototype           = $prototype;
        }

        # reset indentation level if necessary at a sub or package
        # in an attempt to recover from a nesting error
        if ( $level_in_tokenizer < 0 ) {
            if ( $input_line =~ /^\s*(sub|package)\s+(\w+)/ ) {
                reset_indentation_level(0);
                brace_warning("resetting level to 0 at $1 $2\n");
            }
        }

        # all done tokenizing this line ...
        # now prepare the final list of tokens and types

        my @token_type     = ();   # stack of output token types
        my @block_type     = ();   # stack of output code block types
        my @container_type = ();   # stack of output code container types
        my @type_sequence  = ();   # stack of output type sequence numbers
        my @tokens         = ();   # output tokens
        my @levels         = ();   # structural brace levels of output tokens
        my @slevels        = ();   # secondary nesting levels of output tokens
        my @nesting_tokens = ();   # string of tokens leading to this depth
        my @nesting_types  = ();   # string of token types leading to this depth
        my @nesting_blocks = ();   # string of block types leading to this depth
        my @nesting_lists  = ();   # string of list types leading to this depth
        my @ci_string = ();  # string needed to compute continuation indentation
        my @container_environment = ();    # BLOCK or LIST
        my $container_environment = '';
        my $im                    = -1;    # previous $i value
        my $num;
        my $ci_string_sum = ones_count($ci_string_in_tokenizer);

# Computing Token Indentation
#
#     The final section of the tokenizer forms tokens and also computes
#     parameters needed to find indentation.  It is much easier to do it
#     in the tokenizer than elsewhere.  Here is a brief description of how
#     indentation is computed.  Perl::Tidy computes indentation as the sum
#     of 2 terms:
#
#     (1) structural indentation, such as if/else/elsif blocks
#     (2) continuation indentation, such as long parameter call lists.
#
#     These are occasionally called primary and secondary indentation.
#
#     Structural indentation is introduced by tokens of type '{', although
#     the actual tokens might be '{', '(', or '['.  Structural indentation
#     is of two types: BLOCK and non-BLOCK.  Default structural indentation
#     is 4 characters if the standard indentation scheme is used.
#
#     Continuation indentation is introduced whenever a line at BLOCK level
#     is broken before its termination.  Default continuation indentation
#     is 2 characters in the standard indentation scheme.
#
#     Both types of indentation may be nested arbitrarily deep and
#     interlaced.  The distinction between the two is somewhat arbitrary.
#
#     For each token, we will define two variables which would apply if
#     the current statement were broken just before that token, so that
#     that token started a new line:
#
#     $level = the structural indentation level,
#     $ci_level = the continuation indentation level
#
#     The total indentation will be $level * (4 spaces) + $ci_level * (2 spaces),
#     assuming defaults.  However, in some special cases it is customary
#     to modify $ci_level from this strict value.
#
#     The total structural indentation is easy to compute by adding and
#     subtracting 1 from a saved value as types '{' and '}' are seen.  The
#     running value of this variable is $level_in_tokenizer.
#
#     The total continuation is much more difficult to compute, and requires
#     several variables.  These veriables are:
#
#     $ci_string_in_tokenizer = a string of 1's and 0's indicating, for
#       each indentation level, if there are intervening open secondary
#       structures just prior to that level.
#     $continuation_string_in_tokenizer = a string of 1's and 0's indicating
#       if the last token at that level is "continued", meaning that it
#       is not the first token of an expression.
#     $nesting_block_string = a string of 1's and 0's indicating, for each
#       indentation level, if the level is of type BLOCK or not.
#     $nesting_block_flag = the most recent 1 or 0 of $nesting_block_string
#     $nesting_list_string = a string of 1's and 0's indicating, for each
#       indentation level, if it is is appropriate for list formatting.
#       If so, continuation indentation is used to indent long list items.
#     $nesting_list_flag = the most recent 1 or 0 of $nesting_list_string
#     @{$rslevel_stack} = a stack of total nesting depths at each
#       structural indentation level, where "total nesting depth" means
#       the nesting depth that would occur if every nesting token -- '{', '[',
#       and '(' -- , regardless of context, is used to compute a nesting
#       depth.

        #my $nesting_block_flag = ($nesting_block_string =~ /1$/);
        #my $nesting_list_flag = ($nesting_list_string =~ /1$/);

        my ( $ci_string_i, $level_i, $nesting_block_string_i,
            $nesting_list_string_i, $nesting_token_string_i,
            $nesting_type_string_i, );

        foreach $i ( @{$routput_token_list} )
        {    # scan the list of pre-tokens indexes

            # self-checking for valid token types
            my $type                    = $routput_token_type->[$i];
            my $forced_indentation_flag = $routput_indent_flag->[$i];

            # See if we should undo the $forced_indentation_flag.
            # Forced indentation after 'if', 'unless', 'while' and 'until'
            # expressions without trailing parens is optional and doesn't
            # always look good.  It is usually okay for a trailing logical
            # expression, but if the expression is a function call, code block,
            # or some kind of list it puts in an unwanted extra indentation
            # level which is hard to remove.
            #
            # Example where extra indentation looks ok:
            # return 1
            #   if $det_a < 0 and $det_b > 0
            #       or $det_a > 0 and $det_b < 0;
            #
            # Example where extra indentation is not needed because
            # the eval brace also provides indentation:
            # print "not " if defined eval {
            #     reduce { die if $b > 2; $a + $b } 0, 1, 2, 3, 4;
            # };
            #
            # The following rule works fairly well:
            #   Undo the flag if the end of this line, or start of the next
            #   line, is an opening container token or a comma.
            # This almost always works, but if not after another pass it will
            # be stable.
            if ( $forced_indentation_flag && $type eq 'k' ) {
                my $ixlast  = -1;
                my $ilast   = $routput_token_list->[$ixlast];
                my $toklast = $routput_token_type->[$ilast];
                if ( $toklast eq '#' ) {
                    $ixlast--;
                    $ilast   = $routput_token_list->[$ixlast];
                    $toklast = $routput_token_type->[$ilast];
                }
                if ( $toklast eq 'b' ) {
                    $ixlast--;
                    $ilast   = $routput_token_list->[$ixlast];
                    $toklast = $routput_token_type->[$ilast];
                }
                if ( $toklast =~ /^[\{,]$/ ) {
                    $forced_indentation_flag = 0;
                }
                else {
                    ( $toklast, my $i_next ) =
                      find_next_nonblank_token( $max_token_index, $rtokens,
                        $max_token_index );
                    if ( $toklast =~ /^[\{,]$/ ) {
                        $forced_indentation_flag = 0;
                    }
                }
            }

            # if we are already in an indented if, see if we should outdent
            if ($indented_if_level) {

                # don't try to nest trailing if's - shouldn't happen
                if ( $type eq 'k' ) {
                    $forced_indentation_flag = 0;
                }

                # check for the normal case - outdenting at next ';'
                elsif ( $type eq ';' ) {
                    if ( $level_in_tokenizer == $indented_if_level ) {
                        $forced_indentation_flag = -1;
                        $indented_if_level       = 0;
                    }
                }

                # handle case of missing semicolon
                elsif ( $type eq '}' ) {
                    if ( $level_in_tokenizer == $indented_if_level ) {
                        $indented_if_level = 0;

                        # TBD: This could be a subroutine call
                        $level_in_tokenizer--;
                        if ( @{$rslevel_stack} > 1 ) {
                            pop( @{$rslevel_stack} );
                        }
                        if ( length($nesting_block_string) > 1 )
                        {    # true for valid script
                            chop $nesting_block_string;
                            chop $nesting_list_string;
                        }

                    }
                }
            }

            my $tok = $$rtokens[$i];   # the token, but ONLY if same as pretoken
            $level_i = $level_in_tokenizer;

            # This can happen by running perltidy on non-scripts
            # although it could also be bug introduced by programming change.
            # Perl silently accepts a 032 (^Z) and takes it as the end
            if ( !$is_valid_token_type{$type} ) {
                my $val = ord($type);
                warning(
                    "unexpected character decimal $val ($type) in script\n");
                $tokenizer_self->{_in_error} = 1;
            }

            # ----------------------------------------------------------------
            # TOKEN TYPE PATCHES
            #  output __END__, __DATA__, and format as type 'k' instead of ';'
            # to make html colors correct, etc.
            my $fix_type = $type;
            if ( $type eq ';' && $tok =~ /\w/ ) { $fix_type = 'k' }

            # output anonymous 'sub' as keyword
            if ( $type eq 't' && $tok eq 'sub' ) { $fix_type = 'k' }

            # -----------------------------------------------------------------

            $nesting_token_string_i = $nesting_token_string;
            $nesting_type_string_i  = $nesting_type_string;
            $nesting_block_string_i = $nesting_block_string;
            $nesting_list_string_i  = $nesting_list_string;

            # set primary indentation levels based on structural braces
            # Note: these are set so that the leading braces have a HIGHER
            # level than their CONTENTS, which is convenient for indentation
            # Also, define continuation indentation for each token.
            if ( $type eq '{' || $type eq 'L' || $forced_indentation_flag > 0 )
            {

                # use environment before updating
                $container_environment =
                    $nesting_block_flag ? 'BLOCK'
                  : $nesting_list_flag  ? 'LIST'
                  :                       "";

                # if the difference between total nesting levels is not 1,
                # there are intervening non-structural nesting types between
                # this '{' and the previous unclosed '{'
                my $intervening_secondary_structure = 0;
                if ( @{$rslevel_stack} ) {
                    $intervening_secondary_structure =
                      $slevel_in_tokenizer - $rslevel_stack->[-1];
                }

     # Continuation Indentation
     #
     # Having tried setting continuation indentation both in the formatter and
     # in the tokenizer, I can say that setting it in the tokenizer is much,
     # much easier.  The formatter already has too much to do, and can't
     # make decisions on line breaks without knowing what 'ci' will be at
     # arbitrary locations.
     #
     # But a problem with setting the continuation indentation (ci) here
     # in the tokenizer is that we do not know where line breaks will actually
     # be.  As a result, we don't know if we should propagate continuation
     # indentation to higher levels of structure.
     #
     # For nesting of only structural indentation, we never need to do this.
     # For example, in a long if statement, like this
     #
     #   if ( !$output_block_type[$i]
     #     && ($in_statement_continuation) )
     #   {           <--outdented
     #       do_something();
     #   }
     #
     # the second line has ci but we do normally give the lines within the BLOCK
     # any ci.  This would be true if we had blocks nested arbitrarily deeply.
     #
     # But consider something like this, where we have created a break after
     # an opening paren on line 1, and the paren is not (currently) a
     # structural indentation token:
     #
     # my $file = $menubar->Menubutton(
     #   qw/-text File -underline 0 -menuitems/ => [
     #       [
     #           Cascade    => '~View',
     #           -menuitems => [
     #           ...
     #
     # The second line has ci, so it would seem reasonable to propagate it
     # down, giving the third line 1 ci + 1 indentation.  This suggests the
     # following rule, which is currently used to propagating ci down: if there
     # are any non-structural opening parens (or brackets, or braces), before
     # an opening structural brace, then ci is propagated down, and otherwise
     # not.  The variable $intervening_secondary_structure contains this
     # information for the current token, and the string
     # "$ci_string_in_tokenizer" is a stack of previous values of this
     # variable.

                # save the current states
                push( @{$rslevel_stack}, 1 + $slevel_in_tokenizer );
                $level_in_tokenizer++;

                if ($forced_indentation_flag) {

                    # break BEFORE '?' when there is forced indentation
                    if ( $type eq '?' ) { $level_i = $level_in_tokenizer; }
                    if ( $type eq 'k' ) {
                        $indented_if_level = $level_in_tokenizer;
                    }

                    # do not change container environement here if we are not
                    # at a real list. Adding this check prevents "blinkers"
                    # often near 'unless" clauses, such as in the following
                    # code:
##          next
##            unless -e (
##                    $archive =
##                      File::Spec->catdir( $_, "auto", $root, "$sub$lib_ext" )
##            );

                    $nesting_block_string .= "$nesting_block_flag";
                }
                else {

                    if ( $routput_block_type->[$i] ) {
                        $nesting_block_flag = 1;
                        $nesting_block_string .= '1';
                    }
                    else {
                        $nesting_block_flag = 0;
                        $nesting_block_string .= '0';
                    }
                }

                # we will use continuation indentation within containers
                # which are not blocks and not logical expressions
                my $bit = 0;
                if ( !$routput_block_type->[$i] ) {

                    # propagate flag down at nested open parens
                    if ( $routput_container_type->[$i] eq '(' ) {
                        $bit = 1 if $nesting_list_flag;
                    }

                  # use list continuation if not a logical grouping
                  # /^(if|elsif|unless|while|and|or|not|&&|!|\|\||for|foreach)$/
                    else {
                        $bit = 1
                          unless
                          $is_logical_container{ $routput_container_type->[$i]
                          };
                    }
                }
                $nesting_list_string .= $bit;
                $nesting_list_flag = $bit;

                $ci_string_in_tokenizer .=
                  ( $intervening_secondary_structure != 0 ) ? '1' : '0';
                $ci_string_sum = ones_count($ci_string_in_tokenizer);
                $continuation_string_in_tokenizer .=
                  ( $in_statement_continuation > 0 ) ? '1' : '0';

   #  Sometimes we want to give an opening brace continuation indentation,
   #  and sometimes not.  For code blocks, we don't do it, so that the leading
   #  '{' gets outdented, like this:
   #
   #   if ( !$output_block_type[$i]
   #     && ($in_statement_continuation) )
   #   {           <--outdented
   #
   #  For other types, we will give them continuation indentation.  For example,
   #  here is how a list looks with the opening paren indented:
   #
   #     @LoL =
   #       ( [ "fred", "barney" ], [ "george", "jane", "elroy" ],
   #         [ "homer", "marge", "bart" ], );
   #
   #  This looks best when 'ci' is one-half of the indentation  (i.e., 2 and 4)

                my $total_ci = $ci_string_sum;
                if (
                    !$routput_block_type->[$i]    # patch: skip for BLOCK
                    && ($in_statement_continuation)
                    && !( $forced_indentation_flag && $type eq ':' )
                  )
                {
                    $total_ci += $in_statement_continuation
                      unless ( $ci_string_in_tokenizer =~ /1$/ );
                }

                $ci_string_i               = $total_ci;
                $in_statement_continuation = 0;
            }

            elsif ($type eq '}'
                || $type eq 'R'
                || $forced_indentation_flag < 0 )
            {

                # only a nesting error in the script would prevent popping here
                if ( @{$rslevel_stack} > 1 ) { pop( @{$rslevel_stack} ); }

                $level_i = --$level_in_tokenizer;

                # restore previous level values
                if ( length($nesting_block_string) > 1 )
                {    # true for valid script
                    chop $nesting_block_string;
                    $nesting_block_flag = ( $nesting_block_string =~ /1$/ );
                    chop $nesting_list_string;
                    $nesting_list_flag = ( $nesting_list_string =~ /1$/ );

                    chop $ci_string_in_tokenizer;
                    $ci_string_sum = ones_count($ci_string_in_tokenizer);

                    $in_statement_continuation =
                      chop $continuation_string_in_tokenizer;

                    # zero continuation flag at terminal BLOCK '}' which
                    # ends a statement.
                    if ( $routput_block_type->[$i] ) {

                        # ...These include non-anonymous subs
                        # note: could be sub ::abc { or sub 'abc
                        if ( $routput_block_type->[$i] =~ m/^sub\s*/gc ) {

                         # note: older versions of perl require the /gc modifier
                         # here or else the \G does not work.
                            if ( $routput_block_type->[$i] =~ /\G('|::|\w)/gc )
                            {
                                $in_statement_continuation = 0;
                            }
                        }

# ...and include all block types except user subs with
# block prototypes and these: (sort|grep|map|do|eval)
# /^(\}|\{|BEGIN|END|CHECK|INIT|AUTOLOAD|DESTROY|UNITCHECK|continue|;|if|elsif|else|unless|while|until|for|foreach)$/
                        elsif (
                            $is_zero_continuation_block_type{
                                $routput_block_type->[$i]
                            } )
                        {
                            $in_statement_continuation = 0;
                        }

                        # ..but these are not terminal types:
                        #     /^(sort|grep|map|do|eval)$/ )
                        elsif (
                            $is_not_zero_continuation_block_type{
                                $routput_block_type->[$i]
                            } )
                        {
                        }

                        # ..and a block introduced by a label
                        # /^\w+\s*:$/gc ) {
                        elsif ( $routput_block_type->[$i] =~ /:$/ ) {
                            $in_statement_continuation = 0;
                        }

                        # user function with block prototype
                        else {
                            $in_statement_continuation = 0;
                        }
                    }

                    # If we are in a list, then
                    # we must set continuatoin indentation at the closing
                    # paren of something like this (paren after $check):
                    #     assert(
                    #         __LINE__,
                    #         ( not defined $check )
                    #           or ref $check
                    #           or $check eq "new"
                    #           or $check eq "old",
                    #     );
                    elsif ( $tok eq ')' ) {
                        $in_statement_continuation = 1
                          if $routput_container_type->[$i] =~ /^[;,\{\}]$/;
                    }

                    elsif ( $tok eq ';' ) { $in_statement_continuation = 0 }
                }

                # use environment after updating
                $container_environment =
                    $nesting_block_flag ? 'BLOCK'
                  : $nesting_list_flag  ? 'LIST'
                  :                       "";
                $ci_string_i = $ci_string_sum + $in_statement_continuation;
                $nesting_block_string_i = $nesting_block_string;
                $nesting_list_string_i  = $nesting_list_string;
            }

            # not a structural indentation type..
            else {

                $container_environment =
                    $nesting_block_flag ? 'BLOCK'
                  : $nesting_list_flag  ? 'LIST'
                  :                       "";

                # zero the continuation indentation at certain tokens so
                # that they will be at the same level as its container.  For
                # commas, this simplifies the -lp indentation logic, which
                # counts commas.  For ?: it makes them stand out.
                if ($nesting_list_flag) {
                    if ( $type =~ /^[,\?\:]$/ ) {
                        $in_statement_continuation = 0;
                    }
                }

                # be sure binary operators get continuation indentation
                if (
                    $container_environment
                    && (   $type eq 'k' && $is_binary_keyword{$tok}
                        || $is_binary_type{$type} )
                  )
                {
                    $in_statement_continuation = 1;
                }

                # continuation indentation is sum of any open ci from previous
                # levels plus the current level
                $ci_string_i = $ci_string_sum + $in_statement_continuation;

                # update continuation flag ...
                # if this isn't a blank or comment..
                if ( $type ne 'b' && $type ne '#' ) {

                    # and we are in a BLOCK
                    if ($nesting_block_flag) {

                        # the next token after a ';' and label starts a new stmt
                        if ( $type eq ';' || $type eq 'J' ) {
                            $in_statement_continuation = 0;
                        }

                        # otherwise, we are continuing the current statement
                        else {
                            $in_statement_continuation = 1;
                        }
                    }

                    # if we are not in a BLOCK..
                    else {

                        # do not use continuation indentation if not list
                        # environment (could be within if/elsif clause)
                        if ( !$nesting_list_flag ) {
                            $in_statement_continuation = 0;
                        }

                       # otherwise, the next token after a ',' starts a new term
                        elsif ( $type eq ',' ) {
                            $in_statement_continuation = 0;
                        }

                        # otherwise, we are continuing the current term
                        else {
                            $in_statement_continuation = 1;
                        }
                    }
                }
            }

            if ( $level_in_tokenizer < 0 ) {
                unless ( $tokenizer_self->{_saw_negative_indentation} ) {
                    $tokenizer_self->{_saw_negative_indentation} = 1;
                    warning("Starting negative indentation\n");
                }
            }

            # set secondary nesting levels based on all continment token types
            # Note: these are set so that the nesting depth is the depth
            # of the PREVIOUS TOKEN, which is convenient for setting
            # the stength of token bonds
            my $slevel_i = $slevel_in_tokenizer;

            #    /^[L\{\(\[]$/
            if ( $is_opening_type{$type} ) {
                $slevel_in_tokenizer++;
                $nesting_token_string .= $tok;
                $nesting_type_string  .= $type;
            }

            #       /^[R\}\)\]]$/
            elsif ( $is_closing_type{$type} ) {
                $slevel_in_tokenizer--;
                my $char = chop $nesting_token_string;

                if ( $char ne $matching_start_token{$tok} ) {
                    $nesting_token_string .= $char . $tok;
                    $nesting_type_string  .= $type;
                }
                else {
                    chop $nesting_type_string;
                }
            }

            push( @block_type,            $routput_block_type->[$i] );
            push( @ci_string,             $ci_string_i );
            push( @container_environment, $container_environment );
            push( @container_type,        $routput_container_type->[$i] );
            push( @levels,                $level_i );
            push( @nesting_tokens,        $nesting_token_string_i );
            push( @nesting_types,         $nesting_type_string_i );
            push( @slevels,               $slevel_i );
            push( @token_type,            $fix_type );
            push( @type_sequence,         $routput_type_sequence->[$i] );
            push( @nesting_blocks,        $nesting_block_string );
            push( @nesting_lists,         $nesting_list_string );

            # now form the previous token
            if ( $im >= 0 ) {
                $num =
                  $$rtoken_map[$i] - $$rtoken_map[$im];    # how many characters

                if ( $num > 0 ) {
                    push( @tokens,
                        substr( $input_line, $$rtoken_map[$im], $num ) );
                }
            }
            $im = $i;
        }

        $num = length($input_line) - $$rtoken_map[$im];    # make the last token
        if ( $num > 0 ) {
            push( @tokens, substr( $input_line, $$rtoken_map[$im], $num ) );
        }

        $tokenizer_self->{_in_attribute_list} = $in_attribute_list;
        $tokenizer_self->{_in_quote}          = $in_quote;
        $tokenizer_self->{_quote_target} =
          $in_quote ? matching_end_token($quote_character) : "";
        $tokenizer_self->{_rhere_target_list} = $rhere_target_list;

        $line_of_tokens->{_rtoken_type}            = \@token_type;
        $line_of_tokens->{_rtokens}                = \@tokens;
        $line_of_tokens->{_rblock_type}            = \@block_type;
        $line_of_tokens->{_rcontainer_type}        = \@container_type;
        $line_of_tokens->{_rcontainer_environment} = \@container_environment;
        $line_of_tokens->{_rtype_sequence}         = \@type_sequence;
        $line_of_tokens->{_rlevels}                = \@levels;
        $line_of_tokens->{_rslevels}               = \@slevels;
        $line_of_tokens->{_rnesting_tokens}        = \@nesting_tokens;
        $line_of_tokens->{_rci_levels}             = \@ci_string;
        $line_of_tokens->{_rnesting_blocks}        = \@nesting_blocks;

        return;
    }
}    # end tokenize_this_line

#########i#############################################################
# Tokenizer routines which assist in identifying token types
#######################################################################

sub operator_expected {

    # Many perl symbols have two or more meanings.  For example, '<<'
    # can be a shift operator or a here-doc operator.  The
    # interpretation of these symbols depends on the current state of
    # the tokenizer, which may either be expecting a term or an
    # operator.  For this example, a << would be a shift if an operator
    # is expected, and a here-doc if a term is expected.  This routine
    # is called to make this decision for any current token.  It returns
    # one of three possible values:
    #
    #     OPERATOR - operator expected (or at least, not a term)
    #     UNKNOWN  - can't tell
    #     TERM     - a term is expected (or at least, not an operator)
    #
    # The decision is based on what has been seen so far.  This
    # information is stored in the "$last_nonblank_type" and
    # "$last_nonblank_token" variables.  For example, if the
    # $last_nonblank_type is '=~', then we are expecting a TERM, whereas
    # if $last_nonblank_type is 'n' (numeric), we are expecting an
    # OPERATOR.
    #
    # If a UNKNOWN is returned, the calling routine must guess. A major
    # goal of this tokenizer is to minimize the possibility of returning
    # UNKNOWN, because a wrong guess can spoil the formatting of a
    # script.
    #
    # adding NEW_TOKENS: it is critically important that this routine be
    # updated to allow it to determine if an operator or term is to be
    # expected after the new token.  Doing this simply involves adding
    # the new token character to one of the regexes in this routine or
    # to one of the hash lists
    # that it uses, which are initialized in the BEGIN section.
    # USES GLOBAL VARIABLES: $last_nonblank_type, $last_nonblank_token,
    # $statement_type

    my ( $prev_type, $tok, $next_type ) = @_;

    my $op_expected = UNKNOWN;

##print "tok=$tok last type=$last_nonblank_type last tok=$last_nonblank_token\n";

# Note: function prototype is available for token type 'U' for future
# program development.  It contains the leading and trailing parens,
# and no blanks.  It might be used to eliminate token type 'C', for
# example (prototype = '()'). Thus:
# if ($last_nonblank_type eq 'U') {
#     print "previous token=$last_nonblank_token  type=$last_nonblank_type prototype=$last_nonblank_prototype\n";
# }

    # A possible filehandle (or object) requires some care...
    if ( $last_nonblank_type eq 'Z' ) {

        # angle.t
        if ( $last_nonblank_token =~ /^[A-Za-z_]/ ) {
            $op_expected = UNKNOWN;
        }

        # For possible file handle like "$a", Perl uses weird parsing rules.
        # For example:
        # print $a/2,"/hi";   - division
        # print $a / 2,"/hi"; - division
        # print $a/ 2,"/hi";  - division
        # print $a /2,"/hi";  - pattern (and error)!
        elsif ( ( $prev_type eq 'b' ) && ( $next_type ne 'b' ) ) {
            $op_expected = TERM;
        }

        # Note when an operation is being done where a
        # filehandle might be expected, since a change in whitespace
        # could change the interpretation of the statement.
        else {
            if ( $tok =~ /^([x\/\+\-\*\%\&\.\?\<]|\>\>)$/ ) {
                complain("operator in print statement not recommended\n");
                $op_expected = OPERATOR;
            }
        }
    }

    # Check for smartmatch operator before preceding brace or square bracket.
    # For example, at the ? after the ] in the following expressions we are
    # expecting an operator:
    #
    # qr/3/ ~~ ['1234'] ? 1 : 0;
    # map { $_ ~~ [ '0', '1' ] ? 'x' : 'o' } @a;
    elsif ( $last_nonblank_type eq '}' && $last_nonblank_token eq '~~' ) {
        $op_expected = OPERATOR;
    }

    # handle something after 'do' and 'eval'
    elsif ( $is_block_operator{$last_nonblank_token} ) {

        # something like $a = eval "expression";
        #                          ^
        if ( $last_nonblank_type eq 'k' ) {
            $op_expected = TERM;    # expression or list mode following keyword
        }

        # something like $a = do { BLOCK } / 2;
        # or this ? after a smartmatch anonynomous hash or array reference:
        #   qr/3/ ~~ ['1234'] ? 1 : 0;
        #                                  ^
        else {
            $op_expected = OPERATOR;    # block mode following }
        }
    }

    # handle bare word..
    elsif ( $last_nonblank_type eq 'w' ) {

        # unfortunately, we can't tell what type of token to expect next
        # after most bare words
        $op_expected = UNKNOWN;
    }

    # operator, but not term possible after these types
    # Note: moved ')' from type to token because parens in list context
    # get marked as '{' '}' now.  This is a minor glitch in the following:
    #    my %opts = (ref $_[0] eq 'HASH') ? %{shift()} : ();
    #
    elsif (( $last_nonblank_type =~ /^[\]RnviQh]$/ )
        || ( $last_nonblank_token =~ /^(\)|\$|\-\>)/ ) )
    {
        $op_expected = OPERATOR;

        # in a 'use' statement, numbers and v-strings are not true
        # numbers, so to avoid incorrect error messages, we will
        # mark them as unknown for now (use.t)
        # TODO: it would be much nicer to create a new token V for VERSION
        # number in a use statement.  Then this could be a check on type V
        # and related patches which change $statement_type for '=>'
        # and ',' could be removed.  Further, it would clean things up to
        # scan the 'use' statement with a separate subroutine.
        if (   ( $statement_type eq 'use' )
            && ( $last_nonblank_type =~ /^[nv]$/ ) )
        {
            $op_expected = UNKNOWN;
        }

        # expecting VERSION or {} after package NAMESPACE
        elsif ($statement_type =~ /^package\b/
            && $last_nonblank_token =~ /^package\b/ )
        {
            $op_expected = TERM;
        }
    }

    # no operator after many keywords, such as "die", "warn", etc
    elsif ( $expecting_term_token{$last_nonblank_token} ) {

        # patch for dor.t (defined or).
        # perl functions which may be unary operators
        # TODO: This list is incomplete, and these should be put
        # into a hash.
        if (   $tok eq '/'
            && $next_type eq '/'
            && $last_nonblank_type eq 'k'
            && $last_nonblank_token =~ /^eof|undef|shift|pop$/ )
        {
            $op_expected = OPERATOR;
        }
        else {
            $op_expected = TERM;
        }
    }

    # no operator after things like + - **  (i.e., other operators)
    elsif ( $expecting_term_types{$last_nonblank_type} ) {
        $op_expected = TERM;
    }

    # a few operators, like "time", have an empty prototype () and so
    # take no parameters but produce a value to operate on
    elsif ( $expecting_operator_token{$last_nonblank_token} ) {
        $op_expected = OPERATOR;
    }

    # post-increment and decrement produce values to be operated on
    elsif ( $expecting_operator_types{$last_nonblank_type} ) {
        $op_expected = OPERATOR;
    }

    # no value to operate on after sub block
    elsif ( $last_nonblank_token =~ /^sub\s/ ) { $op_expected = TERM; }

    # a right brace here indicates the end of a simple block.
    # all non-structural right braces have type 'R'
    # all braces associated with block operator keywords have been given those
    # keywords as "last_nonblank_token" and caught above.
    # (This statement is order dependent, and must come after checking
    # $last_nonblank_token).
    elsif ( $last_nonblank_type eq '}' ) {

        # patch for dor.t (defined or).
        if (   $tok eq '/'
            && $next_type eq '/'
            && $last_nonblank_token eq ']' )
        {
            $op_expected = OPERATOR;
        }
        else {
            $op_expected = TERM;
        }
    }

    # something else..what did I forget?
    else {

        # collecting diagnostics on unknown operator types..see what was missed
        $op_expected = UNKNOWN;
        write_diagnostics(
"OP: unknown after type=$last_nonblank_type  token=$last_nonblank_token\n"
        );
    }

    TOKENIZER_DEBUG_FLAG_EXPECT && do {
        print STDOUT
"EXPECT: returns $op_expected for last type $last_nonblank_type token $last_nonblank_token\n";
    };
    return $op_expected;
}

sub new_statement_ok {

    # return true if the current token can start a new statement
    # USES GLOBAL VARIABLES: $last_nonblank_type

    return label_ok()    # a label would be ok here

      || $last_nonblank_type eq 'J';    # or we follow a label

}

sub label_ok {

    # Decide if a bare word followed by a colon here is a label
    # USES GLOBAL VARIABLES: $last_nonblank_token, $last_nonblank_type,
    # $brace_depth, @brace_type

    # if it follows an opening or closing code block curly brace..
    if ( ( $last_nonblank_token eq '{' || $last_nonblank_token eq '}' )
        && $last_nonblank_type eq $last_nonblank_token )
    {

        # it is a label if and only if the curly encloses a code block
        return $brace_type[$brace_depth];
    }

    # otherwise, it is a label if and only if it follows a ';' (real or fake)
    # or another label
    else {
        return ( $last_nonblank_type eq ';' || $last_nonblank_type eq 'J' );
    }
}

sub code_block_type {

    # Decide if this is a block of code, and its type.
    # Must be called only when $type = $token = '{'
    # The problem is to distinguish between the start of a block of code
    # and the start of an anonymous hash reference
    # Returns "" if not code block, otherwise returns 'last_nonblank_token'
    # to indicate the type of code block.  (For example, 'last_nonblank_token'
    # might be 'if' for an if block, 'else' for an else block, etc).
    # USES GLOBAL VARIABLES: $last_nonblank_token, $last_nonblank_type,
    # $last_nonblank_block_type, $brace_depth, @brace_type

    # handle case of multiple '{'s

# print "BLOCK_TYPE EXAMINING: type=$last_nonblank_type tok=$last_nonblank_token\n";

    my ( $i, $rtokens, $rtoken_type, $max_token_index ) = @_;
    if (   $last_nonblank_token eq '{'
        && $last_nonblank_type eq $last_nonblank_token )
    {

        # opening brace where a statement may appear is probably
        # a code block but might be and anonymous hash reference
        if ( $brace_type[$brace_depth] ) {
            return decide_if_code_block( $i, $rtokens, $rtoken_type,
                $max_token_index );
        }

        # cannot start a code block within an anonymous hash
        else {
            return "";
        }
    }

    elsif ( $last_nonblank_token eq ';' ) {

        # an opening brace where a statement may appear is probably
        # a code block but might be and anonymous hash reference
        return decide_if_code_block( $i, $rtokens, $rtoken_type,
            $max_token_index );
    }

    # handle case of '}{'
    elsif ($last_nonblank_token eq '}'
        && $last_nonblank_type eq $last_nonblank_token )
    {

        # a } { situation ...
        # could be hash reference after code block..(blktype1.t)
        if ($last_nonblank_block_type) {
            return decide_if_code_block( $i, $rtokens, $rtoken_type,
                $max_token_index );
        }

        # must be a block if it follows a closing hash reference
        else {
            return $last_nonblank_token;
        }
    }

    # NOTE: braces after type characters start code blocks, but for
    # simplicity these are not identified as such.  See also
    # sub is_non_structural_brace.
    # elsif ( $last_nonblank_type eq 't' ) {
    #    return $last_nonblank_token;
    # }

    # brace after label:
    elsif ( $last_nonblank_type eq 'J' ) {
        return $last_nonblank_token;
    }

# otherwise, look at previous token.  This must be a code block if
# it follows any of these:
# /^(BEGIN|END|CHECK|INIT|AUTOLOAD|DESTROY|UNITCHECK|continue|if|elsif|else|unless|do|while|until|eval|for|foreach|map|grep|sort)$/
    elsif ( $is_code_block_token{$last_nonblank_token} ) {

        # Bug Patch: Note that the opening brace after the 'if' in the following
        # snippet is an anonymous hash ref and not a code block!
        #   print 'hi' if { x => 1, }->{x};
        # We can identify this situation because the last nonblank type
        # will be a keyword (instead of a closing peren)
        if (   $last_nonblank_token =~ /^(if|unless)$/
            && $last_nonblank_type eq 'k' )
        {
            return "";
        }
        else {
            return $last_nonblank_token;
        }
    }

    # or a sub or package BLOCK
    elsif ( ( $last_nonblank_type eq 'i' || $last_nonblank_type eq 't' )
        && $last_nonblank_token =~ /^(sub|package)\b/ )
    {
        return $last_nonblank_token;
    }

    elsif ( $statement_type =~ /^(sub|package)\b/ ) {
        return $statement_type;
    }

    # user-defined subs with block parameters (like grep/map/eval)
    elsif ( $last_nonblank_type eq 'G' ) {
        return $last_nonblank_token;
    }

    # check bareword
    elsif ( $last_nonblank_type eq 'w' ) {
        return decide_if_code_block( $i, $rtokens, $rtoken_type,
            $max_token_index );
    }

    # anything else must be anonymous hash reference
    else {
        return "";
    }
}

sub decide_if_code_block {

    # USES GLOBAL VARIABLES: $last_nonblank_token
    my ( $i, $rtokens, $rtoken_type, $max_token_index ) = @_;
    my ( $next_nonblank_token, $i_next ) =
      find_next_nonblank_token( $i, $rtokens, $max_token_index );

    # we are at a '{' where a statement may appear.
    # We must decide if this brace starts an anonymous hash or a code
    # block.
    # return "" if anonymous hash, and $last_nonblank_token otherwise

    # initialize to be code BLOCK
    my $code_block_type = $last_nonblank_token;

    # Check for the common case of an empty anonymous hash reference:
    # Maybe something like sub { { } }
    if ( $next_nonblank_token eq '}' ) {
        $code_block_type = "";
    }

    else {

        # To guess if this '{' is an anonymous hash reference, look ahead
        # and test as follows:
        #
        # it is a hash reference if next come:
        #   - a string or digit followed by a comma or =>
        #   - bareword followed by =>
        # otherwise it is a code block
        #
        # Examples of anonymous hash ref:
        # {'aa',};
        # {1,2}
        #
        # Examples of code blocks:
        # {1; print "hello\n", 1;}
        # {$a,1};

        # We are only going to look ahead one more (nonblank/comment) line.
        # Strange formatting could cause a bad guess, but that's unlikely.
        my @pre_types  = @$rtoken_type[ $i + 1 .. $max_token_index ];
        my @pre_tokens = @$rtokens[ $i + 1 .. $max_token_index ];
        my ( $rpre_tokens, $rpre_types ) =
          peek_ahead_for_n_nonblank_pre_tokens(20);    # 20 is arbitrary but
                                                       # generous, and prevents
                                                       # wasting lots of
                                                       # time in mangled files
        if ( defined($rpre_types) && @$rpre_types ) {
            push @pre_types,  @$rpre_types;
            push @pre_tokens, @$rpre_tokens;
        }

        # put a sentinal token to simplify stopping the search
        push @pre_types, '}';

        my $jbeg = 0;
        $jbeg = 1 if $pre_types[0] eq 'b';

        # first look for one of these
        #  - bareword
        #  - bareword with leading -
        #  - digit
        #  - quoted string
        my $j = $jbeg;
        if ( $pre_types[$j] =~ /^[\'\"]/ ) {

            # find the closing quote; don't worry about escapes
            my $quote_mark = $pre_types[$j];
            for ( my $k = $j + 1 ; $k < $#pre_types ; $k++ ) {
                if ( $pre_types[$k] eq $quote_mark ) {
                    $j = $k + 1;
                    my $next = $pre_types[$j];
                    last;
                }
            }
        }
        elsif ( $pre_types[$j] eq 'd' ) {
            $j++;
        }
        elsif ( $pre_types[$j] eq 'w' ) {
            unless ( $is_keyword{ $pre_tokens[$j] } ) {
                $j++;
            }
        }
        elsif ( $pre_types[$j] eq '-' && $pre_types[ ++$j ] eq 'w' ) {
            $j++;
        }
        if ( $j > $jbeg ) {

            $j++ if $pre_types[$j] eq 'b';

            # it's a hash ref if a comma or => follow next
            if ( $pre_types[$j] eq ','
                || ( $pre_types[$j] eq '=' && $pre_types[ ++$j ] eq '>' ) )
            {
                $code_block_type = "";
            }
        }
    }

    return $code_block_type;
}

sub unexpected {

    # report unexpected token type and show where it is
    # USES GLOBAL VARIABLES: $tokenizer_self
    my ( $found, $expecting, $i_tok, $last_nonblank_i, $rpretoken_map,
        $rpretoken_type, $input_line )
      = @_;

    if ( ++$tokenizer_self->{_unexpected_error_count} <= MAX_NAG_MESSAGES ) {
        my $msg = "found $found where $expecting expected";
        my $pos = $$rpretoken_map[$i_tok];
        interrupt_logfile();
        my $input_line_number = $tokenizer_self->{_last_line_number};
        my ( $offset, $numbered_line, $underline ) =
          make_numbered_line( $input_line_number, $input_line, $pos );
        $underline = write_on_underline( $underline, $pos - $offset, '^' );

        my $trailer = "";
        if ( ( $i_tok > 0 ) && ( $last_nonblank_i >= 0 ) ) {
            my $pos_prev = $$rpretoken_map[$last_nonblank_i];
            my $num;
            if ( $$rpretoken_type[ $i_tok - 1 ] eq 'b' ) {
                $num = $$rpretoken_map[ $i_tok - 1 ] - $pos_prev;
            }
            else {
                $num = $pos - $pos_prev;
            }
            if ( $num > 40 ) { $num = 40; $pos_prev = $pos - 40; }

            $underline =
              write_on_underline( $underline, $pos_prev - $offset, '-' x $num );
            $trailer = " (previous token underlined)";
        }
        warning( $numbered_line . "\n" );
        warning( $underline . "\n" );
        warning( $msg . $trailer . "\n" );
        resume_logfile();
    }
}

sub is_non_structural_brace {

    # Decide if a brace or bracket is structural or non-structural
    # by looking at the previous token and type
    # USES GLOBAL VARIABLES: $last_nonblank_type, $last_nonblank_token

    # EXPERIMENTAL: Mark slices as structural; idea was to improve formatting.
    # Tentatively deactivated because it caused the wrong operator expectation
    # for this code:
    #      $user = @vars[1] / 100;
    # Must update sub operator_expected before re-implementing.
    # if ( $last_nonblank_type eq 'i' && $last_nonblank_token =~ /^@/ ) {
    #    return 0;
    # }

    # NOTE: braces after type characters start code blocks, but for
    # simplicity these are not identified as such.  See also
    # sub code_block_type
    # if ($last_nonblank_type eq 't') {return 0}

    # otherwise, it is non-structural if it is decorated
    # by type information.
    # For example, the '{' here is non-structural:   ${xxx}
    (
        $last_nonblank_token =~ /^([\$\@\*\&\%\)]|->|::)/

          # or if we follow a hash or array closing curly brace or bracket
          # For example, the second '{' in this is non-structural: $a{'x'}{'y'}
          # because the first '}' would have been given type 'R'
          || $last_nonblank_type =~ /^([R\]])$/
    );
}

#########i#############################################################
# Tokenizer routines for tracking container nesting depths
#######################################################################

# The following routines keep track of nesting depths of the nesting
# types, ( [ { and ?.  This is necessary for determining the indentation
# level, and also for debugging programs.  Not only do they keep track of
# nesting depths of the individual brace types, but they check that each
# of the other brace types is balanced within matching pairs.  For
# example, if the program sees this sequence:
#
#         {  ( ( ) }
#
# then it can determine that there is an extra left paren somewhere
# between the { and the }.  And so on with every other possible
# combination of outer and inner brace types.  For another
# example:
#
#         ( [ ..... ]  ] )
#
# which has an extra ] within the parens.
#
# The brace types have indexes 0 .. 3 which are indexes into
# the matrices.
#
# The pair ? : are treated as just another nesting type, with ? acting
# as the opening brace and : acting as the closing brace.
#
# The matrix
#
#         $depth_array[$a][$b][ $current_depth[$a] ] = $current_depth[$b];
#
# saves the nesting depth of brace type $b (where $b is either of the other
# nesting types) when brace type $a enters a new depth.  When this depth
# decreases, a check is made that the current depth of brace types $b is
# unchanged, or otherwise there must have been an error.  This can
# be very useful for localizing errors, particularly when perl runs to
# the end of a large file (such as this one) and announces that there
# is a problem somewhere.
#
# A numerical sequence number is maintained for every nesting type,
# so that each matching pair can be uniquely identified in a simple
# way.

sub increase_nesting_depth {
    my ( $aa, $pos ) = @_;

    # USES GLOBAL VARIABLES: $tokenizer_self, @current_depth,
    # @current_sequence_number, @depth_array, @starting_line_of_current_depth,
    # $statement_type
    my $bb;
    $current_depth[$aa]++;
    $total_depth++;
    $total_depth[$aa][ $current_depth[$aa] ] = $total_depth;
    my $input_line_number = $tokenizer_self->{_last_line_number};
    my $input_line        = $tokenizer_self->{_line_text};

    # Sequence numbers increment by number of items.  This keeps
    # a unique set of numbers but still allows the relative location
    # of any type to be determined.
    $nesting_sequence_number[$aa] += scalar(@closing_brace_names);
    my $seqno = $nesting_sequence_number[$aa];
    $current_sequence_number[$aa][ $current_depth[$aa] ] = $seqno;

    $starting_line_of_current_depth[$aa][ $current_depth[$aa] ] =
      [ $input_line_number, $input_line, $pos ];

    for $bb ( 0 .. $#closing_brace_names ) {
        next if ( $bb == $aa );
        $depth_array[$aa][$bb][ $current_depth[$aa] ] = $current_depth[$bb];
    }

    # set a flag for indenting a nested ternary statement
    my $indent = 0;
    if ( $aa == QUESTION_COLON ) {
        $nested_ternary_flag[ $current_depth[$aa] ] = 0;
        if ( $current_depth[$aa] > 1 ) {
            if ( $nested_ternary_flag[ $current_depth[$aa] - 1 ] == 0 ) {
                my $pdepth = $total_depth[$aa][ $current_depth[$aa] - 1 ];
                if ( $pdepth == $total_depth - 1 ) {
                    $indent = 1;
                    $nested_ternary_flag[ $current_depth[$aa] - 1 ] = -1;
                }
            }
        }
    }
    $nested_statement_type[$aa][ $current_depth[$aa] ] = $statement_type;
    $statement_type = "";
    return ( $seqno, $indent );
}

sub decrease_nesting_depth {

    my ( $aa, $pos ) = @_;

    # USES GLOBAL VARIABLES: $tokenizer_self, @current_depth,
    # @current_sequence_number, @depth_array, @starting_line_of_current_depth
    # $statement_type
    my $bb;
    my $seqno             = 0;
    my $input_line_number = $tokenizer_self->{_last_line_number};
    my $input_line        = $tokenizer_self->{_line_text};

    my $outdent = 0;
    $total_depth--;
    if ( $current_depth[$aa] > 0 ) {

        # set a flag for un-indenting after seeing a nested ternary statement
        $seqno = $current_sequence_number[$aa][ $current_depth[$aa] ];
        if ( $aa == QUESTION_COLON ) {
            $outdent = $nested_ternary_flag[ $current_depth[$aa] ];
        }
        $statement_type = $nested_statement_type[$aa][ $current_depth[$aa] ];

        # check that any brace types $bb contained within are balanced
        for $bb ( 0 .. $#closing_brace_names ) {
            next if ( $bb == $aa );

            unless ( $depth_array[$aa][$bb][ $current_depth[$aa] ] ==
                $current_depth[$bb] )
            {
                my $diff =
                  $current_depth[$bb] -
                  $depth_array[$aa][$bb][ $current_depth[$aa] ];

                # don't whine too many times
                my $saw_brace_error = get_saw_brace_error();
                if (
                    $saw_brace_error <= MAX_NAG_MESSAGES

                    # if too many closing types have occurred, we probably
                    # already caught this error
                    && ( ( $diff > 0 ) || ( $saw_brace_error <= 0 ) )
                  )
                {
                    interrupt_logfile();
                    my $rsl =
                      $starting_line_of_current_depth[$aa]
                      [ $current_depth[$aa] ];
                    my $sl  = $$rsl[0];
                    my $rel = [ $input_line_number, $input_line, $pos ];
                    my $el  = $$rel[0];
                    my ($ess);

                    if ( $diff == 1 || $diff == -1 ) {
                        $ess = '';
                    }
                    else {
                        $ess = 's';
                    }
                    my $bname =
                      ( $diff > 0 )
                      ? $opening_brace_names[$bb]
                      : $closing_brace_names[$bb];
                    write_error_indicator_pair( @$rsl, '^' );
                    my $msg = <<"EOM";
Found $diff extra $bname$ess between $opening_brace_names[$aa] on line $sl and $closing_brace_names[$aa] on line $el
EOM

                    if ( $diff > 0 ) {
                        my $rml =
                          $starting_line_of_current_depth[$bb]
                          [ $current_depth[$bb] ];
                        my $ml = $$rml[0];
                        $msg .=
"    The most recent un-matched $bname is on line $ml\n";
                        write_error_indicator_pair( @$rml, '^' );
                    }
                    write_error_indicator_pair( @$rel, '^' );
                    warning($msg);
                    resume_logfile();
                }
                increment_brace_error();
            }
        }
        $current_depth[$aa]--;
    }
    else {

        my $saw_brace_error = get_saw_brace_error();
        if ( $saw_brace_error <= MAX_NAG_MESSAGES ) {
            my $msg = <<"EOM";
There is no previous $opening_brace_names[$aa] to match a $closing_brace_names[$aa] on line $input_line_number
EOM
            indicate_error( $msg, $input_line_number, $input_line, $pos, '^' );
        }
        increment_brace_error();
    }
    return ( $seqno, $outdent );
}

sub check_final_nesting_depths {
    my ($aa);

    # USES GLOBAL VARIABLES: @current_depth, @starting_line_of_current_depth

    for $aa ( 0 .. $#closing_brace_names ) {

        if ( $current_depth[$aa] ) {
            my $rsl =
              $starting_line_of_current_depth[$aa][ $current_depth[$aa] ];
            my $sl  = $$rsl[0];
            my $msg = <<"EOM";
Final nesting depth of $opening_brace_names[$aa]s is $current_depth[$aa]
The most recent un-matched $opening_brace_names[$aa] is on line $sl
EOM
            indicate_error( $msg, @$rsl, '^' );
            increment_brace_error();
        }
    }
}

#########i#############################################################
# Tokenizer routines for looking ahead in input stream
#######################################################################

sub peek_ahead_for_n_nonblank_pre_tokens {

    # returns next n pretokens if they exist
    # returns undef's if hits eof without seeing any pretokens
    # USES GLOBAL VARIABLES: $tokenizer_self
    my $max_pretokens = shift;
    my $line;
    my $i = 0;
    my ( $rpre_tokens, $rmap, $rpre_types );

    while ( $line = $tokenizer_self->{_line_buffer_object}->peek_ahead( $i++ ) )
    {
        $line =~ s/^\s*//;    # trim leading blanks
        next if ( length($line) <= 0 );    # skip blank
        next if ( $line =~ /^#/ );         # skip comment
        ( $rpre_tokens, $rmap, $rpre_types ) =
          pre_tokenize( $line, $max_pretokens );
        last;
    }
    return ( $rpre_tokens, $rpre_types );
}

# look ahead for next non-blank, non-comment line of code
sub peek_ahead_for_nonblank_token {

    # USES GLOBAL VARIABLES: $tokenizer_self
    my ( $rtokens, $max_token_index ) = @_;
    my $line;
    my $i = 0;

    while ( $line = $tokenizer_self->{_line_buffer_object}->peek_ahead( $i++ ) )
    {
        $line =~ s/^\s*//;    # trim leading blanks
        next if ( length($line) <= 0 );    # skip blank
        next if ( $line =~ /^#/ );         # skip comment
        my ( $rtok, $rmap, $rtype ) =
          pre_tokenize( $line, 2 );        # only need 2 pre-tokens
        my $j = $max_token_index + 1;
        my $tok;

        foreach $tok (@$rtok) {
            last if ( $tok =~ "\n" );
            $$rtokens[ ++$j ] = $tok;
        }
        last;
    }
    return $rtokens;
}

#########i#############################################################
# Tokenizer guessing routines for ambiguous situations
#######################################################################

sub guess_if_pattern_or_conditional {

    # this routine is called when we have encountered a ? following an
    # unknown bareword, and we must decide if it starts a pattern or not
    # input parameters:
    #   $i - token index of the ? starting possible pattern
    # output parameters:
    #   $is_pattern = 0 if probably not pattern,  =1 if probably a pattern
    #   msg = a warning or diagnostic message
    # USES GLOBAL VARIABLES: $last_nonblank_token
    my ( $i, $rtokens, $rtoken_map, $max_token_index ) = @_;
    my $is_pattern = 0;
    my $msg        = "guessing that ? after $last_nonblank_token starts a ";

    if ( $i >= $max_token_index ) {
        $msg .= "conditional (no end to pattern found on the line)\n";
    }
    else {
        my $ibeg = $i;
        $i = $ibeg + 1;
        my $next_token = $$rtokens[$i];    # first token after ?

        # look for a possible ending ? on this line..
        my $in_quote        = 1;
        my $quote_depth     = 0;
        my $quote_character = '';
        my $quote_pos       = 0;
        my $quoted_string;
        (
            $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
            $quoted_string
          )
          = follow_quoted_string( $ibeg, $in_quote, $rtokens, $quote_character,
            $quote_pos, $quote_depth, $max_token_index );

        if ($in_quote) {

            # we didn't find an ending ? on this line,
            # so we bias towards conditional
            $is_pattern = 0;
            $msg .= "conditional (no ending ? on this line)\n";

            # we found an ending ?, so we bias towards a pattern
        }
        else {

            if ( pattern_expected( $i, $rtokens, $max_token_index ) >= 0 ) {
                $is_pattern = 1;
                $msg .= "pattern (found ending ? and pattern expected)\n";
            }
            else {
                $msg .= "pattern (uncertain, but found ending ?)\n";
            }
        }
    }
    return ( $is_pattern, $msg );
}

sub guess_if_pattern_or_division {

    # this routine is called when we have encountered a / following an
    # unknown bareword, and we must decide if it starts a pattern or is a
    # division
    # input parameters:
    #   $i - token index of the / starting possible pattern
    # output parameters:
    #   $is_pattern = 0 if probably division,  =1 if probably a pattern
    #   msg = a warning or diagnostic message
    # USES GLOBAL VARIABLES: $last_nonblank_token
    my ( $i, $rtokens, $rtoken_map, $max_token_index ) = @_;
    my $is_pattern = 0;
    my $msg        = "guessing that / after $last_nonblank_token starts a ";

    if ( $i >= $max_token_index ) {
        "division (no end to pattern found on the line)\n";
    }
    else {
        my $ibeg = $i;
        my $divide_expected =
          numerator_expected( $i, $rtokens, $max_token_index );
        $i = $ibeg + 1;
        my $next_token = $$rtokens[$i];    # first token after slash

        # look for a possible ending / on this line..
        my $in_quote        = 1;
        my $quote_depth     = 0;
        my $quote_character = '';
        my $quote_pos       = 0;
        my $quoted_string;
        (
            $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
            $quoted_string
          )
          = follow_quoted_string( $ibeg, $in_quote, $rtokens, $quote_character,
            $quote_pos, $quote_depth, $max_token_index );

        if ($in_quote) {

            # we didn't find an ending / on this line,
            # so we bias towards division
            if ( $divide_expected >= 0 ) {
                $is_pattern = 0;
                $msg .= "division (no ending / on this line)\n";
            }
            else {
                $msg        = "multi-line pattern (division not possible)\n";
                $is_pattern = 1;
            }

        }

        # we found an ending /, so we bias towards a pattern
        else {

            if ( pattern_expected( $i, $rtokens, $max_token_index ) >= 0 ) {

                if ( $divide_expected >= 0 ) {

                    if ( $i - $ibeg > 60 ) {
                        $msg .= "division (matching / too distant)\n";
                        $is_pattern = 0;
                    }
                    else {
                        $msg .= "pattern (but division possible too)\n";
                        $is_pattern = 1;
                    }
                }
                else {
                    $is_pattern = 1;
                    $msg .= "pattern (division not possible)\n";
                }
            }
            else {

                if ( $divide_expected >= 0 ) {
                    $is_pattern = 0;
                    $msg .= "division (pattern not possible)\n";
                }
                else {
                    $is_pattern = 1;
                    $msg .=
                      "pattern (uncertain, but division would not work here)\n";
                }
            }
        }
    }
    return ( $is_pattern, $msg );
}

# try to resolve here-doc vs. shift by looking ahead for
# non-code or the end token (currently only looks for end token)
# returns 1 if it is probably a here doc, 0 if not
sub guess_if_here_doc {

    # This is how many lines we will search for a target as part of the
    # guessing strategy.  It is a constant because there is probably
    # little reason to change it.
    # USES GLOBAL VARIABLES: $tokenizer_self, $current_package
    # %is_constant,
    use constant HERE_DOC_WINDOW => 40;

    my $next_token        = shift;
    my $here_doc_expected = 0;
    my $line;
    my $k   = 0;
    my $msg = "checking <<";

    while ( $line = $tokenizer_self->{_line_buffer_object}->peek_ahead( $k++ ) )
    {
        chomp $line;

        if ( $line =~ /^$next_token$/ ) {
            $msg .= " -- found target $next_token ahead $k lines\n";
            $here_doc_expected = 1;    # got it
            last;
        }
        last if ( $k >= HERE_DOC_WINDOW );
    }

    unless ($here_doc_expected) {

        if ( !defined($line) ) {
            $here_doc_expected = -1;    # hit eof without seeing target
            $msg .= " -- must be shift; target $next_token not in file\n";

        }
        else {                          # still unsure..taking a wild guess

            if ( !$is_constant{$current_package}{$next_token} ) {
                $here_doc_expected = 1;
                $msg .=
                  " -- guessing it's a here-doc ($next_token not a constant)\n";
            }
            else {
                $msg .=
                  " -- guessing it's a shift ($next_token is a constant)\n";
            }
        }
    }
    write_logfile_entry($msg);
    return $here_doc_expected;
}

#########i#############################################################
# Tokenizer Routines for scanning identifiers and related items
#######################################################################

sub scan_bare_identifier_do {

    # this routine is called to scan a token starting with an alphanumeric
    # variable or package separator, :: or '.
    # USES GLOBAL VARIABLES: $current_package, $last_nonblank_token,
    # $last_nonblank_type,@paren_type, $paren_depth

    my ( $input_line, $i, $tok, $type, $prototype, $rtoken_map,
        $max_token_index )
      = @_;
    my $i_begin = $i;
    my $package = undef;

    my $i_beg = $i;

    # we have to back up one pretoken at a :: since each : is one pretoken
    if ( $tok eq '::' ) { $i_beg-- }
    if ( $tok eq '->' ) { $i_beg-- }
    my $pos_beg = $$rtoken_map[$i_beg];
    pos($input_line) = $pos_beg;

    #  Examples:
    #   A::B::C
    #   A::
    #   ::A
    #   A'B
    if ( $input_line =~ m/\G\s*((?:\w*(?:'|::)))*(?:(?:->)?(\w+))?/gc ) {

        my $pos  = pos($input_line);
        my $numc = $pos - $pos_beg;
        $tok = substr( $input_line, $pos_beg, $numc );

        # type 'w' includes anything without leading type info
        # ($,%,@,*) including something like abc::def::ghi
        $type = 'w';

        my $sub_name = "";
        if ( defined($2) ) { $sub_name = $2; }
        if ( defined($1) ) {
            $package = $1;

            # patch: don't allow isolated package name which just ends
            # in the old style package separator (single quote).  Example:
            #   use CGI':all';
            if ( !($sub_name) && substr( $package, -1, 1 ) eq '\'' ) {
                $pos--;
            }

            $package =~ s/\'/::/g;
            if ( $package =~ /^\:/ ) { $package = 'main' . $package }
            $package =~ s/::$//;
        }
        else {
            $package = $current_package;

            if ( $is_keyword{$tok} ) {
                $type = 'k';
            }
        }

        # if it is a bareword..
        if ( $type eq 'w' ) {

            # check for v-string with leading 'v' type character
            # (This seems to have presidence over filehandle, type 'Y')
            if ( $tok =~ /^v\d[_\d]*$/ ) {

                # we only have the first part - something like 'v101' -
                # look for more
                if ( $input_line =~ m/\G(\.\d[_\d]*)+/gc ) {
                    $pos  = pos($input_line);
                    $numc = $pos - $pos_beg;
                    $tok  = substr( $input_line, $pos_beg, $numc );
                }
                $type = 'v';

                # warn if this version can't handle v-strings
                report_v_string($tok);
            }

            elsif ( $is_constant{$package}{$sub_name} ) {
                $type = 'C';
            }

            # bareword after sort has implied empty prototype; for example:
            # @sorted = sort numerically ( 53, 29, 11, 32, 7 );
            # This has priority over whatever the user has specified.
            elsif ($last_nonblank_token eq 'sort'
                && $last_nonblank_type eq 'k' )
            {
                $type = 'Z';
            }

            # Note: strangely, perl does not seem to really let you create
            # functions which act like eval and do, in the sense that eval
            # and do may have operators following the final }, but any operators
            # that you create with prototype (&) apparently do not allow
            # trailing operators, only terms.  This seems strange.
            # If this ever changes, here is the update
            # to make perltidy behave accordingly:

            # elsif ( $is_block_function{$package}{$tok} ) {
            #    $tok='eval'; # patch to do braces like eval  - doesn't work
            #    $type = 'k';
            #}
            # FIXME: This could become a separate type to allow for different
            # future behavior:
            elsif ( $is_block_function{$package}{$sub_name} ) {
                $type = 'G';
            }

            elsif ( $is_block_list_function{$package}{$sub_name} ) {
                $type = 'G';
            }
            elsif ( $is_user_function{$package}{$sub_name} ) {
                $type      = 'U';
                $prototype = $user_function_prototype{$package}{$sub_name};
            }

            # check for indirect object
            elsif (

                # added 2001-03-27: must not be followed immediately by '('
                # see fhandle.t
                ( $input_line !~ m/\G\(/gc )

                # and
                && (

                    # preceded by keyword like 'print', 'printf' and friends
                    $is_indirect_object_taker{$last_nonblank_token}

                    # or preceded by something like 'print(' or 'printf('
                    || (
                        ( $last_nonblank_token eq '(' )
                        && $is_indirect_object_taker{ $paren_type[$paren_depth]
                        }

                    )
                )
              )
            {

                # may not be indirect object unless followed by a space
                if ( $input_line =~ m/\G\s+/gc ) {
                    $type = 'Y';

                    # Abandon Hope ...
                    # Perl's indirect object notation is a very bad
                    # thing and can cause subtle bugs, especially for
                    # beginning programmers.  And I haven't even been
                    # able to figure out a sane warning scheme which
                    # doesn't get in the way of good scripts.

                    # Complain if a filehandle has any lower case
                    # letters.  This is suggested good practice.
                    # Use 'sub_name' because something like
                    # main::MYHANDLE is ok for filehandle
                    if ( $sub_name =~ /[a-z]/ ) {

                        # could be bug caused by older perltidy if
                        # followed by '('
                        if ( $input_line =~ m/\G\s*\(/gc ) {
                            complain(
"Caution: unknown word '$tok' in indirect object slot\n"
                            );
                        }
                    }
                }

                # bareword not followed by a space -- may not be filehandle
                # (may be function call defined in a 'use' statement)
                else {
                    $type = 'Z';
                }
            }
        }

        # Now we must convert back from character position
        # to pre_token index.
        # I don't think an error flag can occur here ..but who knows
        my $error;
        ( $i, $error ) =
          inverse_pretoken_map( $i, $pos, $rtoken_map, $max_token_index );
        if ($error) {
            warning("scan_bare_identifier: Possibly invalid tokenization\n");
        }
    }

    # no match but line not blank - could be syntax error
    # perl will take '::' alone without complaint
    else {
        $type = 'w';

        # change this warning to log message if it becomes annoying
        warning("didn't find identifier after leading ::\n");
    }
    return ( $i, $tok, $type, $prototype );
}

sub scan_id_do {

# This is the new scanner and will eventually replace scan_identifier.
# Only type 'sub' and 'package' are implemented.
# Token types $ * % @ & -> are not yet implemented.
#
# Scan identifier following a type token.
# The type of call depends on $id_scan_state: $id_scan_state = ''
# for starting call, in which case $tok must be the token defining
# the type.
#
# If the type token is the last nonblank token on the line, a value
# of $id_scan_state = $tok is returned, indicating that further
# calls must be made to get the identifier.  If the type token is
# not the last nonblank token on the line, the identifier is
# scanned and handled and a value of '' is returned.
# USES GLOBAL VARIABLES: $current_package, $last_nonblank_token, $in_attribute_list,
# $statement_type, $tokenizer_self

    my ( $input_line, $i, $tok, $rtokens, $rtoken_map, $id_scan_state,
        $max_token_index )
      = @_;
    my $type = '';
    my ( $i_beg, $pos_beg );

    #print "NSCAN:entering i=$i, tok=$tok, type=$type, state=$id_scan_state\n";
    #my ($a,$b,$c) = caller;
    #print "NSCAN: scan_id called with tok=$tok $a $b $c\n";

    # on re-entry, start scanning at first token on the line
    if ($id_scan_state) {
        $i_beg = $i;
        $type  = '';
    }

    # on initial entry, start scanning just after type token
    else {
        $i_beg         = $i + 1;
        $id_scan_state = $tok;
        $type          = 't';
    }

    # find $i_beg = index of next nonblank token,
    # and handle empty lines
    my $blank_line          = 0;
    my $next_nonblank_token = $$rtokens[$i_beg];
    if ( $i_beg > $max_token_index ) {
        $blank_line = 1;
    }
    else {

        # only a '#' immediately after a '$' is not a comment
        if ( $next_nonblank_token eq '#' ) {
            unless ( $tok eq '$' ) {
                $blank_line = 1;
            }
        }

        if ( $next_nonblank_token =~ /^\s/ ) {
            ( $next_nonblank_token, $i_beg ) =
              find_next_nonblank_token_on_this_line( $i_beg, $rtokens,
                $max_token_index );
            if ( $next_nonblank_token =~ /(^#|^\s*$)/ ) {
                $blank_line = 1;
            }
        }
    }

    # handle non-blank line; identifier, if any, must follow
    unless ($blank_line) {

        if ( $id_scan_state eq 'sub' ) {
            ( $i, $tok, $type, $id_scan_state ) = do_scan_sub(
                $input_line, $i,             $i_beg,
                $tok,        $type,          $rtokens,
                $rtoken_map, $id_scan_state, $max_token_index
            );
        }

        elsif ( $id_scan_state eq 'package' ) {
            ( $i, $tok, $type ) =
              do_scan_package( $input_line, $i, $i_beg, $tok, $type, $rtokens,
                $rtoken_map, $max_token_index );
            $id_scan_state = '';
        }

        else {
            warning("invalid token in scan_id: $tok\n");
            $id_scan_state = '';
        }
    }

    if ( $id_scan_state && ( !defined($type) || !$type ) ) {

        # shouldn't happen:
        warning(
"Program bug in scan_id: undefined type but scan_state=$id_scan_state\n"
        );
        report_definite_bug();
    }

    TOKENIZER_DEBUG_FLAG_NSCAN && do {
        print STDOUT
          "NSCAN: returns i=$i, tok=$tok, type=$type, state=$id_scan_state\n";
    };
    return ( $i, $tok, $type, $id_scan_state );
}

sub check_prototype {
    my ( $proto, $package, $subname ) = @_;
    return unless ( defined($package) && defined($subname) );
    if ( defined($proto) ) {
        $proto =~ s/^\s*\(\s*//;
        $proto =~ s/\s*\)$//;
        if ($proto) {
            $is_user_function{$package}{$subname}        = 1;
            $user_function_prototype{$package}{$subname} = "($proto)";

            # prototypes containing '&' must be treated specially..
            if ( $proto =~ /\&/ ) {

                # right curly braces of prototypes ending in
                # '&' may be followed by an operator
                if ( $proto =~ /\&$/ ) {
                    $is_block_function{$package}{$subname} = 1;
                }

                # right curly braces of prototypes NOT ending in
                # '&' may NOT be followed by an operator
                elsif ( $proto !~ /\&$/ ) {
                    $is_block_list_function{$package}{$subname} = 1;
                }
            }
        }
        else {
            $is_constant{$package}{$subname} = 1;
        }
    }
    else {
        $is_user_function{$package}{$subname} = 1;
    }
}

sub do_scan_package {

    # do_scan_package parses a package name
    # it is called with $i_beg equal to the index of the first nonblank
    # token following a 'package' token.
    # USES GLOBAL VARIABLES: $current_package,

    # package NAMESPACE
    # package NAMESPACE VERSION
    # package NAMESPACE BLOCK
    # package NAMESPACE VERSION BLOCK
    #
    # If VERSION is provided, package sets the $VERSION variable in the given
    # namespace to a version object with the VERSION provided. VERSION must be
    # a "strict" style version number as defined by the version module: a
    # positive decimal number (integer or decimal-fraction) without
    # exponentiation or else a dotted-decimal v-string with a leading 'v'
    # character and at least three components.
    # reference http://perldoc.perl.org/functions/package.html

    my ( $input_line, $i, $i_beg, $tok, $type, $rtokens, $rtoken_map,
        $max_token_index )
      = @_;
    my $package = undef;
    my $pos_beg = $$rtoken_map[$i_beg];
    pos($input_line) = $pos_beg;

    # handle non-blank line; package name, if any, must follow
    if ( $input_line =~ m/\G\s*((?:\w*(?:'|::))*\w+)/gc ) {
        $package = $1;
        $package = ( defined($1) && $1 ) ? $1 : 'main';
        $package =~ s/\'/::/g;
        if ( $package =~ /^\:/ ) { $package = 'main' . $package }
        $package =~ s/::$//;
        my $pos  = pos($input_line);
        my $numc = $pos - $pos_beg;
        $tok = 'package ' . substr( $input_line, $pos_beg, $numc );
        $type = 'i';

        # Now we must convert back from character position
        # to pre_token index.
        # I don't think an error flag can occur here ..but ?
        my $error;
        ( $i, $error ) =
          inverse_pretoken_map( $i, $pos, $rtoken_map, $max_token_index );
        if ($error) { warning("Possibly invalid package\n") }
        $current_package = $package;

        # we should now have package NAMESPACE
        # now expecting VERSION, BLOCK, or ; to follow ...
        # package NAMESPACE VERSION
        # package NAMESPACE BLOCK
        # package NAMESPACE VERSION BLOCK
        my ( $next_nonblank_token, $i_next ) =
          find_next_nonblank_token( $i, $rtokens, $max_token_index );
        if ( $next_nonblank_token =~ /^[v\.\d;\{\}]$/ ) {
            $statement_type = $tok;
        }
        else {
            warning(
                "Unexpected '$next_nonblank_token' after package name '$tok'\n"
            );
        }
    }

    # no match but line not blank --
    # could be a label with name package, like package:  , for example.
    else {
        $type = 'k';
    }

    return ( $i, $tok, $type );
}

sub scan_identifier_do {

    # This routine assembles tokens into identifiers.  It maintains a
    # scan state, id_scan_state.  It updates id_scan_state based upon
    # current id_scan_state and token, and returns an updated
    # id_scan_state and the next index after the identifier.
    # USES GLOBAL VARIABLES: $context, $last_nonblank_token,
    # $last_nonblank_type

    my ( $i, $id_scan_state, $identifier, $rtokens, $max_token_index,
        $expecting )
      = @_;
    my $i_begin   = $i;
    my $type      = '';
    my $tok_begin = $$rtokens[$i_begin];
    if ( $tok_begin eq ':' ) { $tok_begin = '::' }
    my $id_scan_state_begin = $id_scan_state;
    my $identifier_begin    = $identifier;
    my $tok                 = $tok_begin;
    my $message             = "";

    # these flags will be used to help figure out the type:
    my $saw_alpha = ( $tok =~ /^[A-Za-z_]/ );
    my $saw_type;

    # allow old package separator (') except in 'use' statement
    my $allow_tick = ( $last_nonblank_token ne 'use' );

    # get started by defining a type and a state if necessary
    unless ($id_scan_state) {
        $context = UNKNOWN_CONTEXT;

        # fixup for digraph
        if ( $tok eq '>' ) {
            $tok       = '->';
            $tok_begin = $tok;
        }
        $identifier = $tok;

        if ( $tok eq '$' || $tok eq '*' ) {
            $id_scan_state = '$';
            $context       = SCALAR_CONTEXT;
        }
        elsif ( $tok eq '%' || $tok eq '@' ) {
            $id_scan_state = '$';
            $context       = LIST_CONTEXT;
        }
        elsif ( $tok eq '&' ) {
            $id_scan_state = '&';
        }
        elsif ( $tok eq 'sub' or $tok eq 'package' ) {
            $saw_alpha     = 0;     # 'sub' is considered type info here
            $id_scan_state = '$';
            $identifier .= ' ';     # need a space to separate sub from sub name
        }
        elsif ( $tok eq '::' ) {
            $id_scan_state = 'A';
        }
        elsif ( $tok =~ /^[A-Za-z_]/ ) {
            $id_scan_state = ':';
        }
        elsif ( $tok eq '->' ) {
            $id_scan_state = '$';
        }
        else {

            # shouldn't happen
            my ( $a, $b, $c ) = caller;
            warning("Program Bug: scan_identifier given bad token = $tok \n");
            warning("   called from sub $a  line: $c\n");
            report_definite_bug();
        }
        $saw_type = !$saw_alpha;
    }
    else {
        $i--;
        $saw_type = ( $tok =~ /([\$\%\@\*\&])/ );
    }

    # now loop to gather the identifier
    my $i_save = $i;

    while ( $i < $max_token_index ) {
        $i_save = $i unless ( $tok =~ /^\s*$/ );
        $tok = $$rtokens[ ++$i ];

        if ( ( $tok eq ':' ) && ( $$rtokens[ $i + 1 ] eq ':' ) ) {
            $tok = '::';
            $i++;
        }

        if ( $id_scan_state eq '$' ) {    # starting variable name

            if ( $tok eq '$' ) {

                $identifier .= $tok;

                # we've got a punctuation variable if end of line (punct.t)
                if ( $i == $max_token_index ) {
                    $type          = 'i';
                    $id_scan_state = '';
                    last;
                }
            }
            elsif ( $tok =~ /^[A-Za-z_]/ ) {    # alphanumeric ..
                $saw_alpha     = 1;
                $id_scan_state = ':';           # now need ::
                $identifier .= $tok;
            }
            elsif ( $tok eq "'" && $allow_tick ) {    # alphanumeric ..
                $saw_alpha     = 1;
                $id_scan_state = ':';                 # now need ::
                $identifier .= $tok;

                # Perl will accept leading digits in identifiers,
                # although they may not always produce useful results.
                # Something like $main::0 is ok.  But this also works:
                #
                #  sub howdy::123::bubba{ print "bubba $54321!\n" }
                #  howdy::123::bubba();
                #
            }
            elsif ( $tok =~ /^[0-9]/ ) {    # numeric
                $saw_alpha     = 1;
                $id_scan_state = ':';       # now need ::
                $identifier .= $tok;
            }
            elsif ( $tok eq '::' ) {
                $id_scan_state = 'A';
                $identifier .= $tok;
            }
            elsif ( ( $tok eq '#' ) && ( $identifier eq '$' ) ) {    # $#array
                $identifier .= $tok;    # keep same state, a $ could follow
            }
            elsif ( $tok eq '{' ) {

                # check for something like ${#} or ${�}
                ##if (   $identifier eq '$'
                if (
                    (
                           $identifier eq '$'
                        || $identifier eq '@'
                        || $identifier eq '$#'
                    )
                    && $i + 2 <= $max_token_index
                    && $$rtokens[ $i + 2 ] eq '}'
                    && $$rtokens[ $i + 1 ] !~ /[\s\w]/
                  )
                {
                    my $next2 = $$rtokens[ $i + 2 ];
                    my $next1 = $$rtokens[ $i + 1 ];
                    $identifier .= $tok . $next1 . $next2;
                    $i += 2;
                    $id_scan_state = '';
                    last;
                }

                # skip something like ${xxx} or ->{
                $id_scan_state = '';

                # if this is the first token of a line, any tokens for this
                # identifier have already been accumulated
                if ( $identifier eq '$' || $i == 0 ) { $identifier = ''; }
                $i = $i_save;
                last;
            }

            # space ok after leading $ % * & @
            elsif ( $tok =~ /^\s*$/ ) {

                if ( $identifier =~ /^[\$\%\*\&\@]/ ) {

                    if ( length($identifier) > 1 ) {
                        $id_scan_state = '';
                        $i             = $i_save;
                        $type          = 'i';    # probably punctuation variable
                        last;
                    }
                    else {

                        # spaces after $'s are common, and space after @
                        # is harmless, so only complain about space
                        # after other type characters. Space after $ and
                        # @ will be removed in formatting.  Report space
                        # after % and * because they might indicate a
                        # parsing error.  In other words '% ' might be a
                        # modulo operator.  Delete this warning if it
                        # gets annoying.
                        if ( $identifier !~ /^[\@\$]$/ ) {
                            $message =
                              "Space in identifier, following $identifier\n";
                        }
                    }
                }

                # else:
                # space after '->' is ok
            }
            elsif ( $tok eq '^' ) {

                # check for some special variables like $^W
                if ( $identifier =~ /^[\$\*\@\%]$/ ) {
                    $identifier .= $tok;
                    $id_scan_state = 'A';

                    # Perl accepts '$^]' or '@^]', but
                    # there must not be a space before the ']'.
                    my $next1 = $$rtokens[ $i + 1 ];
                    if ( $next1 eq ']' ) {
                        $i++;
                        $identifier .= $next1;
                        $id_scan_state = "";
                        last;
                    }
                }
                else {
                    $id_scan_state = '';
                }
            }
            else {    # something else

                # check for various punctuation variables
                if ( $identifier =~ /^[\$\*\@\%]$/ ) {
                    $identifier .= $tok;
                }

                elsif ( $identifier eq '$#' ) {

                    if ( $tok eq '{' ) { $type = 'i'; $i = $i_save }

                    # perl seems to allow just these: $#: $#- $#+
                    elsif ( $tok =~ /^[\:\-\+]$/ ) {
                        $type = 'i';
                        $identifier .= $tok;
                    }
                    else {
                        $i = $i_save;
                        write_logfile_entry( 'Use of $# is deprecated' . "\n" );
                    }
                }
                elsif ( $identifier eq '$$' ) {

                    # perl does not allow references to punctuation
                    # variables without braces.  For example, this
                    # won't work:
                    #  $:=\4;
                    #  $a = $$:;
                    # You would have to use
                    #  $a = ${$:};

                    $i = $i_save;
                    if   ( $tok eq '{' ) { $type = 't' }
                    else                 { $type = 'i' }
                }
                elsif ( $identifier eq '->' ) {
                    $i = $i_save;
                }
                else {
                    $i = $i_save;
                    if ( length($identifier) == 1 ) { $identifier = ''; }
                }
                $id_scan_state = '';
                last;
            }
        }
        elsif ( $id_scan_state eq '&' ) {    # starting sub call?

            if ( $tok =~ /^[\$A-Za-z_]/ ) {    # alphanumeric ..
                $id_scan_state = ':';          # now need ::
                $saw_alpha     = 1;
                $identifier .= $tok;
            }
            elsif ( $tok eq "'" && $allow_tick ) {    # alphanumeric ..
                $id_scan_state = ':';                 # now need ::
                $saw_alpha     = 1;
                $identifier .= $tok;
            }
            elsif ( $tok =~ /^[0-9]/ ) {    # numeric..see comments above
                $id_scan_state = ':';       # now need ::
                $saw_alpha     = 1;
                $identifier .= $tok;
            }
            elsif ( $tok =~ /^\s*$/ ) {     # allow space
            }
            elsif ( $tok eq '::' ) {        # leading ::
                $id_scan_state = 'A';       # accept alpha next
                $identifier .= $tok;
            }
            elsif ( $tok eq '{' ) {
                if ( $identifier eq '&' || $i == 0 ) { $identifier = ''; }
                $i             = $i_save;
                $id_scan_state = '';
                last;
            }
            else {

                # punctuation variable?
                # testfile: cunningham4.pl
                #
                # We have to be careful here.  If we are in an unknown state,
                # we will reject the punctuation variable.  In the following
                # example the '&' is a binary opeator but we are in an unknown
                # state because there is no sigil on 'Prima', so we don't
                # know what it is.  But it is a bad guess that
                # '&~' is a punction variable.
                # $self->{text}->{colorMap}->[
                #   Prima::PodView::COLOR_CODE_FOREGROUND
                #   & ~tb::COLOR_INDEX ] =
                #   $sec->{ColorCode}
                if ( $identifier eq '&' && $expecting ) {
                    $identifier .= $tok;
                }
                else {
                    $identifier = '';
                    $i          = $i_save;
                    $type       = '&';
                }
                $id_scan_state = '';
                last;
            }
        }
        elsif ( $id_scan_state eq 'A' ) {    # looking for alpha (after ::)

            if ( $tok =~ /^[A-Za-z_]/ ) {    # found it
                $identifier .= $tok;
                $id_scan_state = ':';        # now need ::
                $saw_alpha     = 1;
            }
            elsif ( $tok eq "'" && $allow_tick ) {
                $identifier .= $tok;
                $id_scan_state = ':';        # now need ::
                $saw_alpha     = 1;
            }
            elsif ( $tok =~ /^[0-9]/ ) {     # numeric..see comments above
                $identifier .= $tok;
                $id_scan_state = ':';        # now need ::
                $saw_alpha     = 1;
            }
            elsif ( ( $identifier =~ /^sub / ) && ( $tok =~ /^\s*$/ ) ) {
                $id_scan_state = '(';
                $identifier .= $tok;
            }
            elsif ( ( $identifier =~ /^sub / ) && ( $tok eq '(' ) ) {
                $id_scan_state = ')';
                $identifier .= $tok;
            }
            else {
                $id_scan_state = '';
                $i             = $i_save;
                last;
            }
        }
        elsif ( $id_scan_state eq ':' ) {    # looking for :: after alpha

            if ( $tok eq '::' ) {            # got it
                $identifier .= $tok;
                $id_scan_state = 'A';        # now require alpha
            }
            elsif ( $tok =~ /^[A-Za-z_]/ ) {    # more alphanumeric is ok here
                $identifier .= $tok;
                $id_scan_state = ':';           # now need ::
                $saw_alpha     = 1;
            }
            elsif ( $tok =~ /^[0-9]/ ) {        # numeric..see comments above
                $identifier .= $tok;
                $id_scan_state = ':';           # now need ::
                $saw_alpha     = 1;
            }
            elsif ( $tok eq "'" && $allow_tick ) {    # tick

                if ( $is_keyword{$identifier} ) {
                    $id_scan_state = '';              # that's all
                    $i             = $i_save;
                }
                else {
                    $identifier .= $tok;
                }
            }
            elsif ( ( $identifier =~ /^sub / ) && ( $tok =~ /^\s*$/ ) ) {
                $id_scan_state = '(';
                $identifier .= $tok;
            }
            elsif ( ( $identifier =~ /^sub / ) && ( $tok eq '(' ) ) {
                $id_scan_state = ')';
                $identifier .= $tok;
            }
            else {
                $id_scan_state = '';        # that's all
                $i             = $i_save;
                last;
            }
        }
        elsif ( $id_scan_state eq '(' ) {    # looking for ( of prototype

            if ( $tok eq '(' ) {             # got it
                $identifier .= $tok;
                $id_scan_state = ')';        # now find the end of it
            }
            elsif ( $tok =~ /^\s*$/ ) {      # blank - keep going
                $identifier .= $tok;
            }
            else {
                $id_scan_state = '';         # that's all - no prototype
                $i             = $i_save;
                last;
            }
        }
        elsif ( $id_scan_state eq ')' ) {    # looking for ) to end

            if ( $tok eq ')' ) {             # got it
                $identifier .= $tok;
                $id_scan_state = '';         # all done
                last;
            }
            elsif ( $tok =~ /^[\s\$\%\\\*\@\&\;]/ ) {
                $identifier .= $tok;
            }
            else {    # probable error in script, but keep going
                warning("Unexpected '$tok' while seeking end of prototype\n");
                $identifier .= $tok;
            }
        }
        else {        # can get here due to error in initialization
            $id_scan_state = '';
            $i             = $i_save;
            last;
        }
    }

    if ( $id_scan_state eq ')' ) {
        warning("Hit end of line while seeking ) to end prototype\n");
    }

    # once we enter the actual identifier, it may not extend beyond
    # the end of the current line
    if ( $id_scan_state =~ /^[A\:\(\)]/ ) {
        $id_scan_state = '';
    }
    if ( $i < 0 ) { $i = 0 }

    unless ($type) {

        if ($saw_type) {

            if ($saw_alpha) {
                if ( $identifier =~ /^->/ && $last_nonblank_type eq 'w' ) {
                    $type = 'w';
                }
                else { $type = 'i' }
            }
            elsif ( $identifier eq '->' ) {
                $type = '->';
            }
            elsif (
                ( length($identifier) > 1 )

                # In something like '@$=' we have an identifier '@$'
                # In something like '$${' we have type '$$' (and only
                # part of an identifier)
                && !( $identifier =~ /\$$/ && $tok eq '{' )
                && ( $identifier !~ /^(sub |package )$/ )
              )
            {
                $type = 'i';
            }
            else { $type = 't' }
        }
        elsif ($saw_alpha) {

            # type 'w' includes anything without leading type info
            # ($,%,@,*) including something like abc::def::ghi
            $type = 'w';
        }
        else {
            $type = '';
        }    # this can happen on a restart
    }

    if ($identifier) {
        $tok = $identifier;
        if ($message) { write_logfile_entry($message) }
    }
    else {
        $tok = $tok_begin;
        $i   = $i_begin;
    }

    TOKENIZER_DEBUG_FLAG_SCAN_ID && do {
        my ( $a, $b, $c ) = caller;
        print STDOUT
"SCANID: called from $a $b $c with tok, i, state, identifier =$tok_begin, $i_begin, $id_scan_state_begin, $identifier_begin\n";
        print STDOUT
"SCANID: returned with tok, i, state, identifier =$tok, $i, $id_scan_state, $identifier\n";
    };
    return ( $i, $tok, $type, $id_scan_state, $identifier );
}

{

    # saved package and subnames in case prototype is on separate line
    my ( $package_saved, $subname_saved );

    sub do_scan_sub {

        # do_scan_sub parses a sub name and prototype
        # it is called with $i_beg equal to the index of the first nonblank
        # token following a 'sub' token.

        # TODO: add future error checks to be sure we have a valid
        # sub name.  For example, 'sub &doit' is wrong.  Also, be sure
        # a name is given if and only if a non-anonymous sub is
        # appropriate.
        # USES GLOBAL VARS: $current_package, $last_nonblank_token,
        # $in_attribute_list, %saw_function_definition,
        # $statement_type

        my (
            $input_line, $i,             $i_beg,
            $tok,        $type,          $rtokens,
            $rtoken_map, $id_scan_state, $max_token_index
        ) = @_;
        $id_scan_state = "";    # normally we get everything in one call
        my $subname = undef;
        my $package = undef;
        my $proto   = undef;
        my $attrs   = undef;
        my $match;

        my $pos_beg = $$rtoken_map[$i_beg];
        pos($input_line) = $pos_beg;

        # sub NAME PROTO ATTRS
        if (
            $input_line =~ m/\G\s*
        ((?:\w*(?:'|::))*)  # package - something that ends in :: or '
        (\w+)               # NAME    - required
        (\s*\([^){]*\))?    # PROTO   - something in parens
        (\s*:)?             # ATTRS   - leading : of attribute list
        /gcx
          )
        {
            $match   = 1;
            $subname = $2;
            $proto   = $3;
            $attrs   = $4;

            $package = ( defined($1) && $1 ) ? $1 : $current_package;
            $package =~ s/\'/::/g;
            if ( $package =~ /^\:/ ) { $package = 'main' . $package }
            $package =~ s/::$//;
            my $pos  = pos($input_line);
            my $numc = $pos - $pos_beg;
            $tok = 'sub ' . substr( $input_line, $pos_beg, $numc );
            $type = 'i';
        }

        # Look for prototype/attributes not preceded on this line by subname;
        # This might be an anonymous sub with attributes,
        # or a prototype on a separate line from its sub name
        elsif (
            $input_line =~ m/\G(\s*\([^){]*\))?  # PROTO
            (\s*:)?                              # ATTRS leading ':'
            /gcx
            && ( $1 || $2 )
          )
        {
            $match = 1;
            $proto = $1;
            $attrs = $2;

            # Handle prototype on separate line from subname
            if ($subname_saved) {
                $package = $package_saved;
                $subname = $subname_saved;
                $tok     = $last_nonblank_token;
            }
            $type = 'i';
        }

        if ($match) {

            # ATTRS: if there are attributes, back up and let the ':' be
            # found later by the scanner.
            my $pos = pos($input_line);
            if ($attrs) {
                $pos -= length($attrs);
            }

            my $next_nonblank_token = $tok;

            # catch case of line with leading ATTR ':' after anonymous sub
            if ( $pos == $pos_beg && $tok eq ':' ) {
                $type              = 'A';
                $in_attribute_list = 1;
            }

            # We must convert back from character position
            # to pre_token index.
            else {

                # I don't think an error flag can occur here ..but ?
                my $error;
                ( $i, $error ) = inverse_pretoken_map( $i, $pos, $rtoken_map,
                    $max_token_index );
                if ($error) { warning("Possibly invalid sub\n") }

                # check for multiple definitions of a sub
                ( $next_nonblank_token, my $i_next ) =
                  find_next_nonblank_token_on_this_line( $i, $rtokens,
                    $max_token_index );
            }

            if ( $next_nonblank_token =~ /^(\s*|#)$/ )
            {    # skip blank or side comment
                my ( $rpre_tokens, $rpre_types ) =
                  peek_ahead_for_n_nonblank_pre_tokens(1);
                if ( defined($rpre_tokens) && @$rpre_tokens ) {
                    $next_nonblank_token = $rpre_tokens->[0];
                }
                else {
                    $next_nonblank_token = '}';
                }
            }
            $package_saved = "";
            $subname_saved = "";
            if ( $next_nonblank_token eq '{' ) {
                if ($subname) {

                    # Check for multiple definitions of a sub, but
                    # it is ok to have multiple sub BEGIN, etc,
                    # so we do not complain if name is all caps
                    if (   $saw_function_definition{$package}{$subname}
                        && $subname !~ /^[A-Z]+$/ )
                    {
                        my $lno = $saw_function_definition{$package}{$subname};
                        warning(
"already saw definition of 'sub $subname' in package '$package' at line $lno\n"
                        );
                    }
                    $saw_function_definition{$package}{$subname} =
                      $tokenizer_self->{_last_line_number};
                }
            }
            elsif ( $next_nonblank_token eq ';' ) {
            }
            elsif ( $next_nonblank_token eq '}' ) {
            }

            # ATTRS - if an attribute list follows, remember the name
            # of the sub so the next opening brace can be labeled.
            # Setting 'statement_type' causes any ':'s to introduce
            # attributes.
            elsif ( $next_nonblank_token eq ':' ) {
                $statement_type = $tok;
            }

            # see if PROTO follows on another line:
            elsif ( $next_nonblank_token eq '(' ) {
                if ( $attrs || $proto ) {
                    warning(
"unexpected '(' after definition or declaration of sub '$subname'\n"
                    );
                }
                else {
                    $id_scan_state  = 'sub';    # we must come back to get proto
                    $statement_type = $tok;
                    $package_saved  = $package;
                    $subname_saved  = $subname;
                }
            }
            elsif ($next_nonblank_token) {      # EOF technically ok
                warning(
"expecting ':' or ';' or '{' after definition or declaration of sub '$subname' but saw '$next_nonblank_token'\n"
                );
            }
            check_prototype( $proto, $package, $subname );
        }

        # no match but line not blank
        else {
        }
        return ( $i, $tok, $type, $id_scan_state );
    }
}

#########i###############################################################
# Tokenizer utility routines which may use CONSTANTS but no other GLOBALS
#########################################################################

sub find_next_nonblank_token {
    my ( $i, $rtokens, $max_token_index ) = @_;

    if ( $i >= $max_token_index ) {
        if ( !peeked_ahead() ) {
            peeked_ahead(1);
            $rtokens =
              peek_ahead_for_nonblank_token( $rtokens, $max_token_index );
        }
    }
    my $next_nonblank_token = $$rtokens[ ++$i ];

    if ( $next_nonblank_token =~ /^\s*$/ ) {
        $next_nonblank_token = $$rtokens[ ++$i ];
    }
    return ( $next_nonblank_token, $i );
}

sub numerator_expected {

    # this is a filter for a possible numerator, in support of guessing
    # for the / pattern delimiter token.
    # returns -
    #   1 - yes
    #   0 - can't tell
    #  -1 - no
    # Note: I am using the convention that variables ending in
    # _expected have these 3 possible values.
    my ( $i, $rtokens, $max_token_index ) = @_;
    my $next_token = $$rtokens[ $i + 1 ];
    if ( $next_token eq '=' ) { $i++; }    # handle /=
    my ( $next_nonblank_token, $i_next ) =
      find_next_nonblank_token( $i, $rtokens, $max_token_index );

    if ( $next_nonblank_token =~ /(\(|\$|\w|\.|\@)/ ) {
        1;
    }
    else {

        if ( $next_nonblank_token =~ /^\s*$/ ) {
            0;
        }
        else {
            -1;
        }
    }
}

sub pattern_expected {

    # This is the start of a filter for a possible pattern.
    # It looks at the token after a possbible pattern and tries to
    # determine if that token could end a pattern.
    # returns -
    #   1 - yes
    #   0 - can't tell
    #  -1 - no
    my ( $i, $rtokens, $max_token_index ) = @_;
    my $next_token = $$rtokens[ $i + 1 ];
    if ( $next_token =~ /^[msixpodualgc]/ ) { $i++; }   # skip possible modifier
    my ( $next_nonblank_token, $i_next ) =
      find_next_nonblank_token( $i, $rtokens, $max_token_index );

    # list of tokens which may follow a pattern
    # (can probably be expanded)
    if ( $next_nonblank_token =~ /(\)|\}|\;|\&\&|\|\||and|or|while|if|unless)/ )
    {
        1;
    }
    else {

        if ( $next_nonblank_token =~ /^\s*$/ ) {
            0;
        }
        else {
            -1;
        }
    }
}

sub find_next_nonblank_token_on_this_line {
    my ( $i, $rtokens, $max_token_index ) = @_;
    my $next_nonblank_token;

    if ( $i < $max_token_index ) {
        $next_nonblank_token = $$rtokens[ ++$i ];

        if ( $next_nonblank_token =~ /^\s*$/ ) {

            if ( $i < $max_token_index ) {
                $next_nonblank_token = $$rtokens[ ++$i ];
            }
        }
    }
    else {
        $next_nonblank_token = "";
    }
    return ( $next_nonblank_token, $i );
}

sub find_angle_operator_termination {

    # We are looking at a '<' and want to know if it is an angle operator.
    # We are to return:
    #   $i = pretoken index of ending '>' if found, current $i otherwise
    #   $type = 'Q' if found, '>' otherwise
    my ( $input_line, $i_beg, $rtoken_map, $expecting, $max_token_index ) = @_;
    my $i    = $i_beg;
    my $type = '<';
    pos($input_line) = 1 + $$rtoken_map[$i];

    my $filter;

    # we just have to find the next '>' if a term is expected
    if ( $expecting == TERM ) { $filter = '[\>]' }

    # we have to guess if we don't know what is expected
    elsif ( $expecting == UNKNOWN ) { $filter = '[\>\;\=\#\|\<]' }

    # shouldn't happen - we shouldn't be here if operator is expected
    else { warning("Program Bug in find_angle_operator_termination\n") }

    # To illustrate what we might be looking at, in case we are
    # guessing, here are some examples of valid angle operators
    # (or file globs):
    #  <tmp_imp/*>
    #  <FH>
    #  <$fh>
    #  <*.c *.h>
    #  <_>
    #  <jskdfjskdfj* op/* jskdjfjkosvk*> ( glob.t)
    #  <${PREFIX}*img*.$IMAGE_TYPE>
    #  <img*.$IMAGE_TYPE>
    #  <Timg*.$IMAGE_TYPE>
    #  <$LATEX2HTMLVERSIONS${dd}html[1-9].[0-9].pl>
    #
    # Here are some examples of lines which do not have angle operators:
    #  return undef unless $self->[2]++ < $#{$self->[1]};
    #  < 2  || @$t >
    #
    # the following line from dlister.pl caused trouble:
    #  print'~'x79,"\n",$D<1024?"0.$D":$D>>10,"K, $C files\n\n\n";
    #
    # If the '<' starts an angle operator, it must end on this line and
    # it must not have certain characters like ';' and '=' in it.  I use
    # this to limit the testing.  This filter should be improved if
    # possible.

    if ( $input_line =~ /($filter)/g ) {

        if ( $1 eq '>' ) {

            # We MAY have found an angle operator termination if we get
            # here, but we need to do more to be sure we haven't been
            # fooled.
            my $pos = pos($input_line);

            my $pos_beg = $$rtoken_map[$i];
            my $str = substr( $input_line, $pos_beg, ( $pos - $pos_beg ) );

            # Reject if the closing '>' follows a '-' as in:
            # if ( VERSION < 5.009 && $op-> name eq 'aassign' ) { }
            if ( $expecting eq UNKNOWN ) {
                my $check = substr( $input_line, $pos - 2, 1 );
                if ( $check eq '-' ) {
                    return ( $i, $type );
                }
            }

            ######################################debug#####
            #write_diagnostics( "ANGLE? :$str\n");
            #print "ANGLE: found $1 at pos=$pos str=$str check=$check\n";
            ######################################debug#####
            $type = 'Q';
            my $error;
            ( $i, $error ) =
              inverse_pretoken_map( $i, $pos, $rtoken_map, $max_token_index );

            # It may be possible that a quote ends midway in a pretoken.
            # If this happens, it may be necessary to split the pretoken.
            if ($error) {
                warning(
                    "Possible tokinization error..please check this line\n");
                report_possible_bug();
            }

            # Now let's see where we stand....
            # OK if math op not possible
            if ( $expecting == TERM ) {
            }

            # OK if there are no more than 2 pre-tokens inside
            # (not possible to write 2 token math between < and >)
            # This catches most common cases
            elsif ( $i <= $i_beg + 3 ) {
                write_diagnostics("ANGLE(1 or 2 tokens): $str\n");
            }

            # Not sure..
            else {

                # Let's try a Brace Test: any braces inside must balance
                my $br = 0;
                while ( $str =~ /\{/g ) { $br++ }
                while ( $str =~ /\}/g ) { $br-- }
                my $sb = 0;
                while ( $str =~ /\[/g ) { $sb++ }
                while ( $str =~ /\]/g ) { $sb-- }
                my $pr = 0;
                while ( $str =~ /\(/g ) { $pr++ }
                while ( $str =~ /\)/g ) { $pr-- }

                # if braces do not balance - not angle operator
                if ( $br || $sb || $pr ) {
                    $i    = $i_beg;
                    $type = '<';
                    write_diagnostics(
                        "NOT ANGLE (BRACE={$br ($pr [$sb ):$str\n");
                }

                # we should keep doing more checks here...to be continued
                # Tentatively accepting this as a valid angle operator.
                # There are lots more things that can be checked.
                else {
                    write_diagnostics(
                        "ANGLE-Guessing yes: $str expecting=$expecting\n");
                    write_logfile_entry("Guessing angle operator here: $str\n");
                }
            }
        }

        # didn't find ending >
        else {
            if ( $expecting == TERM ) {
                warning("No ending > for angle operator\n");
            }
        }
    }
    return ( $i, $type );
}

sub scan_number_do {

    #  scan a number in any of the formats that Perl accepts
    #  Underbars (_) are allowed in decimal numbers.
    #  input parameters -
    #      $input_line  - the string to scan
    #      $i           - pre_token index to start scanning
    #    $rtoken_map    - reference to the pre_token map giving starting
    #                    character position in $input_line of token $i
    #  output parameters -
    #    $i            - last pre_token index of the number just scanned
    #    number        - the number (characters); or undef if not a number

    my ( $input_line, $i, $rtoken_map, $input_type, $max_token_index ) = @_;
    my $pos_beg = $$rtoken_map[$i];
    my $pos;
    my $i_begin = $i;
    my $number  = undef;
    my $type    = $input_type;

    my $first_char = substr( $input_line, $pos_beg, 1 );

    # Look for bad starting characters; Shouldn't happen..
    if ( $first_char !~ /[\d\.\+\-Ee]/ ) {
        warning("Program bug - scan_number given character $first_char\n");
        report_definite_bug();
        return ( $i, $type, $number );
    }

    # handle v-string without leading 'v' character ('Two Dot' rule)
    # (vstring.t)
    # TODO: v-strings may contain underscores
    pos($input_line) = $pos_beg;
    if ( $input_line =~ /\G((\d+)?\.\d+(\.\d+)+)/g ) {
        $pos = pos($input_line);
        my $numc = $pos - $pos_beg;
        $number = substr( $input_line, $pos_beg, $numc );
        $type = 'v';
        report_v_string($number);
    }

    # handle octal, hex, binary
    if ( !defined($number) ) {
        pos($input_line) = $pos_beg;
        if ( $input_line =~
            /\G[+-]?0(([xX][0-9a-fA-F_]+)|([0-7_]+)|([bB][01_]+))/g )
        {
            $pos = pos($input_line);
            my $numc = $pos - $pos_beg;
            $number = substr( $input_line, $pos_beg, $numc );
            $type = 'n';
        }
    }

    # handle decimal
    if ( !defined($number) ) {
        pos($input_line) = $pos_beg;

        if ( $input_line =~ /\G([+-]?[\d_]*(\.[\d_]*)?([Ee][+-]?(\d+))?)/g ) {
            $pos = pos($input_line);

            # watch out for things like 0..40 which would give 0. by this;
            if (   ( substr( $input_line, $pos - 1, 1 ) eq '.' )
                && ( substr( $input_line, $pos, 1 ) eq '.' ) )
            {
                $pos--;
            }
            my $numc = $pos - $pos_beg;
            $number = substr( $input_line, $pos_beg, $numc );
            $type = 'n';
        }
    }

    # filter out non-numbers like e + - . e2  .e3 +e6
    # the rule: at least one digit, and any 'e' must be preceded by a digit
    if (
        $number !~ /\d/    # no digits
        || (   $number =~ /^(.*)[eE]/
            && $1 !~ /\d/ )    # or no digits before the 'e'
      )
    {
        $number = undef;
        $type   = $input_type;
        return ( $i, $type, $number );
    }

    # Found a number; now we must convert back from character position
    # to pre_token index. An error here implies user syntax error.
    # An example would be an invalid octal number like '009'.
    my $error;
    ( $i, $error ) =
      inverse_pretoken_map( $i, $pos, $rtoken_map, $max_token_index );
    if ($error) { warning("Possibly invalid number\n") }

    return ( $i, $type, $number );
}

sub inverse_pretoken_map {

    # Starting with the current pre_token index $i, scan forward until
    # finding the index of the next pre_token whose position is $pos.
    my ( $i, $pos, $rtoken_map, $max_token_index ) = @_;
    my $error = 0;

    while ( ++$i <= $max_token_index ) {

        if ( $pos <= $$rtoken_map[$i] ) {

            # Let the calling routine handle errors in which we do not
            # land on a pre-token boundary.  It can happen by running
            # perltidy on some non-perl scripts, for example.
            if ( $pos < $$rtoken_map[$i] ) { $error = 1 }
            $i--;
            last;
        }
    }
    return ( $i, $error );
}

sub find_here_doc {

    # find the target of a here document, if any
    # input parameters:
    #   $i - token index of the second < of <<
    #   ($i must be less than the last token index if this is called)
    # output parameters:
    #   $found_target = 0 didn't find target; =1 found target
    #   HERE_TARGET - the target string (may be empty string)
    #   $i - unchanged if not here doc,
    #    or index of the last token of the here target
    #   $saw_error - flag noting unbalanced quote on here target
    my ( $expecting, $i, $rtokens, $rtoken_map, $max_token_index ) = @_;
    my $ibeg                 = $i;
    my $found_target         = 0;
    my $here_doc_target      = '';
    my $here_quote_character = '';
    my $saw_error            = 0;
    my ( $next_nonblank_token, $i_next_nonblank, $next_token );
    $next_token = $$rtokens[ $i + 1 ];

    # perl allows a backslash before the target string (heredoc.t)
    my $backslash = 0;
    if ( $next_token eq '\\' ) {
        $backslash  = 1;
        $next_token = $$rtokens[ $i + 2 ];
    }

    ( $next_nonblank_token, $i_next_nonblank ) =
      find_next_nonblank_token_on_this_line( $i, $rtokens, $max_token_index );

    if ( $next_nonblank_token =~ /[\'\"\`]/ ) {

        my $in_quote    = 1;
        my $quote_depth = 0;
        my $quote_pos   = 0;
        my $quoted_string;

        (
            $i, $in_quote, $here_quote_character, $quote_pos, $quote_depth,
            $quoted_string
          )
          = follow_quoted_string( $i_next_nonblank, $in_quote, $rtokens,
            $here_quote_character, $quote_pos, $quote_depth, $max_token_index );

        if ($in_quote) {    # didn't find end of quote, so no target found
            $i = $ibeg;
            if ( $expecting == TERM ) {
                warning(
"Did not find here-doc string terminator ($here_quote_character) before end of line \n"
                );
                $saw_error = 1;
            }
        }
        else {              # found ending quote
            my $j;
            $found_target = 1;

            my $tokj;
            for ( $j = $i_next_nonblank + 1 ; $j < $i ; $j++ ) {
                $tokj = $$rtokens[$j];

                # we have to remove any backslash before the quote character
                # so that the here-doc-target exactly matches this string
                next
                  if ( $tokj eq "\\"
                    && $j < $i - 1
                    && $$rtokens[ $j + 1 ] eq $here_quote_character );
                $here_doc_target .= $tokj;
            }
        }
    }

    elsif ( ( $next_token =~ /^\s*$/ ) and ( $expecting == TERM ) ) {
        $found_target = 1;
        write_logfile_entry(
            "found blank here-target after <<; suggest using \"\"\n");
        $i = $ibeg;
    }
    elsif ( $next_token =~ /^\w/ ) {    # simple bareword or integer after <<

        my $here_doc_expected;
        if ( $expecting == UNKNOWN ) {
            $here_doc_expected = guess_if_here_doc($next_token);
        }
        else {
            $here_doc_expected = 1;
        }

        if ($here_doc_expected) {
            $found_target    = 1;
            $here_doc_target = $next_token;
            $i               = $ibeg + 1;
        }

    }
    else {

        if ( $expecting == TERM ) {
            $found_target = 1;
            write_logfile_entry("Note: bare here-doc operator <<\n");
        }
        else {
            $i = $ibeg;
        }
    }

    # patch to neglect any prepended backslash
    if ( $found_target && $backslash ) { $i++ }

    return ( $found_target, $here_doc_target, $here_quote_character, $i,
        $saw_error );
}

sub do_quote {

    # follow (or continue following) quoted string(s)
    # $in_quote return code:
    #   0 - ok, found end
    #   1 - still must find end of quote whose target is $quote_character
    #   2 - still looking for end of first of two quotes
    #
    # Returns updated strings:
    #  $quoted_string_1 = quoted string seen while in_quote=1
    #  $quoted_string_2 = quoted string seen while in_quote=2
    my (
        $i,               $in_quote,    $quote_character,
        $quote_pos,       $quote_depth, $quoted_string_1,
        $quoted_string_2, $rtokens,     $rtoken_map,
        $max_token_index
    ) = @_;

    my $in_quote_starting = $in_quote;

    my $quoted_string;
    if ( $in_quote == 2 ) {    # two quotes/quoted_string_1s to follow
        my $ibeg = $i;
        (
            $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
            $quoted_string
          )
          = follow_quoted_string( $i, $in_quote, $rtokens, $quote_character,
            $quote_pos, $quote_depth, $max_token_index );
        $quoted_string_2 .= $quoted_string;
        if ( $in_quote == 1 ) {
            if ( $quote_character =~ /[\{\[\<\(]/ ) { $i++; }
            $quote_character = '';
        }
        else {
            $quoted_string_2 .= "\n";
        }
    }

    if ( $in_quote == 1 ) {    # one (more) quote to follow
        my $ibeg = $i;
        (
            $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
            $quoted_string
          )
          = follow_quoted_string( $ibeg, $in_quote, $rtokens, $quote_character,
            $quote_pos, $quote_depth, $max_token_index );
        $quoted_string_1 .= $quoted_string;
        if ( $in_quote == 1 ) {
            $quoted_string_1 .= "\n";
        }
    }
    return ( $i, $in_quote, $quote_character, $quote_pos, $quote_depth,
        $quoted_string_1, $quoted_string_2 );
}

sub follow_quoted_string {

    # scan for a specific token, skipping escaped characters
    # if the quote character is blank, use the first non-blank character
    # input parameters:
    #   $rtokens = reference to the array of tokens
    #   $i = the token index of the first character to search
    #   $in_quote = number of quoted strings being followed
    #   $beginning_tok = the starting quote character
    #   $quote_pos = index to check next for alphanumeric delimiter
    # output parameters:
    #   $i = the token index of the ending quote character
    #   $in_quote = decremented if found end, unchanged if not
    #   $beginning_tok = the starting quote character
    #   $quote_pos = index to check next for alphanumeric delimiter
    #   $quote_depth = nesting depth, since delimiters '{ ( [ <' can be nested.
    #   $quoted_string = the text of the quote (without quotation tokens)
    my ( $i_beg, $in_quote, $rtokens, $beginning_tok, $quote_pos, $quote_depth,
        $max_token_index )
      = @_;
    my ( $tok, $end_tok );
    my $i             = $i_beg - 1;
    my $quoted_string = "";

    TOKENIZER_DEBUG_FLAG_QUOTE && do {
        print STDOUT
"QUOTE entering with quote_pos = $quote_pos i=$i beginning_tok =$beginning_tok\n";
    };

    # get the corresponding end token
    if ( $beginning_tok !~ /^\s*$/ ) {
        $end_tok = matching_end_token($beginning_tok);
    }

    # a blank token means we must find and use the first non-blank one
    else {
        my $allow_quote_comments = ( $i < 0 ) ? 1 : 0; # i<0 means we saw a <cr>

        while ( $i < $max_token_index ) {
            $tok = $$rtokens[ ++$i ];

            if ( $tok !~ /^\s*$/ ) {

                if ( ( $tok eq '#' ) && ($allow_quote_comments) ) {
                    $i = $max_token_index;
                }
                else {

                    if ( length($tok) > 1 ) {
                        if ( $quote_pos <= 0 ) { $quote_pos = 1 }
                        $beginning_tok = substr( $tok, $quote_pos - 1, 1 );
                    }
                    else {
                        $beginning_tok = $tok;
                        $quote_pos     = 0;
                    }
                    $end_tok     = matching_end_token($beginning_tok);
                    $quote_depth = 1;
                    last;
                }
            }
            else {
                $allow_quote_comments = 1;
            }
        }
    }

    # There are two different loops which search for the ending quote
    # character.  In the rare case of an alphanumeric quote delimiter, we
    # have to look through alphanumeric tokens character-by-character, since
    # the pre-tokenization process combines multiple alphanumeric
    # characters, whereas for a non-alphanumeric delimiter, only tokens of
    # length 1 can match.

    ###################################################################
    # Case 1 (rare): loop for case of alphanumeric quote delimiter..
    # "quote_pos" is the position the current word to begin searching
    ###################################################################
    if ( $beginning_tok =~ /\w/ ) {

        # Note this because it is not recommended practice except
        # for obfuscated perl contests
        if ( $in_quote == 1 ) {
            write_logfile_entry(
                "Note: alphanumeric quote delimiter ($beginning_tok) \n");
        }

        while ( $i < $max_token_index ) {

            if ( $quote_pos == 0 || ( $i < 0 ) ) {
                $tok = $$rtokens[ ++$i ];

                if ( $tok eq '\\' ) {

                    # retain backslash unless it hides the end token
                    $quoted_string .= $tok
                      unless $$rtokens[ $i + 1 ] eq $end_tok;
                    $quote_pos++;
                    last if ( $i >= $max_token_index );
                    $tok = $$rtokens[ ++$i ];
                }
            }
            my $old_pos = $quote_pos;

            unless ( defined($tok) && defined($end_tok) && defined($quote_pos) )
            {

            }
            $quote_pos = 1 + index( $tok, $end_tok, $quote_pos );

            if ( $quote_pos > 0 ) {

                $quoted_string .=
                  substr( $tok, $old_pos, $quote_pos - $old_pos - 1 );

                $quote_depth--;

                if ( $quote_depth == 0 ) {
                    $in_quote--;
                    last;
                }
            }
            else {
                $quoted_string .= substr( $tok, $old_pos );
            }
        }
    }

    ########################################################################
    # Case 2 (normal): loop for case of a non-alphanumeric quote delimiter..
    ########################################################################
    else {

        while ( $i < $max_token_index ) {
            $tok = $$rtokens[ ++$i ];

            if ( $tok eq $end_tok ) {
                $quote_depth--;

                if ( $quote_depth == 0 ) {
                    $in_quote--;
                    last;
                }
            }
            elsif ( $tok eq $beginning_tok ) {
                $quote_depth++;
            }
            elsif ( $tok eq '\\' ) {

                # retain backslash unless it hides the beginning or end token
                $tok = $$rtokens[ ++$i ];
                $quoted_string .= '\\'
                  unless ( $tok eq $end_tok || $tok eq $beginning_tok );
            }
            $quoted_string .= $tok;
        }
    }
    if ( $i > $max_token_index ) { $i = $max_token_index }
    return ( $i, $in_quote, $beginning_tok, $quote_pos, $quote_depth,
        $quoted_string );
}

sub indicate_error {
    my ( $msg, $line_number, $input_line, $pos, $carrat ) = @_;
    interrupt_logfile();
    warning($msg);
    write_error_indicator_pair( $line_number, $input_line, $pos, $carrat );
    resume_logfile();
}

sub write_error_indicator_pair {
    my ( $line_number, $input_line, $pos, $carrat ) = @_;
    my ( $offset, $numbered_line, $underline ) =
      make_numbered_line( $line_number, $input_line, $pos );
    $underline = write_on_underline( $underline, $pos - $offset, $carrat );
    warning( $numbered_line . "\n" );
    $underline =~ s/\s*$//;
    warning( $underline . "\n" );
}

sub make_numbered_line {

    #  Given an input line, its line number, and a character position of
    #  interest, create a string not longer than 80 characters of the form
    #     $lineno: sub_string
    #  such that the sub_string of $str contains the position of interest
    #
    #  Here is an example of what we want, in this case we add trailing
    #  '...' because the line is long.
    #
    # 2: (One of QAML 2.0's authors is a member of the World Wide Web Con ...
    #
    #  Here is another example, this time in which we used leading '...'
    #  because of excessive length:
    #
    # 2: ... er of the World Wide Web Consortium's
    #
    #  input parameters are:
    #   $lineno = line number
    #   $str = the text of the line
    #   $pos = position of interest (the error) : 0 = first character
    #
    #   We return :
    #     - $offset = an offset which corrects the position in case we only
    #       display part of a line, such that $pos-$offset is the effective
    #       position from the start of the displayed line.
    #     - $numbered_line = the numbered line as above,
    #     - $underline = a blank 'underline' which is all spaces with the same
    #       number of characters as the numbered line.

    my ( $lineno, $str, $pos ) = @_;
    my $offset = ( $pos < 60 ) ? 0 : $pos - 40;
    my $excess = length($str) - $offset - 68;
    my $numc   = ( $excess > 0 ) ? 68 : undef;

    if ( defined($numc) ) {
        if ( $offset == 0 ) {
            $str = substr( $str, $offset, $numc - 4 ) . " ...";
        }
        else {
            $str = "... " . substr( $str, $offset + 4, $numc - 4 ) . " ...";
        }
    }
    else {

        if ( $offset == 0 ) {
        }
        else {
            $str = "... " . substr( $str, $offset + 4 );
        }
    }

    my $numbered_line = sprintf( "%d: ", $lineno );
    $offset -= length($numbered_line);
    $numbered_line .= $str;
    my $underline = " " x length($numbered_line);
    return ( $offset, $numbered_line, $underline );
}

sub write_on_underline {

    # The "underline" is a string that shows where an error is; it starts
    # out as a string of blanks with the same length as the numbered line of
    # code above it, and we have to add marking to show where an error is.
    # In the example below, we want to write the string '--^' just below
    # the line of bad code:
    #
    # 2: (One of QAML 2.0's authors is a member of the World Wide Web Con ...
    #                 ---^
    # We are given the current underline string, plus a position and a
    # string to write on it.
    #
    # In the above example, there will be 2 calls to do this:
    # First call:  $pos=19, pos_chr=^
    # Second call: $pos=16, pos_chr=---
    #
    # This is a trivial thing to do with substr, but there is some
    # checking to do.

    my ( $underline, $pos, $pos_chr ) = @_;

    # check for error..shouldn't happen
    unless ( ( $pos >= 0 ) && ( $pos <= length($underline) ) ) {
        return $underline;
    }
    my $excess = length($pos_chr) + $pos - length($underline);
    if ( $excess > 0 ) {
        $pos_chr = substr( $pos_chr, 0, length($pos_chr) - $excess );
    }
    substr( $underline, $pos, length($pos_chr) ) = $pos_chr;
    return ($underline);
}

sub pre_tokenize {

    # Break a string, $str, into a sequence of preliminary tokens.  We
    # are interested in these types of tokens:
    #   words       (type='w'),            example: 'max_tokens_wanted'
    #   digits      (type = 'd'),          example: '0755'
    #   whitespace  (type = 'b'),          example: '   '
    #   any other single character (i.e. punct; type = the character itself).
    # We cannot do better than this yet because we might be in a quoted
    # string or pattern.  Caller sets $max_tokens_wanted to 0 to get all
    # tokens.
    my ( $str, $max_tokens_wanted ) = @_;

    # we return references to these 3 arrays:
    my @tokens    = ();     # array of the tokens themselves
    my @token_map = (0);    # string position of start of each token
    my @type      = ();     # 'b'=whitespace, 'd'=digits, 'w'=alpha, or punct

    do {

        # whitespace
        if ( $str =~ /\G(\s+)/gc ) { push @type, 'b'; }

        # numbers
        # note that this must come before words!
        elsif ( $str =~ /\G(\d+)/gc ) { push @type, 'd'; }

        # words
        elsif ( $str =~ /\G(\w+)/gc ) { push @type, 'w'; }

        # single-character punctuation
        elsif ( $str =~ /\G(\W)/gc ) { push @type, $1; }

        # that's all..
        else {
            return ( \@tokens, \@token_map, \@type );
        }

        push @tokens,    $1;
        push @token_map, pos($str);

    } while ( --$max_tokens_wanted != 0 );

    return ( \@tokens, \@token_map, \@type );
}

sub show_tokens {

    # this is an old debug routine
    my ( $rtokens, $rtoken_map ) = @_;
    my $num = scalar(@$rtokens);
    my $i;

    for ( $i = 0 ; $i < $num ; $i++ ) {
        my $len = length( $$rtokens[$i] );
        print STDOUT "$i:$len:$$rtoken_map[$i]:$$rtokens[$i]:\n";
    }
}

sub matching_end_token {

    # find closing character for a pattern
    my $beginning_token = shift;

    if ( $beginning_token eq '{' ) {
        '}';
    }
    elsif ( $beginning_token eq '[' ) {
        ']';
    }
    elsif ( $beginning_token eq '<' ) {
        '>';
    }
    elsif ( $beginning_token eq '(' ) {
        ')';
    }
    else {
        $beginning_token;
    }
}

sub dump_token_types {
    my $class = shift;
    my $fh    = shift;

    # This should be the latest list of token types in use
    # adding NEW_TOKENS: add a comment here
    print $fh <<'END_OF_LIST';

Here is a list of the token types currently used for lines of type 'CODE'.
For the following tokens, the "type" of a token is just the token itself.

.. :: << >> ** && .. || // -> => += -= .= %= &= |= ^= *= <>
( ) <= >= == =~ !~ != ++ -- /= x=
... **= <<= >>= &&= ||= //= <=>
, + - / * | % ! x ~ = \ ? : . < > ^ &

The following additional token types are defined:

 type    meaning
    b    blank (white space)
    {    indent: opening structural curly brace or square bracket or paren
         (code block, anonymous hash reference, or anonymous array reference)
    }    outdent: right structural curly brace or square bracket or paren
    [    left non-structural square bracket (enclosing an array index)
    ]    right non-structural square bracket
    (    left non-structural paren (all but a list right of an =)
    )    right non-structural paren
    L    left non-structural curly brace (enclosing a key)
    R    right non-structural curly brace
    ;    terminal semicolon
    f    indicates a semicolon in a "for" statement
    h    here_doc operator <<
    #    a comment
    Q    indicates a quote or pattern
    q    indicates a qw quote block
    k    a perl keyword
    C    user-defined constant or constant function (with void prototype = ())
    U    user-defined function taking parameters
    G    user-defined function taking block parameter (like grep/map/eval)
    M    (unused, but reserved for subroutine definition name)
    P    (unused, but -html uses it to label pod text)
    t    type indicater such as %,$,@,*,&,sub
    w    bare word (perhaps a subroutine call)
    i    identifier of some type (with leading %, $, @, *, &, sub, -> )
    n    a number
    v    a v-string
    F    a file test operator (like -e)
    Y    File handle
    Z    identifier in indirect object slot: may be file handle, object
    J    LABEL:  code block label
    j    LABEL after next, last, redo, goto
    p    unary +
    m    unary -
    pp   pre-increment operator ++
    mm   pre-decrement operator --
    A    : used as attribute separator

    Here are the '_line_type' codes used internally:
    SYSTEM         - system-specific code before hash-bang line
    CODE           - line of perl code (including comments)
    POD_START      - line starting pod, such as '=head'
    POD            - pod documentation text
    POD_END        - last line of pod section, '=cut'
    HERE           - text of here-document
    HERE_END       - last line of here-doc (target word)
    FORMAT         - format section
    FORMAT_END     - last line of format section, '.'
    DATA_START     - __DATA__ line
    DATA           - unidentified text following __DATA__
    END_START      - __END__ line
    END            - unidentified text following __END__
    ERROR          - we are in big trouble, probably not a perl script
END_OF_LIST
}

BEGIN {

    # These names are used in error messages
    @opening_brace_names = qw# '{' '[' '(' '?' #;
    @closing_brace_names = qw# '}' ']' ')' ':' #;

    my @digraphs = qw(
      .. :: << >> ** && .. || // -> => += -= .= %= &= |= ^= *= <>
      <= >= == =~ !~ != ++ -- /= x= ~~
    );
    @is_digraph{@digraphs} = (1) x scalar(@digraphs);

    my @trigraphs = qw( ... **= <<= >>= &&= ||= //= <=> !~~ );
    @is_trigraph{@trigraphs} = (1) x scalar(@trigraphs);

    # make a hash of all valid token types for self-checking the tokenizer
    # (adding NEW_TOKENS : select a new character and add to this list)
    my @valid_token_types = qw#
      A b C G L R f h Q k t w i q n p m F pp mm U j J Y Z v
      { } ( ) [ ] ; + - / * | % ! x ~ = \ ? : . < > ^ &
      #;
    push( @valid_token_types, @digraphs );
    push( @valid_token_types, @trigraphs );
    push( @valid_token_types, ( '#', ',', 'CORE::' ) );
    @is_valid_token_type{@valid_token_types} = (1) x scalar(@valid_token_types);

    # a list of file test letters, as in -e (Table 3-4 of 'camel 3')
    my @file_test_operators =
      qw( A B C M O R S T W X b c d e f g k l o p r s t u w x z);
    @is_file_test_operator{@file_test_operators} =
      (1) x scalar(@file_test_operators);

    # these functions have prototypes of the form (&), so when they are
    # followed by a block, that block MAY BE followed by an operator.
    # Smartmatch operator ~~ may be followed by anonomous hash or array ref
    @_ = qw( do eval );
    @is_block_operator{@_} = (1) x scalar(@_);

    # these functions allow an identifier in the indirect object slot
    @_ = qw( print printf sort exec system say);
    @is_indirect_object_taker{@_} = (1) x scalar(@_);

    # These tokens may precede a code block
    # patched for SWITCH/CASE
    @_ =
      qw( BEGIN END CHECK INIT AUTOLOAD DESTROY UNITCHECK continue if elsif else
      unless do while until eval for foreach map grep sort
      switch case given when);
    @is_code_block_token{@_} = (1) x scalar(@_);

    # I'll build the list of keywords incrementally
    my @Keywords = ();

    # keywords and tokens after which a value or pattern is expected,
    # but not an operator.  In other words, these should consume terms
    # to their right, or at least they are not expected to be followed
    # immediately by operators.
    my @value_requestor = qw(
      AUTOLOAD
      BEGIN
      CHECK
      DESTROY
      END
      EQ
      GE
      GT
      INIT
      LE
      LT
      NE
      UNITCHECK
      abs
      accept
      alarm
      and
      atan2
      bind
      binmode
      bless
      break
      caller
      chdir
      chmod
      chomp
      chop
      chown
      chr
      chroot
      close
      closedir
      cmp
      connect
      continue
      cos
      crypt
      dbmclose
      dbmopen
      defined
      delete
      die
      dump
      each
      else
      elsif
      eof
      eq
      exec
      exists
      exit
      exp
      fcntl
      fileno
      flock
      for
      foreach
      formline
      ge
      getc
      getgrgid
      getgrnam
      gethostbyaddr
      gethostbyname
      getnetbyaddr
      getnetbyname
      getpeername
      getpgrp
      getpriority
      getprotobyname
      getprotobynumber
      getpwnam
      getpwuid
      getservbyname
      getservbyport
      getsockname
      getsockopt
      glob
      gmtime
      goto
      grep
      gt
      hex
      if
      index
      int
      ioctl
      join
      keys
      kill
      last
      lc
      lcfirst
      le
      length
      link
      listen
      local
      localtime
      lock
      log
      lstat
      lt
      map
      mkdir
      msgctl
      msgget
      msgrcv
      msgsnd
      my
      ne
      next
      no
      not
      oct
      open
      opendir
      or
      ord
      our
      pack
      pipe
      pop
      pos
      print
      printf
      prototype
      push
      quotemeta
      rand
      read
      readdir
      readlink
      readline
      readpipe
      recv
      redo
      ref
      rename
      require
      reset
      return
      reverse
      rewinddir
      rindex
      rmdir
      scalar
      seek
      seekdir
      select
      semctl
      semget
      semop
      send
      sethostent
      setnetent
      setpgrp
      setpriority
      setprotoent
      setservent
      setsockopt
      shift
      shmctl
      shmget
      shmread
      shmwrite
      shutdown
      sin
      sleep
      socket
      socketpair
      sort
      splice
      split
      sprintf
      sqrt
      srand
      stat
      study
      substr
      symlink
      syscall
      sysopen
      sysread
      sysseek
      system
      syswrite
      tell
      telldir
      tie
      tied
      truncate
      uc
      ucfirst
      umask
      undef
      unless
      unlink
      unpack
      unshift
      untie
      until
      use
      utime
      values
      vec
      waitpid
      warn
      while
      write
      xor

      switch
      case
      given
      when
      err
      say
    );

    # patched above for SWITCH/CASE given/when err say
    # 'err' is a fairly safe addition.
    # TODO: 'default' still needed if appropriate
    # 'use feature' seen, but perltidy works ok without it.
    # Concerned that 'default' could break code.
    push( @Keywords, @value_requestor );

    # These are treated the same but are not keywords:
    my @extra_vr = qw(
      constant
      vars
    );
    push( @value_requestor, @extra_vr );

    @expecting_term_token{@value_requestor} = (1) x scalar(@value_requestor);

    # this list contains keywords which do not look for arguments,
    # so that they might be followed by an operator, or at least
    # not a term.
    my @operator_requestor = qw(
      endgrent
      endhostent
      endnetent
      endprotoent
      endpwent
      endservent
      fork
      getgrent
      gethostent
      getlogin
      getnetent
      getppid
      getprotoent
      getpwent
      getservent
      setgrent
      setpwent
      time
      times
      wait
      wantarray
    );

    push( @Keywords, @operator_requestor );

    # These are treated the same but are not considered keywords:
    my @extra_or = qw(
      STDERR
      STDIN
      STDOUT
    );

    push( @operator_requestor, @extra_or );

    @expecting_operator_token{@operator_requestor} =
      (1) x scalar(@operator_requestor);

    # these token TYPES expect trailing operator but not a term
    # note: ++ and -- are post-increment and decrement, 'C' = constant
    my @operator_requestor_types = qw( ++ -- C <> q );
    @expecting_operator_types{@operator_requestor_types} =
      (1) x scalar(@operator_requestor_types);

    # these token TYPES consume values (terms)
    # note: pp and mm are pre-increment and decrement
    # f=semicolon in for,  F=file test operator
    my @value_requestor_type = qw#
      L { ( [ ~ !~ =~ ; . .. ... A : && ! || // = + - x
      **= += -= .= /= *= %= x= &= |= ^= <<= >>= &&= ||= //=
      <= >= == != => \ > < % * / ? & | ** <=> ~~ !~~
      f F pp mm Y p m U J G j >> << ^ t
      #;
    push( @value_requestor_type, ',' )
      ;    # (perl doesn't like a ',' in a qw block)
    @expecting_term_types{@value_requestor_type} =
      (1) x scalar(@value_requestor_type);

    # Note: the following valid token types are not assigned here to
    # hashes requesting to be followed by values or terms, but are
    # instead currently hard-coded into sub operator_expected:
    # ) -> :: Q R Z ] b h i k n v w } #

    # For simple syntax checking, it is nice to have a list of operators which
    # will really be unhappy if not followed by a term.  This includes most
    # of the above...
    %really_want_term = %expecting_term_types;

    # with these exceptions...
    delete $really_want_term{'U'}; # user sub, depends on prototype
    delete $really_want_term{'F'}; # file test works on $_ if no following term
    delete $really_want_term{'Y'}; # indirect object, too risky to check syntax;
                                   # let perl do it

    @_ = qw(q qq qw qx qr s y tr m);
    @is_q_qq_qw_qx_qr_s_y_tr_m{@_} = (1) x scalar(@_);

    # These keywords are handled specially in the tokenizer code:
    my @special_keywords = qw(
      do
      eval
      format
      m
      package
      q
      qq
      qr
      qw
      qx
      s
      sub
      tr
      y
    );
    push( @Keywords, @special_keywords );

    # Keywords after which list formatting may be used
    # WARNING: do not include |map|grep|eval or perl may die on
    # syntax errors (map1.t).
    my @keyword_taking_list = qw(
      and
      chmod
      chomp
      chop
      chown
      dbmopen
      die
      elsif
      exec
      fcntl
      for
      foreach
      formline
      getsockopt
      if
      index
      ioctl
      join
      kill
      local
      msgctl
      msgrcv
      msgsnd
      my
      open
      or
      our
      pack
      print
      printf
      push
      read
      readpipe
      recv
      return
      reverse
      rindex
      seek
      select
      semctl
      semget
      send
      setpriority
      setsockopt
      shmctl
      shmget
      shmread
      shmwrite
      socket
      socketpair
      sort
      splice
      split
      sprintf
      substr
      syscall
      sysopen
      sysread
      sysseek
      system
      syswrite
      tie
      unless
      unlink
      unpack
      unshift
      until
      vec
      warn
      while
      given
      when
    );
    @is_keyword_taking_list{@keyword_taking_list} =
      (1) x scalar(@keyword_taking_list);

    # These are not used in any way yet
    #    my @unused_keywords = qw(
    #     __FILE__
    #     __LINE__
    #     __PACKAGE__
    #     );

    #  The list of keywords was originally extracted from function 'keyword' in
    #  perl file toke.c version 5.005.03, using this utility, plus a
    #  little editing: (file getkwd.pl):
    #  while (<>) { while (/\"(.*)\"/g) { print "$1\n"; } }
    #  Add 'get' prefix where necessary, then split into the above lists.
    #  This list should be updated as necessary.
    #  The list should not contain these special variables:
    #  ARGV DATA ENV SIG STDERR STDIN STDOUT
    #  __DATA__ __END__

    @is_keyword{@Keywords} = (1) x scalar(@Keywords);
}
1;
__END__

=head1 NAME

Perl::Tidy - Parses and beautifies perl source

=head1 SYNOPSIS

    use Perl::Tidy;

    my $error_flag = Perl::Tidy::perltidy(
        source            => $source,
        destination       => $destination,
        stderr            => $stderr,
        argv              => $argv,
        perltidyrc        => $perltidyrc,
        logfile           => $logfile,
        errorfile         => $errorfile,
        formatter         => $formatter,           # callback object (see below)
        dump_options      => $dump_options,
        dump_options_type => $dump_options_type,
        prefilter         => $prefilter_coderef,
        postfilter        => $postfilter_coderef,
    );

=head1 DESCRIPTION

This module makes the functionality of the perltidy utility available to perl
scripts.  Any or all of the input parameters may be omitted, in which case the
@ARGV array will be used to provide input parameters as described
in the perltidy(1) man page.

For example, the perltidy script is basically just this:

    use Perl::Tidy;
    Perl::Tidy::perltidy();

The call to B<perltidy> returns a scalar B<$error_flag> which is TRUE if an
error caused premature termination, and FALSE if the process ran to normal
completion.  Additional discuss of errors is contained below in the L<ERROR
HANDLING> section.

The module accepts input and output streams by a variety of methods.
The following list of parameters may be any of the following: a
filename, an ARRAY reference, a SCALAR reference, or an object with
either a B<getline> or B<print> method, as appropriate.

        source            - the source of the script to be formatted
        destination       - the destination of the formatted output
        stderr            - standard error output
        perltidyrc        - the .perltidyrc file
        logfile           - the .LOG file stream, if any
        errorfile         - the .ERR file stream, if any
        dump_options      - ref to a hash to receive parameters (see below),
        dump_options_type - controls contents of dump_options
        dump_getopt_flags - ref to a hash to receive Getopt flags
        dump_options_category - ref to a hash giving category of options
        dump_abbreviations    - ref to a hash giving all abbreviations

The following chart illustrates the logic used to decide how to
treat a parameter.

   ref($param)  $param is assumed to be:
   -----------  ---------------------
   undef        a filename
   SCALAR       ref to string
   ARRAY        ref to array
   (other)      object with getline (if source) or print method

If the parameter is an object, and the object has a B<close> method, that
close method will be called at the end of the stream.

=over 4

=item source

If the B<source> parameter is given, it defines the source of the input stream.
If an input stream is defined with the B<source> parameter then no other source
filenames may be specified in the @ARGV array or B<argv> parameter.

=item destination

If the B<destination> parameter is given, it will be used to define the
file or memory location to receive output of perltidy.

=item stderr

The B<stderr> parameter allows the calling program to redirect the stream that
would otherwise go to the standard error output device to any of the stream
types listed above.  This stream contains important warnings and errors
related to the parameters passed to perltidy.

=item perltidyrc

If the B<perltidyrc> file is given, it will be used instead of any
F<.perltidyrc> configuration file that would otherwise be used.

=item errorfile

The B<errorfile> parameter allows the calling program to capture
the stream that would otherwise go to either a .ERR file.  This
stream contains warnings or errors related to the contents of one
source file or stream.

The reason that this is different from the stderr stream is that when perltidy
is called to process multiple files there will be up to one .ERR file created
for each file and it would be very confusing if they were combined.

However if perltidy is called to process just a single perl script then it may
be more conveninent to combine the B<errorfile> stream with the B<stderr>
stream.  This can be done by setting the B<-se> parameter, in which case this
parameter is ignored.

=item logfile

The B<logfile> parameter allows the calling program to capture
the stream that would otherwise go to a .LOG file.  This
stream is only created if requested with a B<-g> parameter.  It
contains detailed diagnostic information about a script
which may be useful for debugging.

=item argv

If the B<argv> parameter is given, it will be used instead of the
B<@ARGV> array.  The B<argv> parameter may be a string, a reference to a
string, or a reference to an array.  If it is a string or reference to a
string, it will be parsed into an array of items just as if it were a
command line string.

=item dump_options

If the B<dump_options> parameter is given, it must be the reference to a hash.
In this case, the parameters contained in any perltidyrc configuration file
will be placed in this hash and perltidy will return immediately.  This is
equivalent to running perltidy with --dump-options, except that the perameters
are returned in a hash rather than dumped to standard output.  Also, by default
only the parameters in the perltidyrc file are returned, but this can be
changed (see the next parameter).  This parameter provides a convenient method
for external programs to read a perltidyrc file.  An example program using
this feature, F<perltidyrc_dump.pl>, is included in the distribution.

Any combination of the B<dump_> parameters may be used together.

=item dump_options_type

This parameter is a string which can be used to control the parameters placed
in the hash reference supplied by B<dump_options>.  The possible values are
'perltidyrc' (default) and 'full'.  The 'full' parameter causes both the
default options plus any options found in a perltidyrc file to be returned.

=item dump_getopt_flags

If the B<dump_getopt_flags> parameter is given, it must be the reference to a
hash.  This hash will receive all of the parameters that perltidy understands
and flags that are passed to Getopt::Long.  This parameter may be
used alone or with the B<dump_options> flag.  Perltidy will
exit immediately after filling this hash.  See the demo program
F<perltidyrc_dump.pl> for example usage.

=item dump_options_category

If the B<dump_options_category> parameter is given, it must be the reference to a
hash.  This hash will receive a hash with keys equal to all long parameter names
and values equal to the title of the corresponding section of the perltidy manual.
See the demo program F<perltidyrc_dump.pl> for example usage.

=item dump_abbreviations

If the B<dump_abbreviations> parameter is given, it must be the reference to a
hash.  This hash will receive all abbreviations used by Perl::Tidy.  See the
demo program F<perltidyrc_dump.pl> for example usage.

=item prefilter

A code reference that will be applied to the source before tidying. It is
expected to take the full content as a string in its input, and output the
transformed content.

=item postfilter

A code reference that will be applied to the tidied result before outputting.
It is expected to take the full content as a string in its input, and output
the transformed content.

Note: A convenient way to check the function of your custom prefilter and
postfilter code is to use the --notidy option, first with just the prefilter
and then with both the prefilter and postfilter.  See also the file
B<filter_example.pl> in the perltidy distribution.

=back

=head1 ERROR HANDLING

Perltidy will return with an error flag indicating if the process had to be
terminated early due to errors in the input parameters.  This can happen for
example if a parameter is misspelled or given an invalid value.  The calling
program should check this flag because if it is set the destination stream will
be empty or incomplete and should be ignored.  Error messages in the B<stderr>
stream will indicate the cause of any problem.

If the error flag is not set then perltidy ran to completion.   However there
may still be warning messages in the B<stderr> stream related to control
parameters, and there may be warning messages in the B<errorfile> stream
relating to possible syntax errors in the source code being tidied.

In the event of a catastrophic error for which recovery is not possible
B<perltidy> terminates by making calls to B<croak> or B<confess> to help the
programmer localize the problem.  These should normally only occur during
program development.

=head1 NOTES ON FORMATTING PARAMETERS

Parameters which control formatting may be passed in several ways: in a
F<.perltidyrc> configuration file, in the B<perltidyrc> parameter, and in the
B<argv> parameter.

The B<-syn> (B<--check-syntax>) flag may be used with all source and
destination streams except for standard input and output.  However
data streams which are not associated with a filename will
be copied to a temporary file before being be passed to Perl.  This
use of temporary files can cause somewhat confusing output from Perl.

If the B<-pbp> style is used it will typically be necessary to also
specify a B<-nst> flag.  This is necessary to turn off the B<-st> flag
contained in the B<-pbp> parameter set which otherwise would direct
the output stream to the standard output.

=head1 EXAMPLES

The following example uses string references to hold the input and output
code and error streams, and illustrates checking for errors.

  use Perl::Tidy;

  my $source_string = <<'EOT';
  my$error=Perl::Tidy::perltidy(argv=>$argv,source=>\$source_string,
    destination=>\$dest_string,stderr=>\$stderr_string,
  errorfile=>\$errorfile_string,);
  EOT

  my $dest_string;
  my $stderr_string;
  my $errorfile_string;
  my $argv = "-npro";   # Ignore any .perltidyrc at this site
  $argv .= " -pbp";     # Format according to perl best practices
  $argv .= " -nst";     # Must turn off -st in case -pbp is specified
  $argv .= " -se";      # -se appends the errorfile to stderr
  ## $argv .= " --spell-check";  # uncomment to trigger an error

  print "<<RAW SOURCE>>\n$source_string\n";

  my $error = Perl::Tidy::perltidy(
      argv        => $argv,
      source      => \$source_string,
      destination => \$dest_string,
      stderr      => \$stderr_string,
      errorfile   => \$errorfile_string,    # ignored when -se flag is set
      ##phasers   => 'stun',                # uncomment to trigger an error
  );

  if ($error) {

      # serious error in input parameters, no tidied output
      print "<<STDERR>>\n$stderr_string\n";
      die "Exiting because of serious errors\n";
  }

  if ($dest_string)      { print "<<TIDIED SOURCE>>\n$dest_string\n" }
  if ($stderr_string)    { print "<<STDERR>>\n$stderr_string\n" }
  if ($errorfile_string) { print "<<.ERR file>>\n$errorfile_string\n" }

Additional examples are given in examples section of the perltidy distribution.

=head1 Using the B<formatter> Callback Object

The B<formatter> parameter is an optional callback object which allows
the calling program to receive tokenized lines directly from perltidy for
further specialized processing.  When this parameter is used, the two
formatting options which are built into perltidy (beautification or
html) are ignored.  The following diagram illustrates the logical flow:

                    |-- (normal route)   -> code beautification
  caller->perltidy->|-- (-html flag )    -> create html
                    |-- (formatter given)-> callback to write_line

This can be useful for processing perl scripts in some way.  The
parameter C<$formatter> in the perltidy call,

        formatter   => $formatter,

is an object created by the caller with a C<write_line> method which
will accept and process tokenized lines, one line per call.  Here is
a simple example of a C<write_line> which merely prints the line number,
the line type (as determined by perltidy), and the text of the line:

 sub write_line {

     # This is called from perltidy line-by-line
     my $self              = shift;
     my $line_of_tokens    = shift;
     my $line_type         = $line_of_tokens->{_line_type};
     my $input_line_number = $line_of_tokens->{_line_number};
     my $input_line        = $line_of_tokens->{_line_text};
     print "$input_line_number:$line_type:$input_line";
 }

The complete program, B<perllinetype>, is contained in the examples section of
the source distribution.  As this example shows, the callback method
receives a parameter B<$line_of_tokens>, which is a reference to a hash
of other useful information.  This example uses these hash entries:

 $line_of_tokens->{_line_number} - the line number (1,2,...)
 $line_of_tokens->{_line_text}   - the text of the line
 $line_of_tokens->{_line_type}   - the type of the line, one of:

    SYSTEM         - system-specific code before hash-bang line
    CODE           - line of perl code (including comments)
    POD_START      - line starting pod, such as '=head'
    POD            - pod documentation text
    POD_END        - last line of pod section, '=cut'
    HERE           - text of here-document
    HERE_END       - last line of here-doc (target word)
    FORMAT         - format section
    FORMAT_END     - last line of format section, '.'
    DATA_START     - __DATA__ line
    DATA           - unidentified text following __DATA__
    END_START      - __END__ line
    END            - unidentified text following __END__
    ERROR          - we are in big trouble, probably not a perl script

Most applications will be only interested in lines of type B<CODE>.  For
another example, let's write a program which checks for one of the
so-called I<naughty matching variables> C<&`>, C<$&>, and C<$'>, which
can slow down processing.  Here is a B<write_line>, from the example
program B<find_naughty.pl>, which does that:

 sub write_line {

     # This is called back from perltidy line-by-line
     # We're looking for $`, $&, and $'
     my ( $self, $line_of_tokens ) = @_;

     # pull out some stuff we might need
     my $line_type         = $line_of_tokens->{_line_type};
     my $input_line_number = $line_of_tokens->{_line_number};
     my $input_line        = $line_of_tokens->{_line_text};
     my $rtoken_type       = $line_of_tokens->{_rtoken_type};
     my $rtokens           = $line_of_tokens->{_rtokens};
     chomp $input_line;

     # skip comments, pod, etc
     return if ( $line_type ne 'CODE' );

     # loop over tokens looking for $`, $&, and $'
     for ( my $j = 0 ; $j < @$rtoken_type ; $j++ ) {

         # we only want to examine token types 'i' (identifier)
         next unless $$rtoken_type[$j] eq 'i';

         # pull out the actual token text
         my $token = $$rtokens[$j];

         # and check it
         if ( $token =~ /^\$[\`\&\']$/ ) {
             print STDERR
               "$input_line_number: $token\n";
         }
     }
 }

This example pulls out these tokenization variables from the $line_of_tokens
hash reference:

     $rtoken_type = $line_of_tokens->{_rtoken_type};
     $rtokens     = $line_of_tokens->{_rtokens};

The variable C<$rtoken_type> is a reference to an array of token type codes,
and C<$rtokens> is a reference to a corresponding array of token text.
These are obviously only defined for lines of type B<CODE>.
Perltidy classifies tokens into types, and has a brief code for each type.
You can get a complete list at any time by running perltidy from the
command line with

     perltidy --dump-token-types

In the present example, we are only looking for tokens of type B<i>
(identifiers), so the for loop skips past all other types.  When an
identifier is found, its actual text is checked to see if it is one
being sought.  If so, the above write_line prints the token and its
line number.

The B<formatter> feature is relatively new in perltidy, and further
documentation needs to be written to complete its description.  However,
several example programs have been written and can be found in the
B<examples> section of the source distribution.  Probably the best way
to get started is to find one of the examples which most closely matches
your application and start modifying it.

For help with perltidy's pecular way of breaking lines into tokens, you
might run, from the command line,

 perltidy -D filename

where F<filename> is a short script of interest.  This will produce
F<filename.DEBUG> with interleaved lines of text and their token types.
The B<-D> flag has been in perltidy from the beginning for this purpose.
If you want to see the code which creates this file, it is
C<write_debug_entry> in Tidy.pm.

=head1 EXPORT

  &perltidy

=head1 CREDITS

Thanks to Hugh Myers who developed the initial modular interface
to perltidy.

=head1 VERSION

This man page documents Perl::Tidy version 20121207.

=head1 LICENSE

This package is free software; you can redistribute it and/or modify it
under the terms of the "GNU General Public License".

Please refer to the file "COPYING" for details.

=head1 AUTHOR

 Steve Hancock
 perltidy at users.sourceforge.net

=head1 SEE ALSO

The perltidy(1) man page describes all of the features of perltidy.  It
can be found at http://perltidy.sourceforge.net.

=cut