Copy stable/9 to releng/9.3 as part of the 9.3-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

MFC: r229809: Don't include <sys/timeb.h>.
The purpose of not including <sys/timeb.h> is to avoid the compiler
warning we put in <sys/timeb.h>. So this change will cause CVS to
provide its own definition of 'struct timeb'. This is only OK
because the code never calls the libc ftime(3) interface. CVS only
uses 'struct timeb' as a convenience structure -- thus it doesn't
matter how its supplied definition does or does not match the
ftime(3) ABI.

Copy head to stable/9 as part of 9.0-RELEASE release cycle.

Approved by: re (implicit)

Fix buildworld -DNO_CLEAN when using with Perforce, which marks files as
read-only by default, meaning files copied can't be overwritten next time.

Reviewed by: imp
Approved by: ken (mentor)

Instead of using .POSIX:, use the cool looking "@:" rule.

Submitted by: ru

Fix conflicts after heimdal-1.1 import and add build infrastructure. Import
all non-style changes made by heimdal to our own libgssapi.

Update for version 1.11-20080310.
Make our changes to CVS_RSH & CVS_SSH settings here instead of
contrib/cvs/src/cvs.h.

+ Make it clear this was taken from the CVS 1.11 branch on 10-March-2008.
+ Depend on $CVSDIR/configure also - so things get properly remade if we tweak.

Use a phony .POSIX target. This seems to be the way to
turn off the stuff in src/usr.bin/make/main.c rev 1.161.

Use our standard verbose spelling of rules variables.
(as a nice side affect, this will make gnu/usr.bin/cvs/contrib/Makefile
have a later date than contrib/cvs/contrib/Makefile.in - which will help
the build break after the 1.11.22 CVS import...)

Disable contrib - there is weirdness going on with it.

Reimplementation of world/kernel build options. For details, see:

http://lists.freebsd.org/pipermail/freebsd-current/2006-March/061725.html

The src.conf(5) manpage is to follow in a few days.

Brought to you by: imp, jhb, kris, phk, ru (all bugs are mine)

Add a new extensible GSS-API layer which can support GSS-API plugins,
similar the the Solaris implementation. Repackage the krb5 GSS mechanism
as a plugin library for the new implementation. This also includes a
comprehensive set of manpages for the GSS-API functions with text mostly
taken from the RFC.

Reviewed by: Love Hörnquist Åstrand <lha@it.su.se>, ru (build system), des (openssh parts)

Remove kludges intended to support src trees with partial obj trees.

Discussed with: ru

Tidy up cvs entries

NOCRYPT -> NO_CRYPT

Add removal of MKTEMP_SH_FUNCTION to processing for cvsbug script.

Approved by: peter (a while ago)
MFC After: 1 week

For variables that are only checked with defined(), don't provide
any fake value.

Make cvsbug(1) actually work instead of aborting due to missing configure
script substitution markers.

Fix the embedded version string in cvs and cvsbug. This breaks scripts
that parse it.

PR: 72139
Submitted by: Eugene M. Kim <gene@nttmcl.com>

Join the 21st century: Cryptography is no longer an optional component
of releases. The -DNOCRYPT build option still exists for anyone who
really wants to build non-cryptographic binaries, but the "crypto"
release distribution is now part of "base", and anyone installing from a
release will get cryptographic binaries.

Approved by: re (scottl), markm
Discussed on: freebsd-current, in late April 2004

Update bmake glue for cvs-1.11.15, with apologies to Jacques for messing
up his day. The *.patch files are still fine for MFC'ing.

Patch vulnerabilities in the CVS client and server:

A malicious CVS server could cause your CVS client to overwrite
arbitrary files (CAN-2004-0180).

When a CVS client uses the `-p' checkout option, the server could be
fooled into checking out files from outside the given $CVSROOT.

(This patch is applied in an unorthodox manner so as not to complicate
a later vendor import of CVS.)

Oops, use -I${DESTDIR}/usr/include/gnu instead of ... /../../include/gnu

Switch back to regex.h and fix error when mismatched regex.h used with
system libgnuregex

Fixed style of DPADD and LDADD assignments as per style.Makefile(5).

Only evaluate CVS version once, when necessary.
Eliminates the need in sed(1) during install.

Only evaluate CVS version once, when necessary.
Eliminates the need in sed(1) during install.

Only define a default editor if one isn't already defined.

Enable GSSAPI/Kerberos 5 support (aka `gserver').

MFC after: 1 month

Finish update for cvs-1.11.5. Some build-time tunables were moved from
options.h to config.h.in and set via ./configure when built normally.
Export some of the build knobs to the Makefile here, overridable
from /etc/make.conf. Also get the version strings right.
config.h was repocopied to config.h.proto, and we do a limited sed on it
at build time now.

Dont bother with unused rules for version.c

Update for cvs-1.11.5 import (note to self: gotta fix the version tags)

Sigh. Another wrong version string. However, this one isn't used in the
build.

Approved by: re

Oops. Use the correct version number. The number that was here was the
one that I originally used on freefall etc.

Approved by: re

Update bmake glue after 1.11.2.1-20021201 import. options.h is no longer
touched by configure in the autoconf system and a standard version is
now used.

Approved by: re

cvs-1.11.2 seems to bug out if HAVE_MMAP is not defined. It hoses its
internal buffer management somehow (an off by one perhaps). HAVE_MMAP
wasn't detected because configure has a bogus declaration of malloc
which conflicted with stdlib.h. Sigh.

version.h is now autogenerated in 1.11.2, not version.c

Update for 1.11.1p1 -> 1.11.2 import

It seems CVSvn.texi isn't used anymore.

Mark all internal libraries with INTERNALLIB.

Replaced exists() tests with two equivalent defined().

LIBDIR is defined in bsd.own.mk but sys.mk no longer
includes bsd.own.mk as of revision 1.60.

Restore the correct CLEANFILES that got damaged in rev. 1.15.

Allow to build with GCC 3.1.

FILES support for bsd.prog.mk. See bsd.README for details.

Stolen from: NetBSD

Use some slightly less horrible make magic for doing the version stuff.

One more version related tweak after cvs-1.11 -> 1.11.1p1

Update build rules for .in and tidy up.

Tools dir is now empty and unused

Generate version.c on the fly rather than using a generated file.

Get the version source from the (more) correct place.
Remove the obsolete cvsbug.sh script

*** empty log message ***

Add annotate.c, remove rtag.c for 1.11.1p1

beforeinstall -> SCRIPTS.

MAN[1-9] -> MAN.

Typo police.

Regenerated. There are a bunch of changes from this round of ./configure
but none of them are used or make any difference (at least in the FreeBSD
version).

Use libcrypto instead of libdes.

Minor style nits.

Fix typo: MAINAINER -> MAINTAINER

Update files generated by configure.

Add maintainer tags

Call isspace() directly to get rid of an objectionable include.

Support the environtmental var "CVS_OPTIONS". Which can hold a set of
default options for cvs. These options are interpreted first and can be
overwritten by explicit command line parameters.

Obtained from: GNU Grep 2.3

Change distribution to krb4 in preparation for K5.

$Source$,$Revision$ -> $FreeBSD$

Fix for new Common Error system.

$Header$ -> $FreeBSD$

$Id$ -> $FreeBSD$

Update bmake glue for cvs-1.10

Fix LIBDIR so kerberised CVS will work.

Simplified by using new yacc rules and by not generating y.tab.h.

Add an ".ifmake regress" to avoid warning during 'make world'.

Noticed by: bde

Activate CVS regression test.

indicate ability to support the PerservePermissions stuff

add new file (hardlink.c) into sources

comment that pvcs_to_rcs is left out since it needs perl5

Add libdiff

Build the internal libified diff

Update options.h as per configure generated version

Update config.h

Fixed missing dependency on version.c.

Fixed some style bugs ($@ is not recommended ...).

<bsd.prog.mk> has always included ../Makefile.inc, and there are no
complications involving .PATH or dependencies, so don't include it here.

Don't use LDDESTDIR. Just put the -L arg in LDADD.

If kerberos is installed and enabled in make.conf, activate cvs's kserver
hooks. This is a lot safer than 'cvs pserver'.

small change to prevent "modules" to be modified twice, on the following
execution sequences:
% easy-import -n foobaa
% easy-import foobaa

Reviewed by: joerg_wunsch@uriah.heep.sax.de

support remote imports (you must have a local cvs with same path as
remote for this to work)

Approved/Improved by: joerg

Update to build the new cvs version

Revert $FreeBSD$ to $Id$

Simplify using SRCDIR

Make the long-awaited change from $Id$ to $FreeBSD$

This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.

Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.

Same.

Use ${COPY} instead of -c for installing non-source files.

Added reminder to edit the parent Makefile to add what you just imported.

Make it clear that a vendor-branch import adds something to the
existing sources.

Requested by: markm

forgot to reconnect easy-import to the Makefile

Ack! Don't install cvs.info.gz etc in /usr/bin...

Make the CVSvn.texi dependency hack work a little better..

Oops, forgot to 'cvs add' this one.

Update to use the cvs-1.8.1 sources from src/contrib/cvs

Ensure at least one space between the module name and the module path.

Detected by: asami

The default editor is "vi", not "/usr/ucb/vi".

Check for, and disallow, duplicate tags in the "cvs import" command.
RCS cannot deal with duplicate tags; the extra one always becomes
inaccessible and useless.

This will prevent the common mistake of specifying the same name
for the vendor tag and the release tag. The FreeBSD CVS repository
already contains zillions of files with this error. We don't need
any more of them.

checklog - extract your commits from commitlogs archive

Bring in my changes for removing the pestilent obj links (unless you
really want them) from /usr/src. This is the final version of the
patches, incorporating the feedback I've received from -current.

Fix a bogon that prevented easy-import from inserting the new module
name in alphabetical order (broken by my previous commit).

Pointed out by: asami

Make `easy-import' ready for vendor-branch imports.

I'm also going to install the new version on freefall.

Another round of man page cleanups.

Down to only about 100 items left to cleanup! :-)

Whoops. cvsinit was creating a modules file in CVSROOT with an
explicit (and wrong) /usr/local/bin/mkmodules path. We install in /usr/bin.

Noticed by: Faried Nawaz <fn@pain.csrv.uidaho.edu>

Use cvs -Q release, instead of cvs release -Q, for removing the
modules module.

This commit was generated by cvs2svn to compensate for changes in r13122,
which included commits to RCS files with non-trunk default branches.

recording cvs-1.6 file death

This commit was generated by cvs2svn to compensate for changes in r13087,
which included commits to RCS files with non-trunk default branches.

Bah humbug!

Two uninitialised variables were causing a phkmalloc warning (another notch
in phkmalloc's belt) and caused the full rcstemplate to not be constructed
for commits on freefall.

Fix the convert to 'death-support' script to work with un-modified
RCS sources.

A one line hack to prevent 'cvs checkout -c' from triggering a nasty bug
in GNU getopt(). ('cvs checkout -c' cats the modules file to stdout)

Noticed by: Stefan Esser

Whitespace cleanup, add forgotten file to CLEANFILES..

Make CVS use /usr/lib/libmd.a rather than supplying Yet Another Copy(TM)
of md5.[ch]. Note that minor mods were made to md5.h to support this,
so a reinstall of lib/libmd is needed before building this cvs.

Tweak the install targets so that the installs are done in the right place
relative to the SUBDIR recursion.. Otherwise, the results look really odd.
(basically changed a couple of afterinstall's to beforeinstall's)

Argh! Botch alert! I dont *believe* I did this....
(I think I'll blame it on Tristan! :-)

Fix a couple of build warts that I broke...

This commit was generated by cvs2svn to compensate for changes in r12757,
which included commits to RCS files with non-trunk default branches.

Attach easy-import.pl (repostory copied from easy-import.perl) and
use the xPERL_PATHx substitution like the rest of the .pl stuff.

Clean out some files which are no longer part of CVS...

First round of the post-import cleanups...

This commit was generated by cvs2svn to compensate for changes in r12752,
which included commits to RCS files with non-trunk default branches.

Import CVS-1.6.3-951211.. Basically, this is the cvs-1.6.2 release
plus a couple of minor changes..

Some highlights of the new stuff that was not in the old version:
- remote access support.. full checkout/commit/log/etc..
- much improved dead file support..
- speed improvements
- better $CVSROOT handling
- $Name$ support
- support for a "cvsadmin" group to cut down rampant use of "cvs admin -o"
- safer setuid/setgid support
- many bugs fixed.. :-)
- probably some new ones.. :-(
- more that I cannot remember offhand..

This commit was generated by cvs2svn to compensate for changes in r12750,
which included commits to RCS files with non-trunk default branches.

Restore white spaces to original cvs-1.4Alpha2 state so that we can
deal with the cvs-1.6 (and beyond) imports more cleanly and safely.

Put a bandaid on the NULL pointer dereference caused by doing an
"update -jHEAD" when a file has been added on the specified tag.

It doesn't actually make cvs 'handle' it, it just stops it from dying
and leaving stray locks and other wreckage.

This was suggested by the CVS maintainers, and is in cvs-1.5.1-950901.

This changes two things when importing an unreadable file:
1: It stops invalid files being created in the cvs tree
2: It stops the import from aborting without mailing a commit message..

The first is simple, it opens the file for reading before touching the
repository, and the second catches the pieces when it hits an unreadable
file rather than just aborting mid-way through, leaving the repository in
a bit mess.

Reviewed by: rgrimes

Change `install' to `${INSTALL}' so that default install flags can be
specified in the top level Makefiles.

Make the tag checks more strict according to Rod's wishes.

Reviewed by:
Submitted by:
Obtained from:

Remove bogus references to /usr/ucb.

Remove trailing whitespace.

Backed out the only FreeBSD specific change to the stock CVS sources now
that it's taken care of in the CVSROOT/cvsignore file which is FreeBSD
specific.

This will make it easier to track changes in the *official* release.

Added back 'examples' to the subdirs which was bogusly removed.

Added Makefile for example files. They are now installed in
/usr/share/examples/cvs for use by the cvsinit script.

Forgot one FreeBSD specific change needed.

date: 1995/02/04 20:27:23; author: phk; state: Exp; lines: +1 -1
added *.info and *.db to the default "ignore" list in cvs to avoid obviously
common mistakes.

Merged in FreeBSD necessary changes in. Note, all of our source
modifications were *not* merged in (FreeBSD-DEVELOPER and symlink) but
the source is now the stock sources.

This commit was generated by cvs2svn to compensate for changes in r7516,
which included commits to RCS files with non-trunk default branches.

Original sources from CVS-1.4A2 munged to fit our directory structure.

This commit was generated by cvs2svn to compensate for changes in r7514,
which included commits to RCS files with non-trunk default branches.

Made this script a bit more fool-proof, so people like me can better
use it. :-)

It now explicitly requires the specification of a directory to import
from, either as an argument to the script, or by asking the user about
it. (Previously, it implicitly used `.', like cvs import does.)

Also implemented an option `-n', which does essentially the same like
the overall CVS option `-n': show only what would have been done,
don't do any commitment. Note that since the modules' database is
checked out in place (and not commited back), it will erroneously be
reported as to be imported, too:

cvs import: Importing /home/ncvs/ports/foobar/foo/modules
I ports/foobar/foo/modules/CVS
N ports/foobar/foo/modules/modules

This is an unwanted side-effect, but gives the user the option to see
if the `ed' magic did the right thing when editing modules/modules.

Rod, can you please check the function ``checktag'' in the script if it
will be restritctive enough?

Add my script here to make people's life easier when importing new
stuff. I'd like to have it in CVS, and i figured that this might be
the best place to go.

Someone (phk?) could install it into /usr/local/bin on freefall, for
convenience.

Suggested by: phk

added *.info and *.db to the default "ignore" list in cvs to avoid obviously
common mistakes.

Fix bogus install rule.

Use -lgnuregex

__386BSD__ -> __FreeBSD__

I know that many of these entries are bogus and need to be revisited,
but let's get the tree working again for now and then do a pass through
looking at all the __FreeBSD__ entries, shall we?

Adding cvsinit back to the subdirs since it shouldn't have been removed
in the first place.

remove cvsinit from subdir list

Added cvsinit shell script as a runnable command.

Submitted by: CVS mailing list.

Don't build profiled libraries. (NOPROFILE= yes. I don't like the negative
logic or lack of checking of truth values. There are too many ways to spell
"yes". Some Makefiles spell it as "no".)

Add dependencies on libraries to DPADD.

Move another include of bsd.prog.mk to the end.

Centralize declarations of library names.

Clean up small error in the ncvs code that was causing rcs co to be
called with -K-Keoptions -Kioptions. This should fix the problem with
$Id$ still getting changed.

I am also install both ncvs and ocvs on freefall with this change as
ocvs still had the bug with -I \! which I fixed but did not reinstall.

Got this fix from think.com:
The following two patches should allow the documented "-I !" option to
skip the ignore lists and import/update all files in the directory.

I need it to force certain files to import (ie, csh.a which is part of
the csh documentation).

Got this from think.com, nice patch to add Index: lines to cvs diff output
so you can put it right into patch.
Patch by Brian Berliner.

Enhancements to use now -K option to rcs so that certain keywords are
NOT touched by cvs. Only takes effect if -DFREEBSD_DEVELOPER, and installs
as ncvs when this is defined. You must also have the changes to rcs
for this to work.

Add -K option to pass keywordlists to rcs so we can disable and enable
specific keyword expansion on cvs co and cvs patch commands.

Added y.tab.h to the CLEANFILES target.

If we're not logging history information, don't demand write access.

Removed Jordan's DO_LINKS from the Makefile, since this behavior is not
'normal' CVS behavior, and we've never used nor depended on it.
(And at times I've gotten bit by it)

Really fixed the obj/noobj problems in CVS, and also made it so a
make install did not install libcvs in /usr/lib, since it has no
reason to do that.

New manual page system

Added a CLEANFILES entry for getdate.c (it's created by yacc)

Fixed up CVS like RCS to allow obj or no obj directory.

One small change to make the code clearer in update.

Changes to allow simple symlink handling

This commit was generated by cvs2svn to compensate for changes in r12,
which included commits to RCS files with non-trunk default branches.