strings.subr revision 260678
1if [ ! "$_STRINGS_SUBR" ]; then _STRINGS_SUBR=1
2#
3# Copyright (c) 2006-2013 Devin Teske
4# All rights reserved.
5#
6# Redistribution and use in source and binary forms, with or without
7# modification, are permitted provided that the following conditions
8# are met:
9# 1. Redistributions of source code must retain the above copyright
10#    notice, this list of conditions and the following disclaimer.
11# 2. Redistributions in binary form must reproduce the above copyright
12#    notice, this list of conditions and the following disclaimer in the
13#    documentation and/or other materials provided with the distribution.
14#
15# THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
16# ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
17# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
18# ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
19# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
20# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
21# OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
22# HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
23# LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
24# OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
25# SUCH DAMAGE.
26#
27# $FreeBSD: stable/10/usr.sbin/bsdconfig/share/strings.subr 260678 2014-01-15 07:49:17Z dteske $
28#
29############################################################ INCLUDES
30
31BSDCFG_SHARE="/usr/share/bsdconfig"
32. $BSDCFG_SHARE/common.subr || exit 1
33
34############################################################ GLOBALS
35
36#
37# Valid characters that can appear in an sh(1) variable name
38#
39# Please note that the character ranges A-Z and a-z should be avoided because
40# these can include accent characters (which are not valid in a variable name).
41# For example, A-Z matches any character that sorts after A but before Z,
42# including A and Z. Although ASCII order would make more sense, that is not
43# how it works.
44#
45VALID_VARNAME_CHARS="0-9ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz_"
46
47############################################################ FUNCTIONS
48
49# f_substr "$string" $start [$length]
50#
51# Simple wrapper to awk(1)'s `substr' function.
52#
53f_substr()
54{
55	local string="$1" start="${2:-0}" len="${3:-0}"
56	echo "$string" | awk "{ print substr(\$0, $start, $len) }"
57}
58
59# f_snprintf $var_to_set $size $format [$arguments ...]
60#
61# Similar to snprintf(3), write at most $size number of bytes into $var_to_set
62# using printf(1) syntax (`$format [$arguments ...]'). The value of $var_to_set
63# is NULL unless at-least one byte is stored from the output.
64#
65f_snprintf()
66{
67	local __var_to_set="$1" __size="$2"
68	shift 2 # var_to_set size
69	eval "$__var_to_set"=\$\( printf -- \"\$@\" \| \
70		awk -v max=\"\$__size\" \''
71	{
72		len = length($0)
73		max -= len
74		print substr($0,0,(max > 0 ? len : max + len))
75		if ( max < 0 ) exit
76		max--
77	}'\' \)
78}
79
80# f_sprintf $var_to_set $format [$arguments ...]
81#
82# Similar to sprintf(3), write a string into $var_to_set using printf(1) syntax
83# (`$format [$arguments ...]').
84#
85f_sprintf()
86{
87	local __var_to_set="$1"
88	shift 1 # var_to_set
89	eval "$__var_to_set"=\$\( printf -- \"\$@\" \)
90}
91
92# f_vsnprintf $var_to_set $size $format $format_args
93#
94# Similar to vsnprintf(3), write at most $size number of bytes into $var_to_set
95# using printf(1) syntax (`$format $format_args'). The value of $var_to_set is
96# NULL unless at-least one byte is stored from the output.
97#
98# Example 1:
99#
100# 	limit=7 format="%s"
101# 	format_args="'abc   123'" # 3-spaces between abc and 123
102# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[abc   1]
103#
104# Example 2:
105#
106# 	limit=12 format="%s %s"
107# 	format_args="   'doghouse'      'foxhound'   "
108# 		# even more spaces added to illustrate escape-method
109# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[doghouse fox]
110#
111# Example 3:
112#
113# 	limit=13 format="%s %s"
114# 	f_shell_escape arg1 'aaa"aaa' # arg1=[aaa"aaa] (no change)
115# 	f_shell_escape arg2 "aaa'aaa" # arg2=[aaa'\''aaa] (escaped s-quote)
116# 	format_args="'$arg1' '$arg2'" # use single-quotes to surround args
117# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[aaa"aaa aaa'a]
118#
119# In all of the above examples, the call to f_vsnprintf() does not change. Only
120# the contents of $limit, $format, and $format_args changes in each example.
121#
122f_vsnprintf()
123{
124	eval f_snprintf \"\$1\" \"\$2\" \"\$3\" $4
125}
126
127# f_vsprintf $var_to_set $format $format_args
128#
129# Similar to vsprintf(3), write a string into $var_to_set using printf(1)
130# syntax (`$format $format_args').
131#
132f_vsprintf()
133{
134	eval f_sprintf \"\$1\" \"\$2\" $3
135}
136
137# f_longest_line_length
138#
139# Simple wrapper to an awk(1) script to print the length of the longest line of
140# input (read from stdin). Supports the newline escape-sequence `\n' for
141# splitting a single line into multiple lines.
142#
143f_longest_line_length_awk='
144BEGIN { longest = 0 }
145{
146	if (split($0, lines, /\\n/) > 1)
147	{
148		for (n in lines)
149		{
150			len = length(lines[n])
151			longest = ( len > longest ? len : longest )
152		}
153	}
154	else
155	{
156		len = length($0)
157		longest = ( len > longest ? len : longest )
158	}
159}
160END { print longest }
161'
162f_longest_line_length()
163{
164	awk "$f_longest_line_length_awk"
165}
166
167# f_number_of_lines
168#
169# Simple wrapper to an awk(1) script to print the number of lines read from
170# stdin. Supports newline escape-sequence `\n' for splitting a single line into
171# multiple lines.
172#
173f_number_of_lines_awk='
174BEGIN { num_lines = 0 }
175{
176	num_lines += split(" "$0, unused, /\\n/)
177}
178END { print num_lines }
179'
180f_number_of_lines()
181{
182	awk "$f_number_of_lines_awk"
183}
184
185# f_isinteger $arg
186#
187# Returns true if argument is a positive/negative whole integer.
188#
189f_isinteger()
190{
191	local arg="${1#-}"
192	[ "${arg:-x}" = "${arg#[!0-9]*}" ]
193}
194
195# f_uriencode [$text]
196#
197# Encode $text for the purpose of embedding safely into a URL. Non-alphanumeric
198# characters are converted to `%XX' sequence where XX represents the hexa-
199# decimal ordinal of the non-alphanumeric character. If $text is missing, data
200# is instead read from standard input.
201#
202f_uriencode_awk='
203BEGIN {
204	output = ""
205	for (n = 0; n < 256; n++) pack[sprintf("%c", n)] = sprintf("%%%02x", n)
206}
207{
208	sline = ""
209	slen = length($0)
210	for (n = 1; n <= slen; n++) {
211		char = substr($0, n, 1)
212		if ( char !~ /^[[:alnum:]_]$/ ) char = pack[char]
213		sline = sline char
214	}
215	output = output ( output ? "%0a" : "" ) sline
216}
217END { print output }
218'
219f_uriencode()
220{
221	if [ $# -gt 0 ]; then
222		echo "$1" | awk "$f_uriencode_awk"
223	else
224		awk "$f_uriencode_awk"
225	fi
226}
227
228# f_uridecode [$text]
229#
230# Decode $text from a URI. Encoded characters are converted from their `%XX'
231# sequence into original unencoded ASCII sequences. If $text is missing, data
232# is instead read from standard input.
233#
234f_uridecode_awk='
235BEGIN { for (n = 0; n < 256; n++) chr[n] = sprintf("%c", n) }
236{
237	sline = ""
238	slen = length($0)
239	for (n = 1; n <= slen; n++)
240	{
241		seq = substr($0, n, 3)
242		if ( seq ~ /^%[[:xdigit:]][[:xdigit:]]$/ ) {
243			hex = substr(seq, 2, 2)
244			sline = sline chr[sprintf("%u", "0x"hex)]
245			n += 2
246		} else
247			sline = sline substr(seq, 1, 1)
248	}
249	print sline
250}
251'
252f_uridecode()
253{
254	if [ $# -gt 0 ]; then
255		echo "$1" | awk "$f_uridecode_awk"
256	else
257		awk "$f_uridecode_awk"
258	fi
259}
260
261# f_replaceall $string $find $replace [$var_to_set]
262#
263# Replace all occurrences of $find in $string with $replace. If $var_to_set is
264# either missing or NULL, the variable name is produced on standard out for
265# capturing in a sub-shell (which is less recommended due to performance
266# degradation).
267#
268f_replaceall()
269{
270	local __left="" __right="$1"
271	local __find="$2" __replace="$3" __var_to_set="$4"
272	while :; do
273		case "$__right" in *$__find*)
274			__left="$__left${__right%%$__find*}$__replace"
275			__right="${__right#*$__find}"
276			continue
277		esac
278		break
279	done
280	__left="$__left${__right#*$__find}"
281	if [ "$__var_to_set" ]; then
282		setvar "$__var_to_set" "$__left"
283	else
284		echo "$__left"
285	fi
286}
287
288# f_str2varname $string [$var_to_set]
289#
290# Convert a string into a suitable value to be used as a variable name
291# by converting unsuitable characters into the underscrore [_]. If $var_to_set
292# is either missing or NULL, the variable name is produced on standard out for
293# capturing in a sub-shell (which is less recommended due to performance
294# degradation).
295#
296f_str2varname()
297{
298	local __string="$1" __var_to_set="$2"
299	f_replaceall "$__string" "[!$VALID_VARNAME_CHARS]" "_" "$__var_to_set"
300}
301
302# f_shell_escape $string [$var_to_set]
303#
304# Escape $string for shell eval statement(s) by replacing all single-quotes
305# with a special sequence that creates a compound string when interpolated
306# by eval with surrounding single-quotes.
307#
308# For example:
309#
310# 	foo="abc'123"
311# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
312# 	eval echo \'$bar\' # produces abc'123
313#
314# This is helpful when processing an argument list that has to retain its
315# escaped structure for later evaluations.
316#
317# WARNING: Surrounding single-quotes are not added; this is the responsibility
318# of the code passing the escaped values to eval (which also aids readability).
319#
320f_shell_escape()
321{
322	local __string="$1" __var_to_set="$2"
323	f_replaceall "$__string" "'" "'\\''" "$__var_to_set"
324}
325
326# f_shell_unescape $string [$var_to_set]
327#
328# The antithesis of f_shell_escape(), this function takes an escaped $string
329# and expands it.
330#
331# For example:
332#
333# 	foo="abc'123"
334# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
335# 	f_shell_unescape "$bar" # produces abc'123
336#
337f_shell_unescape()
338{
339	local __string="$1" __var_to_set="$2"
340	f_replaceall "$__string" "'\\''" "'" "$__var_to_set"
341}
342
343# f_expand_number $string [$var_to_set]
344#
345# Unformat $string into a number, optionally to be stored in $var_to_set. This
346# function follows the SI power of two convention.
347#
348# The prefixes are:
349#
350# 	Prefix	Description	Multiplier
351# 	k	kilo		1024
352# 	M	mega		1048576
353# 	G	giga		1073741824
354# 	T	tera		1099511627776
355# 	P	peta		1125899906842624
356# 	E	exa		1152921504606846976
357#
358# NOTE: Prefixes are case-insensitive.
359#
360# Upon successful completion, success status is returned; otherwise the number
361# -1 is produced ($var_to_set set to -1 or if $var_to_set is NULL or missing)
362# on standard output. In the case of failure, the error status will be one of:
363#
364# 	Status	Reason
365# 	1	Given $string contains no digits
366# 	2	An unrecognized prefix was given
367# 	3	Result too large to calculate
368#
369f_expand_number()
370{
371	local __string="$1" __var_to_set="$2"
372	local __cp __num __bshift __maxinput
373
374	# Remove any leading non-digits
375	while :; do
376		__cp="$__string"
377		__string="${__cp#[!0-9]}"
378		[ "$__string" = "$__cp" ] && break
379	done
380
381	# Produce `-1' if string didn't contain any digits
382	if [ ! "$__string" ]; then
383		if [ "$__var_to_set" ]; then
384			setvar "$__var_to_set" -1
385		else
386			echo -1
387		fi
388		return 1 # 1 = "Given $string contains no digits"
389	fi
390
391	# Store the numbers
392	__num="${__string%%[!0-9]*}"
393
394	# Shortcut
395	if [ $__num -eq 0 ]; then
396		if [ "$__var_to_set" ]; then
397			setvar "$__var_to_set" 0
398		else
399			echo 0
400		fi
401		return $SUCCESS
402	fi
403
404	# Remove all the leading numbers from the string to get at the prefix
405	while :; do
406		__cp="$__string"
407		__string="${__cp#[0-9]}"
408		[ "$__string" = "$__cp" ] && break
409	done
410
411	#
412	# Test for invalid prefix (and determine bitshift length)
413	#
414	case "$__string" in
415	""|[[:space:]]*) # Shortcut
416		if [ "$__var_to_set" ]; then
417			setvar "$__var_to_set" $__num
418		else
419			echo $__num
420		fi
421		return $SUCCESS ;;
422	[Kk]*) __bshift=10 ;;
423	[Mm]*) __bshift=20 ;;
424	[Gg]*) __bshift=30 ;;
425	[Tt]*) __bshift=40 ;;
426	[Pp]*) __bshift=50 ;;
427	[Ee]*) __bshift=60 ;;
428	*)
429		# Unknown prefix
430		if [ "$__var_to_set" ]; then
431			setvar "$__var_to_set" -1
432		else
433			echo -1
434		fi
435		return 2 # 2 = "An unrecognized prefix was given"
436	esac
437
438	# Determine if the wheels fall off
439	__maxinput=$(( 0x7fffffffffffffff >> $__bshift ))
440	if [ $__num -gt $__maxinput ]; then
441		# Input (before expanding) would exceed 64-bit signed int
442		if [ "$__var_to_set" ]; then
443			setvar "$__var_to_set" -1
444		else
445			echo -1
446		fi
447		return 3 # 3 = "Result too large to calculate"
448	fi
449
450	# Shift the number out and produce it
451	__num=$(( $__num << $__bshift ))
452	if [ "$__var_to_set" ]; then
453		setvar "$__var_to_set" $__num
454	else
455		echo $__num
456	fi
457}
458
459############################################################ MAIN
460
461f_dprintf "%s: Successfully loaded." strings.subr
462
463fi # ! $_STRINGS_SUBR
464