strings.subr revision 258420
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 258420 2013-11-21 03:38:47Z 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="$1"
192
193	# Prevent division-by-zero
194	[ "$arg" = "0" ] && return $SUCCESS
195
196	# Attempt to perform arithmetic divison (an operation which will exit
197	# with error unless arg is a valid positive/negative whole integer).
198	#
199	( : $((0/$arg)) ) > /dev/null 2>&1
200}
201
202# f_uriencode [$text]
203#
204# Encode $text for the purpose of embedding safely into a URL. Non-alphanumeric
205# characters are converted to `%XX' sequence where XX represents the hexa-
206# decimal ordinal of the non-alphanumeric character. If $text is missing, data
207# is instead read from standard input.
208#
209f_uriencode_awk='
210BEGIN {
211	output = ""
212	for (n = 0; n < 256; n++) pack[sprintf("%c", n)] = sprintf("%%%02x", n)
213}
214{
215	sline = ""
216	slen = length($0)
217	for (n = 1; n <= slen; n++) {
218		char = substr($0, n, 1)
219		if ( char !~ /^[[:alnum:]_]$/ ) char = pack[char]
220		sline = sline char
221	}
222	output = output ( output ? "%0a" : "" ) sline
223}
224END { print output }
225'
226f_uriencode()
227{
228	if [ $# -gt 0 ]; then
229		echo "$1" | awk "$f_uriencode_awk"
230	else
231		awk "$f_uriencode_awk"
232	fi
233}
234
235# f_uridecode [$text]
236#
237# Decode $text from a URI. Encoded characters are converted from their `%XX'
238# sequence into original unencoded ASCII sequences. If $text is missing, data
239# is instead read from standard input.
240#
241f_uridecode_awk='
242BEGIN { for (n = 0; n < 256; n++) chr[n] = sprintf("%c", n) }
243{
244	sline = ""
245	slen = length($0)
246	for (n = 1; n <= slen; n++)
247	{
248		seq = substr($0, n, 3)
249		if ( seq ~ /^%[[:xdigit:]][[:xdigit:]]$/ ) {
250			hex = substr(seq, 2, 2)
251			sline = sline chr[sprintf("%u", "0x"hex)]
252			n += 2
253		} else
254			sline = sline substr(seq, 1, 1)
255	}
256	print sline
257}
258'
259f_uridecode()
260{
261	if [ $# -gt 0 ]; then
262		echo "$1" | awk "$f_uridecode_awk"
263	else
264		awk "$f_uridecode_awk"
265	fi
266}
267
268# f_replaceall $string $find $replace [$var_to_set]
269#
270# Replace all occurrences of $find in $string with $replace. If $var_to_set is
271# either missing or NULL, the variable name is produced on standard out for
272# capturing in a sub-shell (which is less recommended due to performance
273# degradation).
274#
275f_replaceall()
276{
277	local __left="" __right="$1"
278	local __find="$2" __replace="$3" __var_to_set="$4"
279	while :; do
280		case "$__right" in *$__find*)
281			__left="$__left${__right%%$__find*}$__replace"
282			__right="${__right#*$__find}"
283			continue
284		esac
285		break
286	done
287	__left="$__left${__right#*$__find}"
288	if [ "$__var_to_set" ]; then
289		setvar "$__var_to_set" "$__left"
290	else
291		echo "$__left"
292	fi
293}
294
295# f_str2varname $string [$var_to_set]
296#
297# Convert a string into a suitable value to be used as a variable name
298# by converting unsuitable characters into the underscrore [_]. If $var_to_set
299# is either missing or NULL, the variable name is produced on standard out for
300# capturing in a sub-shell (which is less recommended due to performance
301# degradation).
302#
303f_str2varname()
304{
305	local __string="$1" __var_to_set="$2"
306	f_replaceall "$__string" "[!$VALID_VARNAME_CHARS]" "_" "$__var_to_set"
307}
308
309# f_shell_escape $string [$var_to_set]
310#
311# Escape $string for shell eval statement(s) by replacing all single-quotes
312# with a special sequence that creates a compound string when interpolated
313# by eval with surrounding single-quotes.
314#
315# For example:
316#
317# 	foo="abc'123"
318# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
319# 	eval echo \'$bar\' # produces abc'123
320#
321# This is helpful when processing an argument list that has to retain its
322# escaped structure for later evaluations.
323#
324# WARNING: Surrounding single-quotes are not added; this is the responsibility
325# of the code passing the escaped values to eval (which also aids readability).
326#
327f_shell_escape()
328{
329	local __string="$1" __var_to_set="$2"
330	f_replaceall "$__string" "'" "'\\''" "$__var_to_set"
331}
332
333# f_shell_unescape $string [$var_to_set]
334#
335# The antithesis of f_shell_escape(), this function takes an escaped $string
336# and expands it.
337#
338# For example:
339#
340# 	foo="abc'123"
341# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
342# 	f_shell_unescape "$bar" # produces abc'123
343#
344f_shell_unescape()
345{
346	local __string="$1" __var_to_set="$2"
347	f_replaceall "$__string" "'\\''" "'" "$__var_to_set"
348}
349
350# f_expand_number $string [$var_to_set]
351#
352# Unformat $string into a number, optionally to be stored in $var_to_set. This
353# function follows the SI power of two convention.
354#
355# The prefixes are:
356#
357# 	Prefix	Description	Multiplier
358# 	k	kilo		1024
359# 	M	mega		1048576
360# 	G	giga		1073741824
361# 	T	tera		1099511627776
362# 	P	peta		1125899906842624
363# 	E	exa		1152921504606846976
364#
365# NOTE: Prefixes are case-insensitive.
366#
367# Upon successful completion, success status is returned; otherwise the number
368# -1 is produced ($var_to_set set to -1 or if $var_to_set is NULL or missing)
369# on standard output. In the case of failure, the error status will be one of:
370#
371# 	Status	Reason
372# 	1	Given $string contains no digits
373# 	2	An unrecognized prefix was given
374# 	3	Result too large to calculate
375#
376f_expand_number()
377{
378	local __string="$1" __var_to_set="$2"
379	local __cp __num __bshift __maxinput
380
381	# Remove any leading non-digits
382	while :; do
383		__cp="$__string"
384		__string="${__cp#[!0-9]}"
385		[ "$__string" = "$__cp" ] && break
386	done
387
388	# Produce `-1' if string didn't contain any digits
389	if [ ! "$__string" ]; then
390		if [ "$__var_to_set" ]; then
391			setvar "$__var_to_set" -1
392		else
393			echo -1
394		fi
395		return 1 # 1 = "Given $string contains no digits"
396	fi
397
398	# Store the numbers
399	__num="${__string%%[!0-9]*}"
400
401	# Shortcut
402	if [ $__num -eq 0 ]; then
403		if [ "$__var_to_set" ]; then
404			setvar "$__var_to_set" 0
405		else
406			echo 0
407		fi
408		return $SUCCESS
409	fi
410
411	# Remove all the leading numbers from the string to get at the prefix
412	while :; do
413		__cp="$__string"
414		__string="${__cp#[0-9]}"
415		[ "$__string" = "$__cp" ] && break
416	done
417
418	#
419	# Test for invalid prefix (and determine bitshift length)
420	#
421	case "$__string" in
422	""|[[:space:]]*) # Shortcut
423		if [ "$__var_to_set" ]; then
424			setvar "$__var_to_set" $__num
425		else
426			echo $__num
427		fi
428		return $SUCCESS ;;
429	[Kk]*) __bshift=10 ;;
430	[Mm]*) __bshift=20 ;;
431	[Gg]*) __bshift=30 ;;
432	[Tt]*) __bshift=40 ;;
433	[Pp]*) __bshift=50 ;;
434	[Ee]*) __bshift=60 ;;
435	*)
436		# Unknown prefix
437		if [ "$__var_to_set" ]; then
438			setvar "$__var_to_set" -1
439		else
440			echo -1
441		fi
442		return 2 # 2 = "An unrecognized prefix was given"
443	esac
444
445	# Determine if the wheels fall off
446	__maxinput=$(( 0x7fffffffffffffff >> $__bshift ))
447	if [ $__num -gt $__maxinput ]; then
448		# Input (before expanding) would exceed 64-bit signed int
449		if [ "$__var_to_set" ]; then
450			setvar "$__var_to_set" -1
451		else
452			echo -1
453		fi
454		return 3 # 3 = "Result too large to calculate"
455	fi
456
457	# Shift the number out and produce it
458	__num=$(( $__num << $__bshift ))
459	if [ "$__var_to_set" ]; then
460		setvar "$__var_to_set" $__num
461	else
462		echo $__num
463	fi
464}
465
466############################################################ MAIN
467
468f_dprintf "%s: Successfully loaded." strings.subr
469
470fi # ! $_STRINGS_SUBR
471