strings.subr revision 256392
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 256392 2013-10-12 19:54:12Z 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 ...
60#
61# Similar to snprintf(3), write at most $size number of bytes into $var_to_set
62# using printf(1) syntax (`$format ...'). The value of $var_to_set is NULL
63# 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"=\$\( printf \"\$@\" \| awk -v max=\"\$__size\" \''
70	{
71		len = length($0)
72		max -= len
73		print substr($0,0,(max > 0 ? len : max + len))
74		if ( max < 0 ) exit
75		max--
76	}'\' \)
77}
78
79# f_vsnprintf $var_to_set $size $format $format_args
80#
81# Similar to vsnprintf(3), write at most $size number of bytes into $var_to_set
82# using printf(1) syntax (`$format $format_args'). The value of $var_to_set is
83# NULL unless at-least one byte is stored from the output.
84#
85# Example 1:
86#
87# 	limit=7 format="%s"
88# 	format_args="'abc   123'" # 3-spaces between abc and 123
89# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[abc   1]
90#
91# Example 2:
92#
93# 	limit=12 format="%s %s"
94# 	format_args="   'doghouse'      'foxhound'   "
95# 		# even more spaces added to illustrate escape-method
96# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[doghouse fox]
97#
98# Example 3:
99#
100# 	limit=13 format="%s %s"
101# 	f_shell_escape arg1 'aaa"aaa' # arg1=[aaa"aaa] (no change)
102# 	f_shell_escape arg2 "aaa'aaa" # arg2=[aaa'\''aaa] (escaped s-quote)
103# 	format_args="'$arg1' '$arg2'" # use single-quotes to surround args
104# 	f_vsnprintf foo $limit "$format" "$format_args" # foo=[aaa"aaa aaa'a]
105#
106# In all of the above examples, the call to f_vsnprintf() does not change. Only
107# the contents of $limit, $format, and $format_args changes in each example.
108#
109f_vsnprintf()
110{
111	eval f_snprintf \"\$1\" \"\$2\" \"\$3\" $4
112}
113
114# f_longest_line_length
115#
116# Simple wrapper to an awk(1) script to print the length of the longest line of
117# input (read from stdin). Supports the newline escape-sequence `\n' for
118# splitting a single line into multiple lines.
119#
120f_longest_line_length_awk='
121BEGIN { longest = 0 }
122{
123	if (split($0, lines, /\\n/) > 1)
124	{
125		for (n in lines)
126		{
127			len = length(lines[n])
128			longest = ( len > longest ? len : longest )
129		}
130	}
131	else
132	{
133		len = length($0)
134		longest = ( len > longest ? len : longest )
135	}
136}
137END { print longest }
138'
139f_longest_line_length()
140{
141	awk "$f_longest_line_length_awk"
142}
143
144# f_number_of_lines
145#
146# Simple wrapper to an awk(1) script to print the number of lines read from
147# stdin. Supports newline escape-sequence `\n' for splitting a single line into
148# multiple lines.
149#
150f_number_of_lines_awk='
151BEGIN { num_lines = 0 }
152{
153	num_lines += split(" "$0, unused, /\\n/)
154}
155END { print num_lines }
156'
157f_number_of_lines()
158{
159	awk "$f_number_of_lines_awk"
160}
161
162# f_isinteger $arg
163#
164# Returns true if argument is a positive/negative whole integer.
165#
166f_isinteger()
167{
168	local arg="$1"
169
170	# Prevent division-by-zero
171	[ "$arg" = "0" ] && return $SUCCESS
172
173	# Attempt to perform arithmetic divison (an operation which will exit
174	# with error unless arg is a valid positive/negative whole integer).
175	#
176	( : $((0/$arg)) ) > /dev/null 2>&1
177}
178
179# f_uriencode [$text]
180#
181# Encode $text for the purpose of embedding safely into a URL. Non-alphanumeric
182# characters are converted to `%XX' sequence where XX represents the hexa-
183# decimal ordinal of the non-alphanumeric character. If $text is missing, data
184# is instead read from standard input.
185#
186f_uriencode_awk='
187BEGIN {
188	output = ""
189	for (n = 0; n < 256; n++) pack[sprintf("%c", n)] = sprintf("%%%02x", n)
190}
191{
192	sline = ""
193	slen = length($0)
194	for (n = 1; n <= slen; n++) {
195		char = substr($0, n, 1)
196		if ( char !~ /^[[:alnum:]_]$/ ) char = pack[char]
197		sline = sline char
198	}
199	output = output ( output ? "%0a" : "" ) sline
200}
201END { print output }
202'
203f_uriencode()
204{
205	if [ $# -gt 0 ]; then
206		echo "$1" | awk "$f_uriencode_awk"
207	else
208		awk "$f_uriencode_awk"
209	fi
210}
211
212# f_uridecode [$text]
213#
214# Decode $text from a URI. Encoded characters are converted from their `%XX'
215# sequence into original unencoded ASCII sequences. If $text is missing, data
216# is instead read from standard input.
217#
218f_uridecode_awk='
219BEGIN { for (n = 0; n < 256; n++) chr[n] = sprintf("%c", n) }
220{
221	sline = ""
222	slen = length($0)
223	for (n = 1; n <= slen; n++)
224	{
225		seq = substr($0, n, 3)
226		if ( seq ~ /^%[[:xdigit:]][[:xdigit:]]$/ ) {
227			hex = substr(seq, 2, 2)
228			sline = sline chr[sprintf("%u", "0x"hex)]
229			n += 2
230		} else
231			sline = sline substr(seq, 1, 1)
232	}
233	print sline
234}
235'
236f_uridecode()
237{
238	if [ $# -gt 0 ]; then
239		echo "$1" | awk "$f_uridecode_awk"
240	else
241		awk "$f_uridecode_awk"
242	fi
243}
244
245# f_replaceall $string $find $replace [$var_to_set]
246#
247# Replace all occurrences of $find in $string with $replace. If $var_to_set is
248# either missing or NULL, the variable name is produced on standard out for
249# capturing in a sub-shell (which is less recommended due to performance
250# degradation).
251#
252f_replaceall()
253{
254	local __left="" __right="$1"
255	local __find="$2" __replace="$3" __var_to_set="$4"
256	while :; do
257		case "$__right" in *$__find*)
258			__left="$__left${__right%%$__find*}$__replace"
259			__right="${__right#*$__find}"
260			continue
261		esac
262		break
263	done
264	__left="$__left${__right#*$__find}"
265	if [ "$__var_to_set" ]; then
266		setvar "$__var_to_set" "$__left"
267	else
268		echo "$__left"
269	fi
270}
271
272# f_str2varname $string [$var_to_set]
273#
274# Convert a string into a suitable value to be used as a variable name
275# by converting unsuitable characters into the underscrore [_]. If $var_to_set
276# is either missing or NULL, the variable name is produced on standard out for
277# capturing in a sub-shell (which is less recommended due to performance
278# degradation).
279#
280f_str2varname()
281{
282	local __string="$1" __var_to_set="$2"
283	f_replaceall "$__string" "[!$VALID_VARNAME_CHARS]" "_" "$__var_to_set"
284}
285
286# f_shell_escape $string [$var_to_set]
287#
288# Escape $string for shell eval statement(s) by replacing all single-quotes
289# with a special sequence that creates a compound string when interpolated
290# by eval with surrounding single-quotes.
291#
292# For example:
293#
294# 	foo="abc'123"
295# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
296# 	eval echo \'$bar\' # produces abc'123
297#
298# This is helpful when processing an argument list that has to retain its
299# escaped structure for later evaluations.
300#
301# WARNING: Surrounding single-quotes are not added; this is the responsibility
302# of the code passing the escaped values to eval (which also aids readability).
303#
304f_shell_escape()
305{
306	local __string="$1" __var_to_set="$2"
307	f_replaceall "$__string" "'" "'\\''" "$__var_to_set"
308}
309
310# f_shell_unescape $string [$var_to_set]
311#
312# The antithesis of f_shell_escape(), this function takes an escaped $string
313# and expands it.
314#
315# For example:
316#
317# 	foo="abc'123"
318# 	f_shell_escape "$foo" bar # bar=[abc'\''123]
319# 	f_shell_unescape "$bar" # produces abc'123
320#
321f_shell_unescape()
322{
323	local __string="$1" __var_to_set="$2"
324	f_replaceall "$__string" "'\\''" "'" "$__var_to_set"
325}
326
327# f_expand_number $string [$var_to_set]
328#
329# Unformat $string into a number, optionally to be stored in $var_to_set. This
330# function follows the SI power of two convention.
331#
332# The prefixes are:
333#
334# 	Prefix	Description	Multiplier
335# 	k	kilo		1024
336# 	M	mega		1048576
337# 	G	giga		1073741824
338# 	T	tera		1099511627776
339# 	P	peta		1125899906842624
340# 	E	exa		1152921504606846976
341#
342# NOTE: Prefixes are case-insensitive.
343#
344# Upon successful completion, success status is returned; otherwise the number
345# -1 is produced ($var_to_set set to -1 or if $var_to_set is NULL or missing)
346# on standard output. In the case of failure, the error status will be one of:
347#
348# 	Status	Reason
349# 	1	Given $string contains no digits
350# 	2	An unrecognized prefix was given
351# 	3	Result too large to calculate
352#
353f_expand_number()
354{
355	local __string="$1" __var_to_set="$2"
356	local __cp __num __bshift __maxinput
357
358	# Remove any leading non-digits
359	while :; do
360		__cp="$__string"
361		__string="${__cp#[!0-9]}"
362		[ "$__string" = "$__cp" ] && break
363	done
364
365	# Produce `-1' if string didn't contain any digits
366	if [ ! "$__string" ]; then
367		if [ "$__var_to_set" ]; then
368			setvar "$__var_to_set" -1
369		else
370			echo -1
371		fi
372		return 1 # 1 = "Given $string contains no digits"
373	fi
374
375	# Store the numbers
376	__num="${__string%%[!0-9]*}"
377
378	# Shortcut
379	if [ $__num -eq 0 ]; then
380		if [ "$__var_to_set" ]; then
381			setvar "$__var_to_set" 0
382		else
383			echo 0
384		fi
385		return $SUCCESS
386	fi
387
388	# Remove all the leading numbers from the string to get at the prefix
389	while :; do
390		__cp="$__string"
391		__string="${__cp#[0-9]}"
392		[ "$__string" = "$__cp" ] && break
393	done
394
395	#
396	# Test for invalid prefix (and determine bitshift length)
397	#
398	case "$__string" in
399	""|[[:space:]]*) # Shortcut
400		if [ "$__var_to_set" ]; then
401			setvar "$__var_to_set" $__num
402		else
403			echo $__num
404		fi
405		return $SUCCESS ;;
406	[Kk]*) __bshift=10 ;;
407	[Mm]*) __bshift=20 ;;
408	[Gg]*) __bshift=30 ;;
409	[Tt]*) __bshift=40 ;;
410	[Pp]*) __bshift=50 ;;
411	[Ee]*) __bshift=60 ;;
412	*)
413		# Unknown prefix
414		if [ "$__var_to_set" ]; then
415			setvar "$__var_to_set" -1
416		else
417			echo -1
418		fi
419		return 2 # 2 = "An unrecognized prefix was given"
420	esac
421
422	# Determine if the wheels fall off
423	__maxinput=$(( 0x7fffffffffffffff >> $__bshift ))
424	if [ $__num -gt $__maxinput ]; then
425		# Input (before expanding) would exceed 64-bit signed int
426		if [ "$__var_to_set" ]; then
427			setvar "$__var_to_set" -1
428		else
429			echo -1
430		fi
431		return 3 # 3 = "Result too large to calculate"
432	fi
433
434	# Shift the number out and produce it
435	__num=$(( $__num << $__bshift ))
436	if [ "$__var_to_set" ]; then
437		setvar "$__var_to_set" $__num
438	else
439		echo $__num
440	fi
441}
442
443############################################################ MAIN
444
445f_dprintf "%s: Successfully loaded." strings.subr
446
447fi # ! $_STRINGS_SUBR
448