1 /* quotearg.c - quote arguments for output
2 Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
4 This program is free software; you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation; either version 2, or (at your option)
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program; if not, write to the Free Software Foundation,
16 Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
18 /* Written by Paul Eggert <eggert@twinsun.com> */
24 #include <sys/types.h>
29 #if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII))
32 # define ISASCII(c) isascii (c)
34 #define ISPRINT(c) (ISASCII (c) && isprint (c))
38 # define _(text) gettext (text)
50 # define UCHAR_MAX ((unsigned char) -1)
61 #if HAVE_MBRTOWC && HAVE_WCHAR_H
64 # define iswprint(wc) 1
65 # define mbrtowc(pwc, s, n, ps) 1
66 # define mbsinit(ps) 1
67 # define mbstate_t int
70 #define INT_BITS (sizeof (int) * CHAR_BIT)
72 struct quoting_options
74 /* Basic quoting style. */
75 enum quoting_style style;
77 /* Quote the characters indicated by this bit vector even if the
78 quoting style would not normally require them to be quoted. */
79 int quote_these_too[((UCHAR_MAX + 1) / INT_BITS
80 + ((UCHAR_MAX + 1) % INT_BITS != 0))];
83 /* Names of quoting styles. */
84 char const *const quoting_style_args[] =
95 /* Correspondences to quoting style names. */
96 enum quoting_style const quoting_style_vals[] =
98 literal_quoting_style,
100 shell_always_quoting_style,
102 escape_quoting_style,
106 /* The default quoting options. */
107 static struct quoting_options default_quoting_options;
109 /* Allocate a new set of quoting options, with contents initially identical
110 to O if O is not null, or to the default if O is null.
111 It is the caller's responsibility to free the result. */
112 struct quoting_options *
113 clone_quoting_options (struct quoting_options *o)
115 struct quoting_options *p
116 = (struct quoting_options *) xmalloc (sizeof (struct quoting_options));
117 *p = *(o ? o : &default_quoting_options);
121 /* Get the value of O's quoting style. If O is null, use the default. */
123 get_quoting_style (struct quoting_options *o)
125 return (o ? o : &default_quoting_options)->style;
128 /* In O (or in the default if O is null),
129 set the value of the quoting style to S. */
131 set_quoting_style (struct quoting_options *o, enum quoting_style s)
133 (o ? o : &default_quoting_options)->style = s;
136 /* In O (or in the default if O is null),
137 set the value of the quoting options for character C to I.
138 Return the old value. Currently, the only values defined for I are
139 0 (the default) and 1 (which means to quote the character even if
140 it would not otherwise be quoted). */
142 set_char_quoting (struct quoting_options *o, char c, int i)
144 unsigned char uc = c;
145 int *p = (o ? o : &default_quoting_options)->quote_these_too + uc / INT_BITS;
146 int shift = uc % INT_BITS;
147 int r = (*p >> shift) & 1;
148 *p ^= ((i & 1) ^ r) << shift;
152 /* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of
153 argument ARG (of size ARGSIZE), using QUOTING_STYLE and the
154 non-quoting-style part of O to control quoting.
155 Terminate the output with a null character, and return the written
156 size of the output, not counting the terminating null.
157 If BUFFERSIZE is too small to store the output string, return the
158 value that would have been returned had BUFFERSIZE been large enough.
159 If ARGSIZE is -1, use the string length of the argument for ARGSIZE.
161 This function acts like quotearg_buffer (BUFFER, BUFFERSIZE, ARG,
162 ARGSIZE, O), except it uses QUOTING_STYLE instead of the quoting
163 style specified by O, and O may not be null. */
166 quotearg_buffer_restyled (char *buffer, size_t buffersize,
167 char const *arg, size_t argsize,
168 enum quoting_style quoting_style,
169 struct quoting_options const *o)
173 char const *quote_string = 0;
174 size_t quote_string_len = 0;
175 int backslash_escapes = 0;
180 if (len < buffersize) \
186 switch (quoting_style)
188 case c_quoting_style:
190 backslash_escapes = 1;
192 quote_string_len = 1;
195 case escape_quoting_style:
196 backslash_escapes = 1;
199 case locale_quoting_style:
200 for (quote_string = _("`"); *quote_string; quote_string++)
201 STORE (*quote_string);
202 backslash_escapes = 1;
203 quote_string = _("'");
204 quote_string_len = strlen (quote_string);
207 case shell_always_quoting_style:
210 quote_string_len = 1;
217 for (i = 0; ! (argsize == (size_t) -1 ? arg[i] == '\0' : i == argsize); i++)
222 if (backslash_escapes
224 && i + quote_string_len <= argsize
225 && memcmp (arg + i, quote_string, quote_string_len) == 0)
232 switch (quoting_style)
234 case shell_quoting_style:
235 goto use_shell_always_quoting_style;
237 case c_quoting_style:
238 if (i + 2 < argsize && arg[i + 1] == '?')
242 case '(': case ')': case '-': case '/':
243 case '<': case '=': case '>':
244 /* Escape the second '?' in what would otherwise be
260 #if HAVE_C_BACKSLASH_A
261 case '\a': esc = 'a'; goto c_escape;
263 case '\b': esc = 'b'; goto c_escape;
264 case '\f': esc = 'f'; goto c_escape;
265 case '\n': esc = 'n'; goto c_escape;
266 case '\r': esc = 'r'; goto c_escape;
267 case '\t': esc = 't'; goto c_escape;
268 case '\v': esc = 'v'; goto c_escape;
269 case '\\': esc = c; goto c_escape;
272 if (backslash_escapes)
277 if (quoting_style == shell_quoting_style)
278 goto use_shell_always_quoting_style;
286 case '!': /* special in bash */
287 case '"': case '$': case '&':
288 case '(': case ')': case '*': case ';':
289 case '<': case '>': case '[':
290 case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */
292 /* A shell special character. In theory, '$' and '`' could
293 be the first bytes of multibyte characters, which means
294 we should check them with mbrtowc, but in practice this
295 doesn't happen so it's not worth worrying about. */
296 if (quoting_style == shell_quoting_style)
297 goto use_shell_always_quoting_style;
301 switch (quoting_style)
303 case shell_quoting_style:
304 goto use_shell_always_quoting_style;
306 case shell_always_quoting_style:
317 case '%': case '+': case ',': case '-': case '.': case '/':
318 case '0': case '1': case '2': case '3': case '4': case '5':
319 case '6': case '7': case '8': case '9': case ':': case '=':
320 case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
321 case 'G': case 'H': case 'I': case 'J': case 'K': case 'L':
322 case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R':
323 case 'S': case 'T': case 'U': case 'V': case 'W': case 'X':
324 case 'Y': case 'Z': case ']': case '_': case 'a': case 'b':
325 case 'c': case 'd': case 'e': case 'f': case 'g': case 'h':
326 case 'i': case 'j': case 'k': case 'l': case 'm': case 'n':
327 case 'o': case 'p': case 'q': case 'r': case 's': case 't':
328 case 'u': case 'v': case 'w': case 'x': case 'y': case 'z':
330 /* These characters don't cause problems, no matter what the
331 quoting style is. They cannot start multibyte sequences. */
335 /* If we have a multibyte sequence, copy it until we reach
336 its end, find an error, or come back to the initial shift
337 state. For C-like styles, if the sequence has
338 unprintable characters, escape the whole sequence, since
339 we can't easily escape single characters within it. */
341 /* Length of multibyte sequence found so far. */
346 memset (&mbstate, 0, sizeof mbstate);
348 if (argsize == (size_t) -1)
349 argsize = strlen (arg);
354 size_t bytes = mbrtowc (&w, &arg[i + m],
355 argsize - (i + m), &mbstate);
358 else if (bytes == (size_t) -1)
363 else if (bytes == (size_t) -2)
366 while (i + m < argsize && arg[i + m])
377 while (! mbsinit (&mbstate));
381 /* Escape a unibyte character like a multibyte
382 sequence if using backslash escapes, and if the
383 character is not printable. */
384 m = backslash_escapes && ! ISPRINT (c);
390 /* Output a multibyte sequence, or an escaped
391 unprintable unibyte character. */
392 size_t imax = i + m - 1;
396 if (backslash_escapes && ! printable)
399 STORE ('0' + (c >> 6));
400 STORE ('0' + ((c >> 3) & 7));
414 if (! (backslash_escapes
415 && o->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS))))
426 for (; *quote_string; quote_string++)
427 STORE (*quote_string);
429 if (len < buffersize)
433 use_shell_always_quoting_style:
434 return quotearg_buffer_restyled (buffer, buffersize, arg, argsize,
435 shell_always_quoting_style, o);
438 /* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of
439 argument ARG (of size ARGSIZE), using O to control quoting.
440 If O is null, use the default.
441 Terminate the output with a null character, and return the written
442 size of the output, not counting the terminating null.
443 If BUFFERSIZE is too small to store the output string, return the
444 value that would have been returned had BUFFERSIZE been large enough.
445 If ARGSIZE is -1, use the string length of the argument for ARGSIZE. */
447 quotearg_buffer (char *buffer, size_t buffersize,
448 char const *arg, size_t argsize,
449 struct quoting_options const *o)
451 struct quoting_options const *p = o ? o : &default_quoting_options;
452 return quotearg_buffer_restyled (buffer, buffersize, arg, argsize,
456 /* Use storage slot N to return a quoted version of the string ARG.
457 OPTIONS specifies the quoting options.
458 The returned value points to static storage that can be
459 reused by the next call to this function with the same value of N.
460 N must be nonnegative. N is deliberately declared with type `int'
461 to allow for future extensions (using negative values). */
463 quotearg_n_options (int n, char const *arg,
464 struct quoting_options const *options)
466 static unsigned int nslots;
467 static struct slotvec
476 size_t s = n1 * sizeof (struct slotvec);
477 if (! (0 < n1 && n1 == s / sizeof (struct slotvec)))
479 slotvec = (struct slotvec *) xrealloc (slotvec, s);
480 memset (slotvec + nslots, 0, (n1 - nslots) * sizeof (struct slotvec));
485 size_t size = slotvec[n].size;
486 char *val = slotvec[n].val;
487 size_t qsize = quotearg_buffer (val, size, arg, (size_t) -1, options);
491 slotvec[n].size = size = qsize + 1;
492 slotvec[n].val = val = xrealloc (val, size);
493 quotearg_buffer (val, size, arg, (size_t) -1, options);
501 quotearg_n (unsigned int n, char const *arg)
503 return quotearg_n_options (n, arg, &default_quoting_options);
507 quotearg (char const *arg)
509 return quotearg_n (0, arg);
513 quotearg_n_style (unsigned int n, enum quoting_style s, char const *arg)
515 struct quoting_options o;
517 memset (o.quote_these_too, 0, sizeof o.quote_these_too);
518 return quotearg_n_options (n, arg, &o);
522 quotearg_style (enum quoting_style s, char const *arg)
524 return quotearg_n_style (0, s, arg);
528 quotearg_char (char const *arg, char ch)
530 struct quoting_options options;
531 options = default_quoting_options;
532 set_char_quoting (&options, ch, 1);
533 return quotearg_n_options (0, arg, &options);
537 quotearg_colon (char const *arg)
539 return quotearg_char (arg, ':');