str.c revision 1.25 1 1.25 christos /* $NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $ */
2 1.10 christos
3 1.1 cgd /*-
4 1.13 christos * Copyright (c) 1988, 1989, 1990, 1993
5 1.13 christos * The Regents of the University of California. All rights reserved.
6 1.20 agc *
7 1.20 agc * This code is derived from software contributed to Berkeley by
8 1.20 agc * Adam de Boor.
9 1.20 agc *
10 1.20 agc * Redistribution and use in source and binary forms, with or without
11 1.20 agc * modification, are permitted provided that the following conditions
12 1.20 agc * are met:
13 1.20 agc * 1. Redistributions of source code must retain the above copyright
14 1.20 agc * notice, this list of conditions and the following disclaimer.
15 1.20 agc * 2. Redistributions in binary form must reproduce the above copyright
16 1.20 agc * notice, this list of conditions and the following disclaimer in the
17 1.20 agc * documentation and/or other materials provided with the distribution.
18 1.20 agc * 3. Neither the name of the University nor the names of its contributors
19 1.20 agc * may be used to endorse or promote products derived from this software
20 1.20 agc * without specific prior written permission.
21 1.20 agc *
22 1.20 agc * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23 1.20 agc * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24 1.20 agc * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25 1.20 agc * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26 1.20 agc * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27 1.20 agc * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28 1.20 agc * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29 1.20 agc * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30 1.20 agc * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31 1.20 agc * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 1.20 agc * SUCH DAMAGE.
33 1.20 agc */
34 1.20 agc
35 1.20 agc /*-
36 1.1 cgd * Copyright (c) 1989 by Berkeley Softworks
37 1.1 cgd * All rights reserved.
38 1.1 cgd *
39 1.1 cgd * This code is derived from software contributed to Berkeley by
40 1.1 cgd * Adam de Boor.
41 1.1 cgd *
42 1.1 cgd * Redistribution and use in source and binary forms, with or without
43 1.1 cgd * modification, are permitted provided that the following conditions
44 1.1 cgd * are met:
45 1.1 cgd * 1. Redistributions of source code must retain the above copyright
46 1.1 cgd * notice, this list of conditions and the following disclaimer.
47 1.1 cgd * 2. Redistributions in binary form must reproduce the above copyright
48 1.1 cgd * notice, this list of conditions and the following disclaimer in the
49 1.1 cgd * documentation and/or other materials provided with the distribution.
50 1.1 cgd * 3. All advertising materials mentioning features or use of this software
51 1.1 cgd * must display the following acknowledgement:
52 1.1 cgd * This product includes software developed by the University of
53 1.1 cgd * California, Berkeley and its contributors.
54 1.1 cgd * 4. Neither the name of the University nor the names of its contributors
55 1.1 cgd * may be used to endorse or promote products derived from this software
56 1.1 cgd * without specific prior written permission.
57 1.1 cgd *
58 1.1 cgd * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
59 1.1 cgd * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
60 1.1 cgd * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
61 1.1 cgd * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
62 1.1 cgd * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
63 1.1 cgd * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
64 1.1 cgd * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
65 1.1 cgd * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
66 1.1 cgd * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
67 1.1 cgd * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
68 1.1 cgd * SUCH DAMAGE.
69 1.1 cgd */
70 1.1 cgd
71 1.22 ross #ifndef MAKE_NATIVE
72 1.25 christos static char rcsid[] = "$NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $";
73 1.15 lukem #else
74 1.14 christos #include <sys/cdefs.h>
75 1.1 cgd #ifndef lint
76 1.10 christos #if 0
77 1.10 christos static char sccsid[] = "@(#)str.c 5.8 (Berkeley) 6/1/90";
78 1.10 christos #else
79 1.25 christos __RCSID("$NetBSD: str.c,v 1.25 2006/08/11 19:11:00 christos Exp $");
80 1.10 christos #endif
81 1.1 cgd #endif /* not lint */
82 1.15 lukem #endif
83 1.1 cgd
84 1.1 cgd #include "make.h"
85 1.1 cgd
86 1.1 cgd /*-
87 1.1 cgd * str_concat --
88 1.1 cgd * concatenate the two strings, inserting a space or slash between them,
89 1.1 cgd * freeing them if requested.
90 1.1 cgd *
91 1.1 cgd * returns --
92 1.1 cgd * the resulting string in allocated space.
93 1.1 cgd */
94 1.1 cgd char *
95 1.19 christos str_concat(const char *s1, const char *s2, int flags)
96 1.1 cgd {
97 1.18 wiz int len1, len2;
98 1.18 wiz char *result;
99 1.1 cgd
100 1.1 cgd /* get the length of both strings */
101 1.1 cgd len1 = strlen(s1);
102 1.1 cgd len2 = strlen(s2);
103 1.1 cgd
104 1.1 cgd /* allocate length plus separator plus EOS */
105 1.1 cgd result = emalloc((u_int)(len1 + len2 + 2));
106 1.1 cgd
107 1.1 cgd /* copy first string into place */
108 1.4 cgd memcpy(result, s1, len1);
109 1.1 cgd
110 1.1 cgd /* add separator character */
111 1.1 cgd if (flags & STR_ADDSPACE) {
112 1.1 cgd result[len1] = ' ';
113 1.1 cgd ++len1;
114 1.1 cgd } else if (flags & STR_ADDSLASH) {
115 1.1 cgd result[len1] = '/';
116 1.1 cgd ++len1;
117 1.1 cgd }
118 1.1 cgd
119 1.1 cgd /* copy second string plus EOS into place */
120 1.4 cgd memcpy(result + len1, s2, len2 + 1);
121 1.1 cgd
122 1.1 cgd return(result);
123 1.1 cgd }
124 1.1 cgd
125 1.1 cgd /*-
126 1.1 cgd * brk_string --
127 1.1 cgd * Fracture a string into an array of words (as delineated by tabs or
128 1.1 cgd * spaces) taking quotation marks into account. Leading tabs/spaces
129 1.1 cgd * are ignored.
130 1.1 cgd *
131 1.1 cgd * returns --
132 1.21 sjg * Pointer to the array of pointers to the words.
133 1.1 cgd */
134 1.1 cgd char **
135 1.19 christos brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
136 1.1 cgd {
137 1.18 wiz int argc, ch;
138 1.19 christos char inquote, *start, *t;
139 1.19 christos const char *p;
140 1.1 cgd int len;
141 1.16 christos int argmax = 50, curlen = 0;
142 1.23 christos char **argv = emalloc((argmax + 1) * sizeof(char *));
143 1.1 cgd
144 1.4 cgd /* skip leading space chars. */
145 1.4 cgd for (; *str == ' ' || *str == '\t'; ++str)
146 1.4 cgd continue;
147 1.1 cgd
148 1.1 cgd /* allocate room for a copy of the string */
149 1.16 christos if ((len = strlen(str) + 1) > curlen)
150 1.16 christos *buffer = emalloc(curlen = len);
151 1.1 cgd
152 1.1 cgd /*
153 1.1 cgd * copy the string; at the same time, parse backslashes,
154 1.1 cgd * quotes and build the argument list.
155 1.1 cgd */
156 1.16 christos argc = 0;
157 1.1 cgd inquote = '\0';
158 1.16 christos for (p = str, start = t = *buffer;; ++p) {
159 1.1 cgd switch(ch = *p) {
160 1.1 cgd case '"':
161 1.1 cgd case '\'':
162 1.17 christos if (inquote) {
163 1.1 cgd if (inquote == ch)
164 1.4 cgd inquote = '\0';
165 1.1 cgd else
166 1.1 cgd break;
167 1.17 christos }
168 1.6 jtc else {
169 1.4 cgd inquote = (char) ch;
170 1.6 jtc /* Don't miss "" or '' */
171 1.6 jtc if (start == NULL && p[1] == inquote) {
172 1.6 jtc start = t + 1;
173 1.25 christos p++;
174 1.25 christos inquote = '\0';
175 1.6 jtc break;
176 1.6 jtc }
177 1.6 jtc }
178 1.8 jtc if (!expand) {
179 1.8 jtc if (!start)
180 1.8 jtc start = t;
181 1.8 jtc *t++ = ch;
182 1.8 jtc }
183 1.1 cgd continue;
184 1.1 cgd case ' ':
185 1.1 cgd case '\t':
186 1.8 jtc case '\n':
187 1.1 cgd if (inquote)
188 1.1 cgd break;
189 1.1 cgd if (!start)
190 1.1 cgd continue;
191 1.1 cgd /* FALLTHROUGH */
192 1.1 cgd case '\0':
193 1.1 cgd /*
194 1.1 cgd * end of a token -- make sure there's enough argv
195 1.1 cgd * space and save off a pointer.
196 1.1 cgd */
197 1.8 jtc if (!start)
198 1.8 jtc goto done;
199 1.8 jtc
200 1.1 cgd *t++ = '\0';
201 1.1 cgd if (argc == argmax) {
202 1.1 cgd argmax *= 2; /* ramp up fast */
203 1.12 jtc argv = (char **)erealloc(argv,
204 1.12 jtc (argmax + 1) * sizeof(char *));
205 1.1 cgd }
206 1.1 cgd argv[argc++] = start;
207 1.24 christos start = NULL;
208 1.1 cgd if (ch == '\n' || ch == '\0')
209 1.1 cgd goto done;
210 1.1 cgd continue;
211 1.1 cgd case '\\':
212 1.8 jtc if (!expand) {
213 1.8 jtc if (!start)
214 1.8 jtc start = t;
215 1.8 jtc *t++ = '\\';
216 1.8 jtc ch = *++p;
217 1.8 jtc break;
218 1.8 jtc }
219 1.13 christos
220 1.1 cgd switch (ch = *++p) {
221 1.1 cgd case '\0':
222 1.1 cgd case '\n':
223 1.1 cgd /* hmmm; fix it up as best we can */
224 1.1 cgd ch = '\\';
225 1.1 cgd --p;
226 1.1 cgd break;
227 1.1 cgd case 'b':
228 1.1 cgd ch = '\b';
229 1.1 cgd break;
230 1.1 cgd case 'f':
231 1.1 cgd ch = '\f';
232 1.1 cgd break;
233 1.1 cgd case 'n':
234 1.1 cgd ch = '\n';
235 1.1 cgd break;
236 1.1 cgd case 'r':
237 1.1 cgd ch = '\r';
238 1.1 cgd break;
239 1.1 cgd case 't':
240 1.1 cgd ch = '\t';
241 1.1 cgd break;
242 1.1 cgd }
243 1.1 cgd break;
244 1.1 cgd }
245 1.1 cgd if (!start)
246 1.1 cgd start = t;
247 1.4 cgd *t++ = (char) ch;
248 1.1 cgd }
249 1.24 christos done: argv[argc] = NULL;
250 1.1 cgd *store_argc = argc;
251 1.1 cgd return(argv);
252 1.1 cgd }
253 1.1 cgd
254 1.1 cgd /*
255 1.1 cgd * Str_FindSubstring -- See if a string contains a particular substring.
256 1.13 christos *
257 1.18 wiz * Input:
258 1.18 wiz * string String to search.
259 1.18 wiz * substring Substring to find in string.
260 1.18 wiz *
261 1.1 cgd * Results: If string contains substring, the return value is the location of
262 1.1 cgd * the first matching instance of substring in string. If string doesn't
263 1.1 cgd * contain substring, the return value is NULL. Matching is done on an exact
264 1.1 cgd * character-for-character basis with no wildcards or special characters.
265 1.13 christos *
266 1.1 cgd * Side effects: None.
267 1.1 cgd */
268 1.1 cgd char *
269 1.19 christos Str_FindSubstring(const char *string, const char *substring)
270 1.1 cgd {
271 1.19 christos const char *a, *b;
272 1.1 cgd
273 1.1 cgd /*
274 1.1 cgd * First scan quickly through the two strings looking for a single-
275 1.1 cgd * character match. When it's found, then compare the rest of the
276 1.1 cgd * substring.
277 1.1 cgd */
278 1.1 cgd
279 1.1 cgd for (b = substring; *string != 0; string += 1) {
280 1.1 cgd if (*string != *b)
281 1.1 cgd continue;
282 1.1 cgd a = string;
283 1.1 cgd for (;;) {
284 1.1 cgd if (*b == 0)
285 1.19 christos return UNCONST(string);
286 1.1 cgd if (*a++ != *b++)
287 1.1 cgd break;
288 1.1 cgd }
289 1.1 cgd b = substring;
290 1.1 cgd }
291 1.19 christos return NULL;
292 1.1 cgd }
293 1.1 cgd
294 1.1 cgd /*
295 1.1 cgd * Str_Match --
296 1.13 christos *
297 1.1 cgd * See if a particular string matches a particular pattern.
298 1.13 christos *
299 1.1 cgd * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
300 1.1 cgd * matching operation permits the following special characters in the
301 1.1 cgd * pattern: *?\[] (see the man page for details on what these mean).
302 1.13 christos *
303 1.1 cgd * Side effects: None.
304 1.1 cgd */
305 1.4 cgd int
306 1.19 christos Str_Match(const char *string, const char *pattern)
307 1.1 cgd {
308 1.1 cgd char c2;
309 1.1 cgd
310 1.1 cgd for (;;) {
311 1.1 cgd /*
312 1.1 cgd * See if we're at the end of both the pattern and the
313 1.1 cgd * string. If, we succeeded. If we're at the end of the
314 1.1 cgd * pattern but not at the end of the string, we failed.
315 1.1 cgd */
316 1.1 cgd if (*pattern == 0)
317 1.1 cgd return(!*string);
318 1.1 cgd if (*string == 0 && *pattern != '*')
319 1.1 cgd return(0);
320 1.1 cgd /*
321 1.1 cgd * Check for a "*" as the next pattern character. It matches
322 1.1 cgd * any substring. We handle this by calling ourselves
323 1.1 cgd * recursively for each postfix of string, until either we
324 1.1 cgd * match or we reach the end of the string.
325 1.1 cgd */
326 1.1 cgd if (*pattern == '*') {
327 1.1 cgd pattern += 1;
328 1.1 cgd if (*pattern == 0)
329 1.1 cgd return(1);
330 1.1 cgd while (*string != 0) {
331 1.1 cgd if (Str_Match(string, pattern))
332 1.1 cgd return(1);
333 1.1 cgd ++string;
334 1.1 cgd }
335 1.1 cgd return(0);
336 1.1 cgd }
337 1.1 cgd /*
338 1.1 cgd * Check for a "?" as the next pattern character. It matches
339 1.1 cgd * any single character.
340 1.1 cgd */
341 1.1 cgd if (*pattern == '?')
342 1.1 cgd goto thisCharOK;
343 1.1 cgd /*
344 1.1 cgd * Check for a "[" as the next pattern character. It is
345 1.1 cgd * followed by a list of characters that are acceptable, or
346 1.1 cgd * by a range (two characters separated by "-").
347 1.1 cgd */
348 1.1 cgd if (*pattern == '[') {
349 1.1 cgd ++pattern;
350 1.1 cgd for (;;) {
351 1.1 cgd if ((*pattern == ']') || (*pattern == 0))
352 1.1 cgd return(0);
353 1.1 cgd if (*pattern == *string)
354 1.1 cgd break;
355 1.1 cgd if (pattern[1] == '-') {
356 1.1 cgd c2 = pattern[2];
357 1.1 cgd if (c2 == 0)
358 1.1 cgd return(0);
359 1.1 cgd if ((*pattern <= *string) &&
360 1.1 cgd (c2 >= *string))
361 1.1 cgd break;
362 1.1 cgd if ((*pattern >= *string) &&
363 1.1 cgd (c2 <= *string))
364 1.1 cgd break;
365 1.1 cgd pattern += 2;
366 1.1 cgd }
367 1.1 cgd ++pattern;
368 1.1 cgd }
369 1.1 cgd while ((*pattern != ']') && (*pattern != 0))
370 1.1 cgd ++pattern;
371 1.1 cgd goto thisCharOK;
372 1.1 cgd }
373 1.1 cgd /*
374 1.1 cgd * If the next pattern character is '/', just strip off the
375 1.1 cgd * '/' so we do exact matching on the character that follows.
376 1.1 cgd */
377 1.1 cgd if (*pattern == '\\') {
378 1.1 cgd ++pattern;
379 1.1 cgd if (*pattern == 0)
380 1.1 cgd return(0);
381 1.1 cgd }
382 1.1 cgd /*
383 1.1 cgd * There's no special character. Just make sure that the
384 1.1 cgd * next characters of each string match.
385 1.1 cgd */
386 1.1 cgd if (*pattern != *string)
387 1.1 cgd return(0);
388 1.1 cgd thisCharOK: ++pattern;
389 1.1 cgd ++string;
390 1.1 cgd }
391 1.4 cgd }
392 1.4 cgd
393 1.4 cgd
394 1.4 cgd /*-
395 1.4 cgd *-----------------------------------------------------------------------
396 1.4 cgd * Str_SYSVMatch --
397 1.13 christos * Check word against pattern for a match (% is wild),
398 1.13 christos *
399 1.18 wiz * Input:
400 1.18 wiz * word Word to examine
401 1.18 wiz * pattern Pattern to examine against
402 1.18 wiz * len Number of characters to substitute
403 1.18 wiz *
404 1.4 cgd * Results:
405 1.4 cgd * Returns the beginning position of a match or null. The number
406 1.4 cgd * of characters matched is returned in len.
407 1.4 cgd *
408 1.4 cgd * Side Effects:
409 1.4 cgd * None
410 1.4 cgd *
411 1.4 cgd *-----------------------------------------------------------------------
412 1.4 cgd */
413 1.4 cgd char *
414 1.19 christos Str_SYSVMatch(const char *word, const char *pattern, int *len)
415 1.4 cgd {
416 1.19 christos const char *p = pattern;
417 1.19 christos const char *w = word;
418 1.19 christos const char *m;
419 1.4 cgd
420 1.5 jtc if (*p == '\0') {
421 1.5 jtc /* Null pattern is the whole string */
422 1.5 jtc *len = strlen(w);
423 1.19 christos return UNCONST(w);
424 1.5 jtc }
425 1.4 cgd
426 1.4 cgd if ((m = strchr(p, '%')) != NULL) {
427 1.4 cgd /* check that the prefix matches */
428 1.4 cgd for (; p != m && *w && *w == *p; w++, p++)
429 1.4 cgd continue;
430 1.4 cgd
431 1.4 cgd if (p != m)
432 1.4 cgd return NULL; /* No match */
433 1.4 cgd
434 1.4 cgd if (*++p == '\0') {
435 1.4 cgd /* No more pattern, return the rest of the string */
436 1.4 cgd *len = strlen(w);
437 1.19 christos return UNCONST(w);
438 1.4 cgd }
439 1.4 cgd }
440 1.4 cgd
441 1.4 cgd m = w;
442 1.4 cgd
443 1.4 cgd /* Find a matching tail */
444 1.4 cgd do
445 1.4 cgd if (strcmp(p, w) == 0) {
446 1.4 cgd *len = w - m;
447 1.19 christos return UNCONST(m);
448 1.4 cgd }
449 1.4 cgd while (*w++ != '\0');
450 1.13 christos
451 1.4 cgd return NULL;
452 1.4 cgd }
453 1.4 cgd
454 1.4 cgd
455 1.4 cgd /*-
456 1.4 cgd *-----------------------------------------------------------------------
457 1.4 cgd * Str_SYSVSubst --
458 1.4 cgd * Substitute '%' on the pattern with len characters from src.
459 1.4 cgd * If the pattern does not contain a '%' prepend len characters
460 1.4 cgd * from src.
461 1.13 christos *
462 1.4 cgd * Results:
463 1.4 cgd * None
464 1.4 cgd *
465 1.4 cgd * Side Effects:
466 1.4 cgd * Places result on buf
467 1.4 cgd *
468 1.4 cgd *-----------------------------------------------------------------------
469 1.4 cgd */
470 1.4 cgd void
471 1.18 wiz Str_SYSVSubst(Buffer buf, char *pat, char *src, int len)
472 1.4 cgd {
473 1.4 cgd char *m;
474 1.4 cgd
475 1.4 cgd if ((m = strchr(pat, '%')) != NULL) {
476 1.4 cgd /* Copy the prefix */
477 1.24 christos Buf_AddBytes(buf, m - pat, (Byte *)pat);
478 1.4 cgd /* skip the % */
479 1.4 cgd pat = m + 1;
480 1.4 cgd }
481 1.4 cgd
482 1.4 cgd /* Copy the pattern */
483 1.24 christos Buf_AddBytes(buf, len, (Byte *)src);
484 1.4 cgd
485 1.4 cgd /* append the rest */
486 1.24 christos Buf_AddBytes(buf, strlen(pat), (Byte *)pat);
487 1.1 cgd }
488