str.c revision 1.10 1 /* $NetBSD: str.c,v 1.10 1995/06/14 15:19:56 christos Exp $ */
2
3 /*-
4 * Copyright (c) 1988, 1989, 1990 The Regents of the University of California.
5 * Copyright (c) 1988, 1989 by Adam de Boor
6 * Copyright (c) 1989 by Berkeley Softworks
7 * All rights reserved.
8 *
9 * This code is derived from software contributed to Berkeley by
10 * Adam de Boor.
11 *
12 * Redistribution and use in source and binary forms, with or without
13 * modification, are permitted provided that the following conditions
14 * are met:
15 * 1. Redistributions of source code must retain the above copyright
16 * notice, this list of conditions and the following disclaimer.
17 * 2. Redistributions in binary form must reproduce the above copyright
18 * notice, this list of conditions and the following disclaimer in the
19 * documentation and/or other materials provided with the distribution.
20 * 3. All advertising materials mentioning features or use of this software
21 * must display the following acknowledgement:
22 * This product includes software developed by the University of
23 * California, Berkeley and its contributors.
24 * 4. Neither the name of the University nor the names of its contributors
25 * may be used to endorse or promote products derived from this software
26 * without specific prior written permission.
27 *
28 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
29 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
30 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
31 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
32 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
33 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
34 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
35 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
36 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
37 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
38 * SUCH DAMAGE.
39 */
40
41 #ifndef lint
42 #if 0
43 static char sccsid[] = "@(#)str.c 5.8 (Berkeley) 6/1/90";
44 #else
45 static char rcsid[] = "$NetBSD: str.c,v 1.10 1995/06/14 15:19:56 christos Exp $";
46 #endif
47 #endif /* not lint */
48
49 #include "make.h"
50
51 static char **argv, *buffer;
52 static int argmax, curlen;
53
54 /*
55 * str_init --
56 * Initialize the strings package
57 *
58 */
59 void
60 str_init()
61 {
62 char *p1;
63 argv = (char **)emalloc((argmax = 50) * sizeof(char *));
64 argv[0] = Var_Value(".MAKE", VAR_GLOBAL, &p1);
65 }
66
67
68 /*
69 * str_end --
70 * Cleanup the strings package
71 *
72 */
73 void
74 str_end()
75 {
76 if (argv[0]) {
77 free(argv[0]);
78 free((Address) argv);
79 }
80 if (buffer)
81 free(buffer);
82 }
83
84 /*-
85 * str_concat --
86 * concatenate the two strings, inserting a space or slash between them,
87 * freeing them if requested.
88 *
89 * returns --
90 * the resulting string in allocated space.
91 */
92 char *
93 str_concat(s1, s2, flags)
94 char *s1, *s2;
95 int flags;
96 {
97 register int len1, len2;
98 register char *result;
99
100 /* get the length of both strings */
101 len1 = strlen(s1);
102 len2 = strlen(s2);
103
104 /* allocate length plus separator plus EOS */
105 result = emalloc((u_int)(len1 + len2 + 2));
106
107 /* copy first string into place */
108 memcpy(result, s1, len1);
109
110 /* add separator character */
111 if (flags & STR_ADDSPACE) {
112 result[len1] = ' ';
113 ++len1;
114 } else if (flags & STR_ADDSLASH) {
115 result[len1] = '/';
116 ++len1;
117 }
118
119 /* copy second string plus EOS into place */
120 memcpy(result + len1, s2, len2 + 1);
121
122 /* free original strings */
123 if (flags & STR_DOFREE) {
124 (void)free(s1);
125 (void)free(s2);
126 }
127 return(result);
128 }
129
130 /*-
131 * brk_string --
132 * Fracture a string into an array of words (as delineated by tabs or
133 * spaces) taking quotation marks into account. Leading tabs/spaces
134 * are ignored.
135 *
136 * returns --
137 * Pointer to the array of pointers to the words. To make life easier,
138 * the first word is always the value of the .MAKE variable.
139 */
140 char **
141 brk_string(str, store_argc, expand)
142 register char *str;
143 int *store_argc;
144 Boolean expand;
145 {
146 register int argc, ch;
147 register char inquote, *p, *start, *t;
148 int len;
149
150 /* skip leading space chars. */
151 for (; *str == ' ' || *str == '\t'; ++str)
152 continue;
153
154 /* allocate room for a copy of the string */
155 if ((len = strlen(str) + 1) > curlen) {
156 if (buffer)
157 free(buffer);
158 buffer = emalloc(curlen = len);
159 }
160
161 /*
162 * copy the string; at the same time, parse backslashes,
163 * quotes and build the argument list.
164 */
165 argc = 1;
166 inquote = '\0';
167 for (p = str, start = t = buffer;; ++p) {
168 switch(ch = *p) {
169 case '"':
170 case '\'':
171 if (inquote)
172 if (inquote == ch)
173 inquote = '\0';
174 else
175 break;
176 else {
177 inquote = (char) ch;
178 /* Don't miss "" or '' */
179 if (start == NULL && p[1] == inquote) {
180 start = t + 1;
181 break;
182 }
183 }
184 if (!expand) {
185 if (!start)
186 start = t;
187 *t++ = ch;
188 }
189 continue;
190 case ' ':
191 case '\t':
192 case '\n':
193 if (inquote)
194 break;
195 if (!start)
196 continue;
197 /* FALLTHROUGH */
198 case '\0':
199 /*
200 * end of a token -- make sure there's enough argv
201 * space and save off a pointer.
202 */
203 if (!start)
204 goto done;
205
206 *t++ = '\0';
207 if (argc == argmax) {
208 argmax *= 2; /* ramp up fast */
209 if (!(argv = (char **)realloc(argv,
210 argmax * sizeof(char *))))
211 enomem();
212 }
213 argv[argc++] = start;
214 start = (char *)NULL;
215 if (ch == '\n' || ch == '\0')
216 goto done;
217 continue;
218 case '\\':
219 if (!expand) {
220 if (!start)
221 start = t;
222 *t++ = '\\';
223 ch = *++p;
224 break;
225 }
226
227 switch (ch = *++p) {
228 case '\0':
229 case '\n':
230 /* hmmm; fix it up as best we can */
231 ch = '\\';
232 --p;
233 break;
234 case 'b':
235 ch = '\b';
236 break;
237 case 'f':
238 ch = '\f';
239 break;
240 case 'n':
241 ch = '\n';
242 break;
243 case 'r':
244 ch = '\r';
245 break;
246 case 't':
247 ch = '\t';
248 break;
249 }
250 break;
251 }
252 if (!start)
253 start = t;
254 *t++ = (char) ch;
255 }
256 done: argv[argc] = (char *)NULL;
257 *store_argc = argc;
258 return(argv);
259 }
260
261 /*
262 * Str_FindSubstring -- See if a string contains a particular substring.
263 *
264 * Results: If string contains substring, the return value is the location of
265 * the first matching instance of substring in string. If string doesn't
266 * contain substring, the return value is NULL. Matching is done on an exact
267 * character-for-character basis with no wildcards or special characters.
268 *
269 * Side effects: None.
270 */
271 char *
272 Str_FindSubstring(string, substring)
273 register char *string; /* String to search. */
274 char *substring; /* Substring to find in string */
275 {
276 register char *a, *b;
277
278 /*
279 * First scan quickly through the two strings looking for a single-
280 * character match. When it's found, then compare the rest of the
281 * substring.
282 */
283
284 for (b = substring; *string != 0; string += 1) {
285 if (*string != *b)
286 continue;
287 a = string;
288 for (;;) {
289 if (*b == 0)
290 return(string);
291 if (*a++ != *b++)
292 break;
293 }
294 b = substring;
295 }
296 return((char *) NULL);
297 }
298
299 /*
300 * Str_Match --
301 *
302 * See if a particular string matches a particular pattern.
303 *
304 * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
305 * matching operation permits the following special characters in the
306 * pattern: *?\[] (see the man page for details on what these mean).
307 *
308 * Side effects: None.
309 */
310 int
311 Str_Match(string, pattern)
312 register char *string; /* String */
313 register char *pattern; /* Pattern */
314 {
315 char c2;
316
317 for (;;) {
318 /*
319 * See if we're at the end of both the pattern and the
320 * string. If, we succeeded. If we're at the end of the
321 * pattern but not at the end of the string, we failed.
322 */
323 if (*pattern == 0)
324 return(!*string);
325 if (*string == 0 && *pattern != '*')
326 return(0);
327 /*
328 * Check for a "*" as the next pattern character. It matches
329 * any substring. We handle this by calling ourselves
330 * recursively for each postfix of string, until either we
331 * match or we reach the end of the string.
332 */
333 if (*pattern == '*') {
334 pattern += 1;
335 if (*pattern == 0)
336 return(1);
337 while (*string != 0) {
338 if (Str_Match(string, pattern))
339 return(1);
340 ++string;
341 }
342 return(0);
343 }
344 /*
345 * Check for a "?" as the next pattern character. It matches
346 * any single character.
347 */
348 if (*pattern == '?')
349 goto thisCharOK;
350 /*
351 * Check for a "[" as the next pattern character. It is
352 * followed by a list of characters that are acceptable, or
353 * by a range (two characters separated by "-").
354 */
355 if (*pattern == '[') {
356 ++pattern;
357 for (;;) {
358 if ((*pattern == ']') || (*pattern == 0))
359 return(0);
360 if (*pattern == *string)
361 break;
362 if (pattern[1] == '-') {
363 c2 = pattern[2];
364 if (c2 == 0)
365 return(0);
366 if ((*pattern <= *string) &&
367 (c2 >= *string))
368 break;
369 if ((*pattern >= *string) &&
370 (c2 <= *string))
371 break;
372 pattern += 2;
373 }
374 ++pattern;
375 }
376 while ((*pattern != ']') && (*pattern != 0))
377 ++pattern;
378 goto thisCharOK;
379 }
380 /*
381 * If the next pattern character is '/', just strip off the
382 * '/' so we do exact matching on the character that follows.
383 */
384 if (*pattern == '\\') {
385 ++pattern;
386 if (*pattern == 0)
387 return(0);
388 }
389 /*
390 * There's no special character. Just make sure that the
391 * next characters of each string match.
392 */
393 if (*pattern != *string)
394 return(0);
395 thisCharOK: ++pattern;
396 ++string;
397 }
398 }
399
400
401 /*-
402 *-----------------------------------------------------------------------
403 * Str_SYSVMatch --
404 * Check word against pattern for a match (% is wild),
405 *
406 * Results:
407 * Returns the beginning position of a match or null. The number
408 * of characters matched is returned in len.
409 *
410 * Side Effects:
411 * None
412 *
413 *-----------------------------------------------------------------------
414 */
415 char *
416 Str_SYSVMatch(word, pattern, len)
417 char *word; /* Word to examine */
418 char *pattern; /* Pattern to examine against */
419 int *len; /* Number of characters to substitute */
420 {
421 char *p = pattern;
422 char *w = word;
423 char *m;
424
425 if (*p == '\0') {
426 /* Null pattern is the whole string */
427 *len = strlen(w);
428 return w;
429 }
430
431 if ((m = strchr(p, '%')) != NULL) {
432 /* check that the prefix matches */
433 for (; p != m && *w && *w == *p; w++, p++)
434 continue;
435
436 if (p != m)
437 return NULL; /* No match */
438
439 if (*++p == '\0') {
440 /* No more pattern, return the rest of the string */
441 *len = strlen(w);
442 return w;
443 }
444 }
445
446 m = w;
447
448 /* Find a matching tail */
449 do
450 if (strcmp(p, w) == 0) {
451 *len = w - m;
452 return m;
453 }
454 while (*w++ != '\0');
455
456 return NULL;
457 }
458
459
460 /*-
461 *-----------------------------------------------------------------------
462 * Str_SYSVSubst --
463 * Substitute '%' on the pattern with len characters from src.
464 * If the pattern does not contain a '%' prepend len characters
465 * from src.
466 *
467 * Results:
468 * None
469 *
470 * Side Effects:
471 * Places result on buf
472 *
473 *-----------------------------------------------------------------------
474 */
475 void
476 Str_SYSVSubst(buf, pat, src, len)
477 Buffer buf;
478 char *pat;
479 char *src;
480 int len;
481 {
482 char *m;
483
484 if ((m = strchr(pat, '%')) != NULL) {
485 /* Copy the prefix */
486 Buf_AddBytes(buf, m - pat, (Byte *) pat);
487 /* skip the % */
488 pat = m + 1;
489 }
490
491 /* Copy the pattern */
492 Buf_AddBytes(buf, len, (Byte *) src);
493
494 /* append the rest */
495 Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
496 }
497