xref: /freebsd/contrib/bmake/str.c (revision 724b4bfdf1306e4f2c451b6d146fe0fe0353b2c8)
1 /*	$NetBSD: str.c,v 1.34 2012/03/03 23:16:47 dholland Exp $	*/
2 
3 /*-
4  * Copyright (c) 1988, 1989, 1990, 1993
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * This code is derived from software contributed to Berkeley by
8  * Adam de Boor.
9  *
10  * Redistribution and use in source and binary forms, with or without
11  * modification, are permitted provided that the following conditions
12  * are met:
13  * 1. Redistributions of source code must retain the above copyright
14  *    notice, this list of conditions and the following disclaimer.
15  * 2. Redistributions in binary form must reproduce the above copyright
16  *    notice, this list of conditions and the following disclaimer in the
17  *    documentation and/or other materials provided with the distribution.
18  * 3. Neither the name of the University nor the names of its contributors
19  *    may be used to endorse or promote products derived from this software
20  *    without specific prior written permission.
21  *
22  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
23  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
24  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
25  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
26  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
27  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
28  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
29  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
31  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32  * SUCH DAMAGE.
33  */
34 
35 /*-
36  * Copyright (c) 1989 by Berkeley Softworks
37  * All rights reserved.
38  *
39  * This code is derived from software contributed to Berkeley by
40  * Adam de Boor.
41  *
42  * Redistribution and use in source and binary forms, with or without
43  * modification, are permitted provided that the following conditions
44  * are met:
45  * 1. Redistributions of source code must retain the above copyright
46  *    notice, this list of conditions and the following disclaimer.
47  * 2. Redistributions in binary form must reproduce the above copyright
48  *    notice, this list of conditions and the following disclaimer in the
49  *    documentation and/or other materials provided with the distribution.
50  * 3. All advertising materials mentioning features or use of this software
51  *    must display the following acknowledgement:
52  *	This product includes software developed by the University of
53  *	California, Berkeley and its contributors.
54  * 4. Neither the name of the University nor the names of its contributors
55  *    may be used to endorse or promote products derived from this software
56  *    without specific prior written permission.
57  *
58  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
59  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
60  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
61  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
62  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
63  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
64  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
65  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
66  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
67  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
68  * SUCH DAMAGE.
69  */
70 
71 #ifndef MAKE_NATIVE
72 static char rcsid[] = "$NetBSD: str.c,v 1.34 2012/03/03 23:16:47 dholland Exp $";
73 #else
74 #include <sys/cdefs.h>
75 #ifndef lint
76 #if 0
77 static char     sccsid[] = "@(#)str.c	5.8 (Berkeley) 6/1/90";
78 #else
79 __RCSID("$NetBSD: str.c,v 1.34 2012/03/03 23:16:47 dholland Exp $");
80 #endif
81 #endif				/* not lint */
82 #endif
83 
84 #include "make.h"
85 
86 /*-
87  * str_concat --
88  *	concatenate the two strings, inserting a space or slash between them,
89  *	freeing them if requested.
90  *
91  * returns --
92  *	the resulting string in allocated space.
93  */
94 char *
95 str_concat(const char *s1, const char *s2, int flags)
96 {
97 	int len1, len2;
98 	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 = bmake_malloc((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 	return(result);
123 }
124 
125 /*-
126  * brk_string --
127  *	Fracture a string into an array of words (as delineated by tabs or
128  *	spaces) taking quotation marks into account.  Leading tabs/spaces
129  *	are ignored.
130  *
131  * If expand is TRUE, quotes are removed and escape sequences
132  *  such as \r, \t, etc... are expanded.
133  *
134  * returns --
135  *	Pointer to the array of pointers to the words.
136  *      Memory containing the actual words in *buffer.
137  *		Both of these must be free'd by the caller.
138  *      Number of words in *store_argc.
139  */
140 char **
141 brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
142 {
143 	int argc, ch;
144 	char inquote, *start, *t;
145 	const char *p;
146 	int len;
147 	int argmax = 50, curlen = 0;
148     	char **argv = bmake_malloc((argmax + 1) * sizeof(char *));
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 		*buffer = bmake_malloc(curlen = len);
157 
158 	/*
159 	 * copy the string; at the same time, parse backslashes,
160 	 * quotes and build the argument list.
161 	 */
162 	argc = 0;
163 	inquote = '\0';
164 	for (p = str, start = t = *buffer;; ++p) {
165 		switch(ch = *p) {
166 		case '"':
167 		case '\'':
168 			if (inquote) {
169 				if (inquote == ch)
170 					inquote = '\0';
171 				else
172 					break;
173 			}
174 			else {
175 				inquote = (char) ch;
176 				/* Don't miss "" or '' */
177 				if (start == NULL && p[1] == inquote) {
178 					if (!expand) {
179 						start = t;
180 						*t++ = ch;
181 					} else
182 						start = t + 1;
183 					p++;
184 					inquote = '\0';
185 					break;
186 				}
187 			}
188 			if (!expand) {
189 				if (!start)
190 					start = t;
191 				*t++ = ch;
192 			}
193 			continue;
194 		case ' ':
195 		case '\t':
196 		case '\n':
197 			if (inquote)
198 				break;
199 			if (!start)
200 				continue;
201 			/* FALLTHROUGH */
202 		case '\0':
203 			/*
204 			 * end of a token -- make sure there's enough argv
205 			 * space and save off a pointer.
206 			 */
207 			if (!start)
208 			    goto done;
209 
210 			*t++ = '\0';
211 			if (argc == argmax) {
212 				argmax *= 2;		/* ramp up fast */
213 				argv = (char **)bmake_realloc(argv,
214 				    (argmax + 1) * sizeof(char *));
215 			}
216 			argv[argc++] = start;
217 			start = NULL;
218 			if (ch == '\n' || ch == '\0') {
219 				if (expand && inquote) {
220 					free(argv);
221 					free(*buffer);
222 					*buffer = NULL;
223 					return NULL;
224 				}
225 				goto done;
226 			}
227 			continue;
228 		case '\\':
229 			if (!expand) {
230 				if (!start)
231 					start = t;
232 				*t++ = '\\';
233 				if (*(p+1) == '\0') /* catch '\' at end of line */
234 					continue;
235 				ch = *++p;
236 				break;
237 			}
238 
239 			switch (ch = *++p) {
240 			case '\0':
241 			case '\n':
242 				/* hmmm; fix it up as best we can */
243 				ch = '\\';
244 				--p;
245 				break;
246 			case 'b':
247 				ch = '\b';
248 				break;
249 			case 'f':
250 				ch = '\f';
251 				break;
252 			case 'n':
253 				ch = '\n';
254 				break;
255 			case 'r':
256 				ch = '\r';
257 				break;
258 			case 't':
259 				ch = '\t';
260 				break;
261 			}
262 			break;
263 		}
264 		if (!start)
265 			start = t;
266 		*t++ = (char) ch;
267 	}
268 done:	argv[argc] = NULL;
269 	*store_argc = argc;
270 	return(argv);
271 }
272 
273 /*
274  * Str_FindSubstring -- See if a string contains a particular substring.
275  *
276  * Input:
277  *	string		String to search.
278  *	substring	Substring to find in string.
279  *
280  * Results: If string contains substring, the return value is the location of
281  * the first matching instance of substring in string.  If string doesn't
282  * contain substring, the return value is NULL.  Matching is done on an exact
283  * character-for-character basis with no wildcards or special characters.
284  *
285  * Side effects: None.
286  */
287 char *
288 Str_FindSubstring(const char *string, const char *substring)
289 {
290 	const char *a, *b;
291 
292 	/*
293 	 * First scan quickly through the two strings looking for a single-
294 	 * character match.  When it's found, then compare the rest of the
295 	 * substring.
296 	 */
297 
298 	for (b = substring; *string != 0; string += 1) {
299 		if (*string != *b)
300 			continue;
301 		a = string;
302 		for (;;) {
303 			if (*b == 0)
304 				return UNCONST(string);
305 			if (*a++ != *b++)
306 				break;
307 		}
308 		b = substring;
309 	}
310 	return NULL;
311 }
312 
313 /*
314  * Str_Match --
315  *
316  * See if a particular string matches a particular pattern.
317  *
318  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
319  * matching operation permits the following special characters in the
320  * pattern: *?\[] (see the man page for details on what these mean).
321  *
322  * XXX this function does not detect or report malformed patterns.
323  *
324  * Side effects: None.
325  */
326 int
327 Str_Match(const char *string, const char *pattern)
328 {
329 	char c2;
330 
331 	for (;;) {
332 		/*
333 		 * See if we're at the end of both the pattern and the
334 		 * string. If, we succeeded.  If we're at the end of the
335 		 * pattern but not at the end of the string, we failed.
336 		 */
337 		if (*pattern == 0)
338 			return(!*string);
339 		if (*string == 0 && *pattern != '*')
340 			return(0);
341 		/*
342 		 * Check for a "*" as the next pattern character.  It matches
343 		 * any substring.  We handle this by calling ourselves
344 		 * recursively for each postfix of string, until either we
345 		 * match or we reach the end of the string.
346 		 */
347 		if (*pattern == '*') {
348 			pattern += 1;
349 			if (*pattern == 0)
350 				return(1);
351 			while (*string != 0) {
352 				if (Str_Match(string, pattern))
353 					return(1);
354 				++string;
355 			}
356 			return(0);
357 		}
358 		/*
359 		 * Check for a "?" as the next pattern character.  It matches
360 		 * any single character.
361 		 */
362 		if (*pattern == '?')
363 			goto thisCharOK;
364 		/*
365 		 * Check for a "[" as the next pattern character.  It is
366 		 * followed by a list of characters that are acceptable, or
367 		 * by a range (two characters separated by "-").
368 		 */
369 		if (*pattern == '[') {
370 			++pattern;
371 			for (;;) {
372 				if ((*pattern == ']') || (*pattern == 0))
373 					return(0);
374 				if (*pattern == *string)
375 					break;
376 				if (pattern[1] == '-') {
377 					c2 = pattern[2];
378 					if (c2 == 0)
379 						return(0);
380 					if ((*pattern <= *string) &&
381 					    (c2 >= *string))
382 						break;
383 					if ((*pattern >= *string) &&
384 					    (c2 <= *string))
385 						break;
386 					pattern += 2;
387 				}
388 				++pattern;
389 			}
390 			while ((*pattern != ']') && (*pattern != 0))
391 				++pattern;
392 			goto thisCharOK;
393 		}
394 		/*
395 		 * If the next pattern character is '/', just strip off the
396 		 * '/' so we do exact matching on the character that follows.
397 		 */
398 		if (*pattern == '\\') {
399 			++pattern;
400 			if (*pattern == 0)
401 				return(0);
402 		}
403 		/*
404 		 * There's no special character.  Just make sure that the
405 		 * next characters of each string match.
406 		 */
407 		if (*pattern != *string)
408 			return(0);
409 thisCharOK:	++pattern;
410 		++string;
411 	}
412 }
413 
414 
415 /*-
416  *-----------------------------------------------------------------------
417  * Str_SYSVMatch --
418  *	Check word against pattern for a match (% is wild),
419  *
420  * Input:
421  *	word		Word to examine
422  *	pattern		Pattern to examine against
423  *	len		Number of characters to substitute
424  *
425  * Results:
426  *	Returns the beginning position of a match or null. The number
427  *	of characters matched is returned in len.
428  *
429  * Side Effects:
430  *	None
431  *
432  *-----------------------------------------------------------------------
433  */
434 char *
435 Str_SYSVMatch(const char *word, const char *pattern, int *len)
436 {
437     const char *p = pattern;
438     const char *w = word;
439     const char *m;
440 
441     if (*p == '\0') {
442 	/* Null pattern is the whole string */
443 	*len = strlen(w);
444 	return UNCONST(w);
445     }
446 
447     if ((m = strchr(p, '%')) != NULL) {
448 	/* check that the prefix matches */
449 	for (; p != m && *w && *w == *p; w++, p++)
450 	     continue;
451 
452 	if (p != m)
453 	    return NULL;	/* No match */
454 
455 	if (*++p == '\0') {
456 	    /* No more pattern, return the rest of the string */
457 	    *len = strlen(w);
458 	    return UNCONST(w);
459 	}
460     }
461 
462     m = w;
463 
464     /* Find a matching tail */
465     do
466 	if (strcmp(p, w) == 0) {
467 	    *len = w - m;
468 	    return UNCONST(m);
469 	}
470     while (*w++ != '\0');
471 
472     return NULL;
473 }
474 
475 
476 /*-
477  *-----------------------------------------------------------------------
478  * Str_SYSVSubst --
479  *	Substitute '%' on the pattern with len characters from src.
480  *	If the pattern does not contain a '%' prepend len characters
481  *	from src.
482  *
483  * Results:
484  *	None
485  *
486  * Side Effects:
487  *	Places result on buf
488  *
489  *-----------------------------------------------------------------------
490  */
491 void
492 Str_SYSVSubst(Buffer *buf, char *pat, char *src, int len)
493 {
494     char *m;
495 
496     if ((m = strchr(pat, '%')) != NULL) {
497 	/* Copy the prefix */
498 	Buf_AddBytes(buf, m - pat, pat);
499 	/* skip the % */
500 	pat = m + 1;
501     }
502 
503     /* Copy the pattern */
504     Buf_AddBytes(buf, len, src);
505 
506     /* append the rest */
507     Buf_AddBytes(buf, strlen(pat), pat);
508 }
509