xref: /freebsd/contrib/bmake/str.c (revision f5147e312f43a9050468de539aeafa072caa1a60)
1 /*	$NetBSD: str.c,v 1.38 2017/04/21 22:15:44 sjg 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 1.38 2017/04/21 22:15:44 sjg 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 1.38 2017/04/21 22:15:44 sjg 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((unsigned 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;
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 	 * initial argmax based on len
160 	 */
161 	argmax = MAX((len / 5), 50);
162 	argv = bmake_malloc((argmax + 1) * sizeof(char *));
163 
164 	/*
165 	 * copy the string; at the same time, parse backslashes,
166 	 * quotes and build the argument list.
167 	 */
168 	argc = 0;
169 	inquote = '\0';
170 	for (p = str, start = t = *buffer;; ++p) {
171 		switch(ch = *p) {
172 		case '"':
173 		case '\'':
174 			if (inquote) {
175 				if (inquote == ch)
176 					inquote = '\0';
177 				else
178 					break;
179 			}
180 			else {
181 				inquote = (char) ch;
182 				/* Don't miss "" or '' */
183 				if (start == NULL && p[1] == inquote) {
184 					if (!expand) {
185 						start = t;
186 						*t++ = ch;
187 					} else
188 						start = t + 1;
189 					p++;
190 					inquote = '\0';
191 					break;
192 				}
193 			}
194 			if (!expand) {
195 				if (!start)
196 					start = t;
197 				*t++ = ch;
198 			}
199 			continue;
200 		case ' ':
201 		case '\t':
202 		case '\n':
203 			if (inquote)
204 				break;
205 			if (!start)
206 				continue;
207 			/* FALLTHROUGH */
208 		case '\0':
209 			/*
210 			 * end of a token -- make sure there's enough argv
211 			 * space and save off a pointer.
212 			 */
213 			if (!start)
214 			    goto done;
215 
216 			*t++ = '\0';
217 			if (argc == argmax) {
218 				argmax *= 2;		/* ramp up fast */
219 				argv = (char **)bmake_realloc(argv,
220 				    (argmax + 1) * sizeof(char *));
221 			}
222 			argv[argc++] = start;
223 			start = NULL;
224 			if (ch == '\n' || ch == '\0') {
225 				if (expand && inquote) {
226 					free(argv);
227 					free(*buffer);
228 					*buffer = NULL;
229 					return NULL;
230 				}
231 				goto done;
232 			}
233 			continue;
234 		case '\\':
235 			if (!expand) {
236 				if (!start)
237 					start = t;
238 				*t++ = '\\';
239 				if (*(p+1) == '\0') /* catch '\' at end of line */
240 					continue;
241 				ch = *++p;
242 				break;
243 			}
244 
245 			switch (ch = *++p) {
246 			case '\0':
247 			case '\n':
248 				/* hmmm; fix it up as best we can */
249 				ch = '\\';
250 				--p;
251 				break;
252 			case 'b':
253 				ch = '\b';
254 				break;
255 			case 'f':
256 				ch = '\f';
257 				break;
258 			case 'n':
259 				ch = '\n';
260 				break;
261 			case 'r':
262 				ch = '\r';
263 				break;
264 			case 't':
265 				ch = '\t';
266 				break;
267 			}
268 			break;
269 		}
270 		if (!start)
271 			start = t;
272 		*t++ = (char) ch;
273 	}
274 done:	argv[argc] = NULL;
275 	*store_argc = argc;
276 	return(argv);
277 }
278 
279 /*
280  * Str_FindSubstring -- See if a string contains a particular substring.
281  *
282  * Input:
283  *	string		String to search.
284  *	substring	Substring to find in string.
285  *
286  * Results: If string contains substring, the return value is the location of
287  * the first matching instance of substring in string.  If string doesn't
288  * contain substring, the return value is NULL.  Matching is done on an exact
289  * character-for-character basis with no wildcards or special characters.
290  *
291  * Side effects: None.
292  */
293 char *
294 Str_FindSubstring(const char *string, const char *substring)
295 {
296 	const char *a, *b;
297 
298 	/*
299 	 * First scan quickly through the two strings looking for a single-
300 	 * character match.  When it's found, then compare the rest of the
301 	 * substring.
302 	 */
303 
304 	for (b = substring; *string != 0; string += 1) {
305 		if (*string != *b)
306 			continue;
307 		a = string;
308 		for (;;) {
309 			if (*b == 0)
310 				return UNCONST(string);
311 			if (*a++ != *b++)
312 				break;
313 		}
314 		b = substring;
315 	}
316 	return NULL;
317 }
318 
319 /*
320  * Str_Match --
321  *
322  * See if a particular string matches a particular pattern.
323  *
324  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
325  * matching operation permits the following special characters in the
326  * pattern: *?\[] (see the man page for details on what these mean).
327  *
328  * XXX this function does not detect or report malformed patterns.
329  *
330  * Side effects: None.
331  */
332 int
333 Str_Match(const char *string, const char *pattern)
334 {
335 	char c2;
336 
337 	for (;;) {
338 		/*
339 		 * See if we're at the end of both the pattern and the
340 		 * string. If, we succeeded.  If we're at the end of the
341 		 * pattern but not at the end of the string, we failed.
342 		 */
343 		if (*pattern == 0)
344 			return(!*string);
345 		if (*string == 0 && *pattern != '*')
346 			return(0);
347 		/*
348 		 * Check for a "*" as the next pattern character.  It matches
349 		 * any substring.  We handle this by calling ourselves
350 		 * recursively for each postfix of string, until either we
351 		 * match or we reach the end of the string.
352 		 */
353 		if (*pattern == '*') {
354 			pattern += 1;
355 			if (*pattern == 0)
356 				return(1);
357 			while (*string != 0) {
358 				if (Str_Match(string, pattern))
359 					return(1);
360 				++string;
361 			}
362 			return(0);
363 		}
364 		/*
365 		 * Check for a "?" as the next pattern character.  It matches
366 		 * any single character.
367 		 */
368 		if (*pattern == '?')
369 			goto thisCharOK;
370 		/*
371 		 * Check for a "[" as the next pattern character.  It is
372 		 * followed by a list of characters that are acceptable, or
373 		 * by a range (two characters separated by "-").
374 		 */
375 		if (*pattern == '[') {
376 			int nomatch;
377 
378 			++pattern;
379 			if (*pattern == '^') {
380 				++pattern;
381 				nomatch = 1;
382 			} else
383 				nomatch = 0;
384 			for (;;) {
385 				if ((*pattern == ']') || (*pattern == 0)) {
386 					if (nomatch)
387 						break;
388 					return(0);
389 				}
390 				if (*pattern == *string)
391 					break;
392 				if (pattern[1] == '-') {
393 					c2 = pattern[2];
394 					if (c2 == 0)
395 						return(nomatch);
396 					if ((*pattern <= *string) &&
397 					    (c2 >= *string))
398 						break;
399 					if ((*pattern >= *string) &&
400 					    (c2 <= *string))
401 						break;
402 					pattern += 2;
403 				}
404 				++pattern;
405 			}
406 			if (nomatch && (*pattern != ']') && (*pattern != 0))
407 				return 0;
408 			while ((*pattern != ']') && (*pattern != 0))
409 				++pattern;
410 			goto thisCharOK;
411 		}
412 		/*
413 		 * If the next pattern character is '/', just strip off the
414 		 * '/' so we do exact matching on the character that follows.
415 		 */
416 		if (*pattern == '\\') {
417 			++pattern;
418 			if (*pattern == 0)
419 				return(0);
420 		}
421 		/*
422 		 * There's no special character.  Just make sure that the
423 		 * next characters of each string match.
424 		 */
425 		if (*pattern != *string)
426 			return(0);
427 thisCharOK:	++pattern;
428 		++string;
429 	}
430 }
431 
432 
433 /*-
434  *-----------------------------------------------------------------------
435  * Str_SYSVMatch --
436  *	Check word against pattern for a match (% is wild),
437  *
438  * Input:
439  *	word		Word to examine
440  *	pattern		Pattern to examine against
441  *	len		Number of characters to substitute
442  *
443  * Results:
444  *	Returns the beginning position of a match or null. The number
445  *	of characters matched is returned in len.
446  *
447  * Side Effects:
448  *	None
449  *
450  *-----------------------------------------------------------------------
451  */
452 char *
453 Str_SYSVMatch(const char *word, const char *pattern, int *len)
454 {
455     const char *p = pattern;
456     const char *w = word;
457     const char *m;
458 
459     if (*p == '\0') {
460 	/* Null pattern is the whole string */
461 	*len = strlen(w);
462 	return UNCONST(w);
463     }
464 
465     if ((m = strchr(p, '%')) != NULL) {
466 	/* check that the prefix matches */
467 	for (; p != m && *w && *w == *p; w++, p++)
468 	     continue;
469 
470 	if (p != m)
471 	    return NULL;	/* No match */
472 
473 	if (*++p == '\0') {
474 	    /* No more pattern, return the rest of the string */
475 	    *len = strlen(w);
476 	    return UNCONST(w);
477 	}
478     }
479 
480     m = w;
481 
482     /* Find a matching tail */
483     do
484 	if (strcmp(p, w) == 0) {
485 	    *len = w - m;
486 	    return UNCONST(m);
487 	}
488     while (*w++ != '\0');
489 
490     return NULL;
491 }
492 
493 
494 /*-
495  *-----------------------------------------------------------------------
496  * Str_SYSVSubst --
497  *	Substitute '%' on the pattern with len characters from src.
498  *	If the pattern does not contain a '%' prepend len characters
499  *	from src.
500  *
501  * Results:
502  *	None
503  *
504  * Side Effects:
505  *	Places result on buf
506  *
507  *-----------------------------------------------------------------------
508  */
509 void
510 Str_SYSVSubst(Buffer *buf, char *pat, char *src, int len)
511 {
512     char *m;
513 
514     if ((m = strchr(pat, '%')) != NULL) {
515 	/* Copy the prefix */
516 	Buf_AddBytes(buf, m - pat, pat);
517 	/* skip the % */
518 	pat = m + 1;
519     }
520 
521     /* Copy the pattern */
522     Buf_AddBytes(buf, len, src);
523 
524     /* append the rest */
525     Buf_AddBytes(buf, strlen(pat), pat);
526 }
527