xref: /netbsd-src/usr.bin/make/str.c (revision 23c8222edbfb0f0932d88a8351d3a0cf817dfb9e)
1 /*	$NetBSD: str.c,v 1.22 2004/05/07 00:04:40 ross 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 1.22 2004/05/07 00:04:40 ross 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 1.22 2004/05/07 00:04:40 ross 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 = 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 	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  * returns --
132  *	Pointer to the array of pointers to the words.
133  */
134 char **
135 brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
136 {
137 	int argc, ch;
138 	char inquote, *start, *t;
139 	const char *p;
140 	int len;
141 	int argmax = 50, curlen = 0;
142     	char **argv = (char **)emalloc((argmax + 1) * sizeof(char *));
143 
144 	/* skip leading space chars. */
145 	for (; *str == ' ' || *str == '\t'; ++str)
146 		continue;
147 
148 	/* allocate room for a copy of the string */
149 	if ((len = strlen(str) + 1) > curlen)
150 		*buffer = emalloc(curlen = len);
151 
152 	/*
153 	 * copy the string; at the same time, parse backslashes,
154 	 * quotes and build the argument list.
155 	 */
156 	argc = 0;
157 	inquote = '\0';
158 	for (p = str, start = t = *buffer;; ++p) {
159 		switch(ch = *p) {
160 		case '"':
161 		case '\'':
162 			if (inquote) {
163 				if (inquote == ch)
164 					inquote = '\0';
165 				else
166 					break;
167 			}
168 			else {
169 				inquote = (char) ch;
170 				/* Don't miss "" or '' */
171 				if (start == NULL && p[1] == inquote) {
172 					start = t + 1;
173 					break;
174 				}
175 			}
176 			if (!expand) {
177 				if (!start)
178 					start = t;
179 				*t++ = ch;
180 			}
181 			continue;
182 		case ' ':
183 		case '\t':
184 		case '\n':
185 			if (inquote)
186 				break;
187 			if (!start)
188 				continue;
189 			/* FALLTHROUGH */
190 		case '\0':
191 			/*
192 			 * end of a token -- make sure there's enough argv
193 			 * space and save off a pointer.
194 			 */
195 			if (!start)
196 			    goto done;
197 
198 			*t++ = '\0';
199 			if (argc == argmax) {
200 				argmax *= 2;		/* ramp up fast */
201 				argv = (char **)erealloc(argv,
202 				    (argmax + 1) * sizeof(char *));
203 			}
204 			argv[argc++] = start;
205 			start = (char *)NULL;
206 			if (ch == '\n' || ch == '\0')
207 				goto done;
208 			continue;
209 		case '\\':
210 			if (!expand) {
211 				if (!start)
212 					start = t;
213 				*t++ = '\\';
214 				ch = *++p;
215 				break;
216 			}
217 
218 			switch (ch = *++p) {
219 			case '\0':
220 			case '\n':
221 				/* hmmm; fix it up as best we can */
222 				ch = '\\';
223 				--p;
224 				break;
225 			case 'b':
226 				ch = '\b';
227 				break;
228 			case 'f':
229 				ch = '\f';
230 				break;
231 			case 'n':
232 				ch = '\n';
233 				break;
234 			case 'r':
235 				ch = '\r';
236 				break;
237 			case 't':
238 				ch = '\t';
239 				break;
240 			}
241 			break;
242 		}
243 		if (!start)
244 			start = t;
245 		*t++ = (char) ch;
246 	}
247 done:	argv[argc] = (char *)NULL;
248 	*store_argc = argc;
249 	return(argv);
250 }
251 
252 /*
253  * Str_FindSubstring -- See if a string contains a particular substring.
254  *
255  * Input:
256  *	string		String to search.
257  *	substring	Substring to find in string.
258  *
259  * Results: If string contains substring, the return value is the location of
260  * the first matching instance of substring in string.  If string doesn't
261  * contain substring, the return value is NULL.  Matching is done on an exact
262  * character-for-character basis with no wildcards or special characters.
263  *
264  * Side effects: None.
265  */
266 char *
267 Str_FindSubstring(const char *string, const char *substring)
268 {
269 	const char *a, *b;
270 
271 	/*
272 	 * First scan quickly through the two strings looking for a single-
273 	 * character match.  When it's found, then compare the rest of the
274 	 * substring.
275 	 */
276 
277 	for (b = substring; *string != 0; string += 1) {
278 		if (*string != *b)
279 			continue;
280 		a = string;
281 		for (;;) {
282 			if (*b == 0)
283 				return UNCONST(string);
284 			if (*a++ != *b++)
285 				break;
286 		}
287 		b = substring;
288 	}
289 	return NULL;
290 }
291 
292 /*
293  * Str_Match --
294  *
295  * See if a particular string matches a particular pattern.
296  *
297  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
298  * matching operation permits the following special characters in the
299  * pattern: *?\[] (see the man page for details on what these mean).
300  *
301  * Side effects: None.
302  */
303 int
304 Str_Match(const char *string, const char *pattern)
305 {
306 	char c2;
307 
308 	for (;;) {
309 		/*
310 		 * See if we're at the end of both the pattern and the
311 		 * string. If, we succeeded.  If we're at the end of the
312 		 * pattern but not at the end of the string, we failed.
313 		 */
314 		if (*pattern == 0)
315 			return(!*string);
316 		if (*string == 0 && *pattern != '*')
317 			return(0);
318 		/*
319 		 * Check for a "*" as the next pattern character.  It matches
320 		 * any substring.  We handle this by calling ourselves
321 		 * recursively for each postfix of string, until either we
322 		 * match or we reach the end of the string.
323 		 */
324 		if (*pattern == '*') {
325 			pattern += 1;
326 			if (*pattern == 0)
327 				return(1);
328 			while (*string != 0) {
329 				if (Str_Match(string, pattern))
330 					return(1);
331 				++string;
332 			}
333 			return(0);
334 		}
335 		/*
336 		 * Check for a "?" as the next pattern character.  It matches
337 		 * any single character.
338 		 */
339 		if (*pattern == '?')
340 			goto thisCharOK;
341 		/*
342 		 * Check for a "[" as the next pattern character.  It is
343 		 * followed by a list of characters that are acceptable, or
344 		 * by a range (two characters separated by "-").
345 		 */
346 		if (*pattern == '[') {
347 			++pattern;
348 			for (;;) {
349 				if ((*pattern == ']') || (*pattern == 0))
350 					return(0);
351 				if (*pattern == *string)
352 					break;
353 				if (pattern[1] == '-') {
354 					c2 = pattern[2];
355 					if (c2 == 0)
356 						return(0);
357 					if ((*pattern <= *string) &&
358 					    (c2 >= *string))
359 						break;
360 					if ((*pattern >= *string) &&
361 					    (c2 <= *string))
362 						break;
363 					pattern += 2;
364 				}
365 				++pattern;
366 			}
367 			while ((*pattern != ']') && (*pattern != 0))
368 				++pattern;
369 			goto thisCharOK;
370 		}
371 		/*
372 		 * If the next pattern character is '/', just strip off the
373 		 * '/' so we do exact matching on the character that follows.
374 		 */
375 		if (*pattern == '\\') {
376 			++pattern;
377 			if (*pattern == 0)
378 				return(0);
379 		}
380 		/*
381 		 * There's no special character.  Just make sure that the
382 		 * next characters of each string match.
383 		 */
384 		if (*pattern != *string)
385 			return(0);
386 thisCharOK:	++pattern;
387 		++string;
388 	}
389 }
390 
391 
392 /*-
393  *-----------------------------------------------------------------------
394  * Str_SYSVMatch --
395  *	Check word against pattern for a match (% is wild),
396  *
397  * Input:
398  *	word		Word to examine
399  *	pattern		Pattern to examine against
400  *	len		Number of characters to substitute
401  *
402  * Results:
403  *	Returns the beginning position of a match or null. The number
404  *	of characters matched is returned in len.
405  *
406  * Side Effects:
407  *	None
408  *
409  *-----------------------------------------------------------------------
410  */
411 char *
412 Str_SYSVMatch(const char *word, const char *pattern, int *len)
413 {
414     const char *p = pattern;
415     const char *w = word;
416     const char *m;
417 
418     if (*p == '\0') {
419 	/* Null pattern is the whole string */
420 	*len = strlen(w);
421 	return UNCONST(w);
422     }
423 
424     if ((m = strchr(p, '%')) != NULL) {
425 	/* check that the prefix matches */
426 	for (; p != m && *w && *w == *p; w++, p++)
427 	     continue;
428 
429 	if (p != m)
430 	    return NULL;	/* No match */
431 
432 	if (*++p == '\0') {
433 	    /* No more pattern, return the rest of the string */
434 	    *len = strlen(w);
435 	    return UNCONST(w);
436 	}
437     }
438 
439     m = w;
440 
441     /* Find a matching tail */
442     do
443 	if (strcmp(p, w) == 0) {
444 	    *len = w - m;
445 	    return UNCONST(m);
446 	}
447     while (*w++ != '\0');
448 
449     return NULL;
450 }
451 
452 
453 /*-
454  *-----------------------------------------------------------------------
455  * Str_SYSVSubst --
456  *	Substitute '%' on the pattern with len characters from src.
457  *	If the pattern does not contain a '%' prepend len characters
458  *	from src.
459  *
460  * Results:
461  *	None
462  *
463  * Side Effects:
464  *	Places result on buf
465  *
466  *-----------------------------------------------------------------------
467  */
468 void
469 Str_SYSVSubst(Buffer buf, char *pat, char *src, int len)
470 {
471     char *m;
472 
473     if ((m = strchr(pat, '%')) != NULL) {
474 	/* Copy the prefix */
475 	Buf_AddBytes(buf, m - pat, (Byte *) pat);
476 	/* skip the % */
477 	pat = m + 1;
478     }
479 
480     /* Copy the pattern */
481     Buf_AddBytes(buf, len, (Byte *) src);
482 
483     /* append the rest */
484     Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
485 }
486