xref: /netbsd-src/usr.bin/make/str.c (revision da5f4674a3fc214be3572d358b66af40ab9401e7)
1 /*	$NetBSD: str.c,v 1.20 2003/08/07 11:14:57 agc 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 #ifdef MAKE_BOOTSTRAP
72 static char rcsid[] = "$NetBSD: str.c,v 1.20 2003/08/07 11:14:57 agc 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 1.20 2003/08/07 11:14:57 agc 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.  To make life easier,
133  *	the first word is always the value of the .MAKE variable.
134  */
135 char **
136 brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
137 {
138 	int argc, ch;
139 	char inquote, *start, *t;
140 	const char *p;
141 	int len;
142 	int argmax = 50, curlen = 0;
143     	char **argv = (char **)emalloc((argmax + 1) * sizeof(char *));
144 
145 	/* skip leading space chars. */
146 	for (; *str == ' ' || *str == '\t'; ++str)
147 		continue;
148 
149 	/* allocate room for a copy of the string */
150 	if ((len = strlen(str) + 1) > curlen)
151 		*buffer = emalloc(curlen = len);
152 
153 	/*
154 	 * copy the string; at the same time, parse backslashes,
155 	 * quotes and build the argument list.
156 	 */
157 	argc = 0;
158 	inquote = '\0';
159 	for (p = str, start = t = *buffer;; ++p) {
160 		switch(ch = *p) {
161 		case '"':
162 		case '\'':
163 			if (inquote) {
164 				if (inquote == ch)
165 					inquote = '\0';
166 				else
167 					break;
168 			}
169 			else {
170 				inquote = (char) ch;
171 				/* Don't miss "" or '' */
172 				if (start == NULL && p[1] == inquote) {
173 					start = t + 1;
174 					break;
175 				}
176 			}
177 			if (!expand) {
178 				if (!start)
179 					start = t;
180 				*t++ = ch;
181 			}
182 			continue;
183 		case ' ':
184 		case '\t':
185 		case '\n':
186 			if (inquote)
187 				break;
188 			if (!start)
189 				continue;
190 			/* FALLTHROUGH */
191 		case '\0':
192 			/*
193 			 * end of a token -- make sure there's enough argv
194 			 * space and save off a pointer.
195 			 */
196 			if (!start)
197 			    goto done;
198 
199 			*t++ = '\0';
200 			if (argc == argmax) {
201 				argmax *= 2;		/* ramp up fast */
202 				argv = (char **)erealloc(argv,
203 				    (argmax + 1) * sizeof(char *));
204 			}
205 			argv[argc++] = start;
206 			start = (char *)NULL;
207 			if (ch == '\n' || ch == '\0')
208 				goto done;
209 			continue;
210 		case '\\':
211 			if (!expand) {
212 				if (!start)
213 					start = t;
214 				*t++ = '\\';
215 				ch = *++p;
216 				break;
217 			}
218 
219 			switch (ch = *++p) {
220 			case '\0':
221 			case '\n':
222 				/* hmmm; fix it up as best we can */
223 				ch = '\\';
224 				--p;
225 				break;
226 			case 'b':
227 				ch = '\b';
228 				break;
229 			case 'f':
230 				ch = '\f';
231 				break;
232 			case 'n':
233 				ch = '\n';
234 				break;
235 			case 'r':
236 				ch = '\r';
237 				break;
238 			case 't':
239 				ch = '\t';
240 				break;
241 			}
242 			break;
243 		}
244 		if (!start)
245 			start = t;
246 		*t++ = (char) ch;
247 	}
248 done:	argv[argc] = (char *)NULL;
249 	*store_argc = argc;
250 	return(argv);
251 }
252 
253 /*
254  * Str_FindSubstring -- See if a string contains a particular substring.
255  *
256  * Input:
257  *	string		String to search.
258  *	substring	Substring to find in string.
259  *
260  * Results: If string contains substring, the return value is the location of
261  * the first matching instance of substring in string.  If string doesn't
262  * contain substring, the return value is NULL.  Matching is done on an exact
263  * character-for-character basis with no wildcards or special characters.
264  *
265  * Side effects: None.
266  */
267 char *
268 Str_FindSubstring(const char *string, const char *substring)
269 {
270 	const char *a, *b;
271 
272 	/*
273 	 * First scan quickly through the two strings looking for a single-
274 	 * character match.  When it's found, then compare the rest of the
275 	 * substring.
276 	 */
277 
278 	for (b = substring; *string != 0; string += 1) {
279 		if (*string != *b)
280 			continue;
281 		a = string;
282 		for (;;) {
283 			if (*b == 0)
284 				return UNCONST(string);
285 			if (*a++ != *b++)
286 				break;
287 		}
288 		b = substring;
289 	}
290 	return NULL;
291 }
292 
293 /*
294  * Str_Match --
295  *
296  * See if a particular string matches a particular pattern.
297  *
298  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
299  * matching operation permits the following special characters in the
300  * pattern: *?\[] (see the man page for details on what these mean).
301  *
302  * Side effects: None.
303  */
304 int
305 Str_Match(const char *string, const char *pattern)
306 {
307 	char c2;
308 
309 	for (;;) {
310 		/*
311 		 * See if we're at the end of both the pattern and the
312 		 * string. If, we succeeded.  If we're at the end of the
313 		 * pattern but not at the end of the string, we failed.
314 		 */
315 		if (*pattern == 0)
316 			return(!*string);
317 		if (*string == 0 && *pattern != '*')
318 			return(0);
319 		/*
320 		 * Check for a "*" as the next pattern character.  It matches
321 		 * any substring.  We handle this by calling ourselves
322 		 * recursively for each postfix of string, until either we
323 		 * match or we reach the end of the string.
324 		 */
325 		if (*pattern == '*') {
326 			pattern += 1;
327 			if (*pattern == 0)
328 				return(1);
329 			while (*string != 0) {
330 				if (Str_Match(string, pattern))
331 					return(1);
332 				++string;
333 			}
334 			return(0);
335 		}
336 		/*
337 		 * Check for a "?" as the next pattern character.  It matches
338 		 * any single character.
339 		 */
340 		if (*pattern == '?')
341 			goto thisCharOK;
342 		/*
343 		 * Check for a "[" as the next pattern character.  It is
344 		 * followed by a list of characters that are acceptable, or
345 		 * by a range (two characters separated by "-").
346 		 */
347 		if (*pattern == '[') {
348 			++pattern;
349 			for (;;) {
350 				if ((*pattern == ']') || (*pattern == 0))
351 					return(0);
352 				if (*pattern == *string)
353 					break;
354 				if (pattern[1] == '-') {
355 					c2 = pattern[2];
356 					if (c2 == 0)
357 						return(0);
358 					if ((*pattern <= *string) &&
359 					    (c2 >= *string))
360 						break;
361 					if ((*pattern >= *string) &&
362 					    (c2 <= *string))
363 						break;
364 					pattern += 2;
365 				}
366 				++pattern;
367 			}
368 			while ((*pattern != ']') && (*pattern != 0))
369 				++pattern;
370 			goto thisCharOK;
371 		}
372 		/*
373 		 * If the next pattern character is '/', just strip off the
374 		 * '/' so we do exact matching on the character that follows.
375 		 */
376 		if (*pattern == '\\') {
377 			++pattern;
378 			if (*pattern == 0)
379 				return(0);
380 		}
381 		/*
382 		 * There's no special character.  Just make sure that the
383 		 * next characters of each string match.
384 		 */
385 		if (*pattern != *string)
386 			return(0);
387 thisCharOK:	++pattern;
388 		++string;
389 	}
390 }
391 
392 
393 /*-
394  *-----------------------------------------------------------------------
395  * Str_SYSVMatch --
396  *	Check word against pattern for a match (% is wild),
397  *
398  * Input:
399  *	word		Word to examine
400  *	pattern		Pattern to examine against
401  *	len		Number of characters to substitute
402  *
403  * Results:
404  *	Returns the beginning position of a match or null. The number
405  *	of characters matched is returned in len.
406  *
407  * Side Effects:
408  *	None
409  *
410  *-----------------------------------------------------------------------
411  */
412 char *
413 Str_SYSVMatch(const char *word, const char *pattern, int *len)
414 {
415     const char *p = pattern;
416     const char *w = word;
417     const char *m;
418 
419     if (*p == '\0') {
420 	/* Null pattern is the whole string */
421 	*len = strlen(w);
422 	return UNCONST(w);
423     }
424 
425     if ((m = strchr(p, '%')) != NULL) {
426 	/* check that the prefix matches */
427 	for (; p != m && *w && *w == *p; w++, p++)
428 	     continue;
429 
430 	if (p != m)
431 	    return NULL;	/* No match */
432 
433 	if (*++p == '\0') {
434 	    /* No more pattern, return the rest of the string */
435 	    *len = strlen(w);
436 	    return UNCONST(w);
437 	}
438     }
439 
440     m = w;
441 
442     /* Find a matching tail */
443     do
444 	if (strcmp(p, w) == 0) {
445 	    *len = w - m;
446 	    return UNCONST(m);
447 	}
448     while (*w++ != '\0');
449 
450     return NULL;
451 }
452 
453 
454 /*-
455  *-----------------------------------------------------------------------
456  * Str_SYSVSubst --
457  *	Substitute '%' on the pattern with len characters from src.
458  *	If the pattern does not contain a '%' prepend len characters
459  *	from src.
460  *
461  * Results:
462  *	None
463  *
464  * Side Effects:
465  *	Places result on buf
466  *
467  *-----------------------------------------------------------------------
468  */
469 void
470 Str_SYSVSubst(Buffer buf, char *pat, char *src, int len)
471 {
472     char *m;
473 
474     if ((m = strchr(pat, '%')) != NULL) {
475 	/* Copy the prefix */
476 	Buf_AddBytes(buf, m - pat, (Byte *) pat);
477 	/* skip the % */
478 	pat = m + 1;
479     }
480 
481     /* Copy the pattern */
482     Buf_AddBytes(buf, len, (Byte *) src);
483 
484     /* append the rest */
485     Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
486 }
487