xref: /netbsd-src/usr.bin/make/str.c (revision 81b108b45f75f89f1e3ffad9fb6f074e771c0935)
1 /*	$NetBSD: str.c,v 1.12 1996/03/29 02:17:34 jtc 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 1.12 1996/03/29 02:17:34 jtc 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) + 1) * 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 				argv = (char **)erealloc(argv,
210 				    (argmax + 1) * sizeof(char *));
211 			}
212 			argv[argc++] = start;
213 			start = (char *)NULL;
214 			if (ch == '\n' || ch == '\0')
215 				goto done;
216 			continue;
217 		case '\\':
218 			if (!expand) {
219 				if (!start)
220 					start = t;
221 				*t++ = '\\';
222 				ch = *++p;
223 				break;
224 			}
225 
226 			switch (ch = *++p) {
227 			case '\0':
228 			case '\n':
229 				/* hmmm; fix it up as best we can */
230 				ch = '\\';
231 				--p;
232 				break;
233 			case 'b':
234 				ch = '\b';
235 				break;
236 			case 'f':
237 				ch = '\f';
238 				break;
239 			case 'n':
240 				ch = '\n';
241 				break;
242 			case 'r':
243 				ch = '\r';
244 				break;
245 			case 't':
246 				ch = '\t';
247 				break;
248 			}
249 			break;
250 		}
251 		if (!start)
252 			start = t;
253 		*t++ = (char) ch;
254 	}
255 done:	argv[argc] = (char *)NULL;
256 	*store_argc = argc;
257 	return(argv);
258 }
259 
260 /*
261  * Str_FindSubstring -- See if a string contains a particular substring.
262  *
263  * Results: If string contains substring, the return value is the location of
264  * the first matching instance of substring in string.  If string doesn't
265  * contain substring, the return value is NULL.  Matching is done on an exact
266  * character-for-character basis with no wildcards or special characters.
267  *
268  * Side effects: None.
269  */
270 char *
271 Str_FindSubstring(string, substring)
272 	register char *string;		/* String to search. */
273 	char *substring;		/* Substring to find in string */
274 {
275 	register char *a, *b;
276 
277 	/*
278 	 * First scan quickly through the two strings looking for a single-
279 	 * character match.  When it's found, then compare the rest of the
280 	 * substring.
281 	 */
282 
283 	for (b = substring; *string != 0; string += 1) {
284 		if (*string != *b)
285 			continue;
286 		a = string;
287 		for (;;) {
288 			if (*b == 0)
289 				return(string);
290 			if (*a++ != *b++)
291 				break;
292 		}
293 		b = substring;
294 	}
295 	return((char *) NULL);
296 }
297 
298 /*
299  * Str_Match --
300  *
301  * See if a particular string matches a particular pattern.
302  *
303  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
304  * matching operation permits the following special characters in the
305  * pattern: *?\[] (see the man page for details on what these mean).
306  *
307  * Side effects: None.
308  */
309 int
310 Str_Match(string, pattern)
311 	register char *string;		/* String */
312 	register char *pattern;		/* Pattern */
313 {
314 	char c2;
315 
316 	for (;;) {
317 		/*
318 		 * See if we're at the end of both the pattern and the
319 		 * string. If, we succeeded.  If we're at the end of the
320 		 * pattern but not at the end of the string, we failed.
321 		 */
322 		if (*pattern == 0)
323 			return(!*string);
324 		if (*string == 0 && *pattern != '*')
325 			return(0);
326 		/*
327 		 * Check for a "*" as the next pattern character.  It matches
328 		 * any substring.  We handle this by calling ourselves
329 		 * recursively for each postfix of string, until either we
330 		 * match or we reach the end of the string.
331 		 */
332 		if (*pattern == '*') {
333 			pattern += 1;
334 			if (*pattern == 0)
335 				return(1);
336 			while (*string != 0) {
337 				if (Str_Match(string, pattern))
338 					return(1);
339 				++string;
340 			}
341 			return(0);
342 		}
343 		/*
344 		 * Check for a "?" as the next pattern character.  It matches
345 		 * any single character.
346 		 */
347 		if (*pattern == '?')
348 			goto thisCharOK;
349 		/*
350 		 * Check for a "[" as the next pattern character.  It is
351 		 * followed by a list of characters that are acceptable, or
352 		 * by a range (two characters separated by "-").
353 		 */
354 		if (*pattern == '[') {
355 			++pattern;
356 			for (;;) {
357 				if ((*pattern == ']') || (*pattern == 0))
358 					return(0);
359 				if (*pattern == *string)
360 					break;
361 				if (pattern[1] == '-') {
362 					c2 = pattern[2];
363 					if (c2 == 0)
364 						return(0);
365 					if ((*pattern <= *string) &&
366 					    (c2 >= *string))
367 						break;
368 					if ((*pattern >= *string) &&
369 					    (c2 <= *string))
370 						break;
371 					pattern += 2;
372 				}
373 				++pattern;
374 			}
375 			while ((*pattern != ']') && (*pattern != 0))
376 				++pattern;
377 			goto thisCharOK;
378 		}
379 		/*
380 		 * If the next pattern character is '/', just strip off the
381 		 * '/' so we do exact matching on the character that follows.
382 		 */
383 		if (*pattern == '\\') {
384 			++pattern;
385 			if (*pattern == 0)
386 				return(0);
387 		}
388 		/*
389 		 * There's no special character.  Just make sure that the
390 		 * next characters of each string match.
391 		 */
392 		if (*pattern != *string)
393 			return(0);
394 thisCharOK:	++pattern;
395 		++string;
396 	}
397 }
398 
399 
400 /*-
401  *-----------------------------------------------------------------------
402  * Str_SYSVMatch --
403  *	Check word against pattern for a match (% is wild),
404  *
405  * Results:
406  *	Returns the beginning position of a match or null. The number
407  *	of characters matched is returned in len.
408  *
409  * Side Effects:
410  *	None
411  *
412  *-----------------------------------------------------------------------
413  */
414 char *
415 Str_SYSVMatch(word, pattern, len)
416     char	*word;		/* Word to examine */
417     char	*pattern;	/* Pattern to examine against */
418     int		*len;		/* Number of characters to substitute */
419 {
420     char *p = pattern;
421     char *w = word;
422     char *m;
423 
424     if (*p == '\0') {
425 	/* Null pattern is the whole string */
426 	*len = strlen(w);
427 	return w;
428     }
429 
430     if ((m = strchr(p, '%')) != NULL) {
431 	/* check that the prefix matches */
432 	for (; p != m && *w && *w == *p; w++, p++)
433 	     continue;
434 
435 	if (p != m)
436 	    return NULL;	/* No match */
437 
438 	if (*++p == '\0') {
439 	    /* No more pattern, return the rest of the string */
440 	    *len = strlen(w);
441 	    return w;
442 	}
443     }
444 
445     m = w;
446 
447     /* Find a matching tail */
448     do
449 	if (strcmp(p, w) == 0) {
450 	    *len = w - m;
451 	    return m;
452 	}
453     while (*w++ != '\0');
454 
455     return NULL;
456 }
457 
458 
459 /*-
460  *-----------------------------------------------------------------------
461  * Str_SYSVSubst --
462  *	Substitute '%' on the pattern with len characters from src.
463  *	If the pattern does not contain a '%' prepend len characters
464  *	from src.
465  *
466  * Results:
467  *	None
468  *
469  * Side Effects:
470  *	Places result on buf
471  *
472  *-----------------------------------------------------------------------
473  */
474 void
475 Str_SYSVSubst(buf, pat, src, len)
476     Buffer buf;
477     char *pat;
478     char *src;
479     int   len;
480 {
481     char *m;
482 
483     if ((m = strchr(pat, '%')) != NULL) {
484 	/* Copy the prefix */
485 	Buf_AddBytes(buf, m - pat, (Byte *) pat);
486 	/* skip the % */
487 	pat = m + 1;
488     }
489 
490     /* Copy the pattern */
491     Buf_AddBytes(buf, len, (Byte *) src);
492 
493     /* append the rest */
494     Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
495 }
496