xref: /netbsd-src/usr.bin/make/str.c (revision 76dfffe33547c37f8bdd446e3e4ab0f3c16cea4b)
1 /*	$NetBSD: str.c,v 1.13 1996/11/06 17:59:23 christos Exp $	*/
2 
3 /*-
4  * Copyright (c) 1988, 1989, 1990, 1993
5  *	The Regents of the University of California.  All rights reserved.
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 1.13 1996/11/06 17:59:23 christos 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) {
77 	if (argv[0])
78 	    free(argv[0]);
79 	free((Address) argv);
80     }
81     if (buffer)
82 	free(buffer);
83 }
84 
85 /*-
86  * str_concat --
87  *	concatenate the two strings, inserting a space or slash between them,
88  *	freeing them if requested.
89  *
90  * returns --
91  *	the resulting string in allocated space.
92  */
93 char *
94 str_concat(s1, s2, flags)
95 	char *s1, *s2;
96 	int flags;
97 {
98 	register int len1, len2;
99 	register char *result;
100 
101 	/* get the length of both strings */
102 	len1 = strlen(s1);
103 	len2 = strlen(s2);
104 
105 	/* allocate length plus separator plus EOS */
106 	result = emalloc((u_int)(len1 + len2 + 2));
107 
108 	/* copy first string into place */
109 	memcpy(result, s1, len1);
110 
111 	/* add separator character */
112 	if (flags & STR_ADDSPACE) {
113 		result[len1] = ' ';
114 		++len1;
115 	} else if (flags & STR_ADDSLASH) {
116 		result[len1] = '/';
117 		++len1;
118 	}
119 
120 	/* copy second string plus EOS into place */
121 	memcpy(result + len1, s2, len2 + 1);
122 
123 	/* free original strings */
124 	if (flags & STR_DOFREE) {
125 		(void)free(s1);
126 		(void)free(s2);
127 	}
128 	return(result);
129 }
130 
131 /*-
132  * brk_string --
133  *	Fracture a string into an array of words (as delineated by tabs or
134  *	spaces) taking quotation marks into account.  Leading tabs/spaces
135  *	are ignored.
136  *
137  * returns --
138  *	Pointer to the array of pointers to the words.  To make life easier,
139  *	the first word is always the value of the .MAKE variable.
140  */
141 char **
142 brk_string(str, store_argc, expand)
143 	register char *str;
144 	int *store_argc;
145 	Boolean expand;
146 {
147 	register int argc, ch;
148 	register char inquote, *p, *start, *t;
149 	int len;
150 
151 	/* skip leading space chars. */
152 	for (; *str == ' ' || *str == '\t'; ++str)
153 		continue;
154 
155 	/* allocate room for a copy of the string */
156 	if ((len = strlen(str) + 1) > curlen) {
157 		if (buffer)
158 		    free(buffer);
159 		buffer = emalloc(curlen = len);
160 	}
161 
162 	/*
163 	 * copy the string; at the same time, parse backslashes,
164 	 * quotes and build the argument list.
165 	 */
166 	argc = 1;
167 	inquote = '\0';
168 	for (p = str, start = t = buffer;; ++p) {
169 		switch(ch = *p) {
170 		case '"':
171 		case '\'':
172 			if (inquote)
173 				if (inquote == ch)
174 					inquote = '\0';
175 				else
176 					break;
177 			else {
178 				inquote = (char) ch;
179 				/* Don't miss "" or '' */
180 				if (start == NULL && p[1] == inquote) {
181 					start = t + 1;
182 					break;
183 				}
184 			}
185 			if (!expand) {
186 				if (!start)
187 					start = t;
188 				*t++ = ch;
189 			}
190 			continue;
191 		case ' ':
192 		case '\t':
193 		case '\n':
194 			if (inquote)
195 				break;
196 			if (!start)
197 				continue;
198 			/* FALLTHROUGH */
199 		case '\0':
200 			/*
201 			 * end of a token -- make sure there's enough argv
202 			 * space and save off a pointer.
203 			 */
204 			if (!start)
205 			    goto done;
206 
207 			*t++ = '\0';
208 			if (argc == argmax) {
209 				argmax *= 2;		/* ramp up fast */
210 				argv = (char **)erealloc(argv,
211 				    (argmax + 1) * sizeof(char *));
212 			}
213 			argv[argc++] = start;
214 			start = (char *)NULL;
215 			if (ch == '\n' || ch == '\0')
216 				goto done;
217 			continue;
218 		case '\\':
219 			if (!expand) {
220 				if (!start)
221 					start = t;
222 				*t++ = '\\';
223 				ch = *++p;
224 				break;
225 			}
226 
227 			switch (ch = *++p) {
228 			case '\0':
229 			case '\n':
230 				/* hmmm; fix it up as best we can */
231 				ch = '\\';
232 				--p;
233 				break;
234 			case 'b':
235 				ch = '\b';
236 				break;
237 			case 'f':
238 				ch = '\f';
239 				break;
240 			case 'n':
241 				ch = '\n';
242 				break;
243 			case 'r':
244 				ch = '\r';
245 				break;
246 			case 't':
247 				ch = '\t';
248 				break;
249 			}
250 			break;
251 		}
252 		if (!start)
253 			start = t;
254 		*t++ = (char) ch;
255 	}
256 done:	argv[argc] = (char *)NULL;
257 	*store_argc = argc;
258 	return(argv);
259 }
260 
261 /*
262  * Str_FindSubstring -- See if a string contains a particular substring.
263  *
264  * Results: If string contains substring, the return value is the location of
265  * the first matching instance of substring in string.  If string doesn't
266  * contain substring, the return value is NULL.  Matching is done on an exact
267  * character-for-character basis with no wildcards or special characters.
268  *
269  * Side effects: None.
270  */
271 char *
272 Str_FindSubstring(string, substring)
273 	register char *string;		/* String to search. */
274 	char *substring;		/* Substring to find in string */
275 {
276 	register char *a, *b;
277 
278 	/*
279 	 * First scan quickly through the two strings looking for a single-
280 	 * character match.  When it's found, then compare the rest of the
281 	 * substring.
282 	 */
283 
284 	for (b = substring; *string != 0; string += 1) {
285 		if (*string != *b)
286 			continue;
287 		a = string;
288 		for (;;) {
289 			if (*b == 0)
290 				return(string);
291 			if (*a++ != *b++)
292 				break;
293 		}
294 		b = substring;
295 	}
296 	return((char *) NULL);
297 }
298 
299 /*
300  * Str_Match --
301  *
302  * See if a particular string matches a particular pattern.
303  *
304  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
305  * matching operation permits the following special characters in the
306  * pattern: *?\[] (see the man page for details on what these mean).
307  *
308  * Side effects: None.
309  */
310 int
311 Str_Match(string, pattern)
312 	register char *string;		/* String */
313 	register char *pattern;		/* Pattern */
314 {
315 	char c2;
316 
317 	for (;;) {
318 		/*
319 		 * See if we're at the end of both the pattern and the
320 		 * string. If, we succeeded.  If we're at the end of the
321 		 * pattern but not at the end of the string, we failed.
322 		 */
323 		if (*pattern == 0)
324 			return(!*string);
325 		if (*string == 0 && *pattern != '*')
326 			return(0);
327 		/*
328 		 * Check for a "*" as the next pattern character.  It matches
329 		 * any substring.  We handle this by calling ourselves
330 		 * recursively for each postfix of string, until either we
331 		 * match or we reach the end of the string.
332 		 */
333 		if (*pattern == '*') {
334 			pattern += 1;
335 			if (*pattern == 0)
336 				return(1);
337 			while (*string != 0) {
338 				if (Str_Match(string, pattern))
339 					return(1);
340 				++string;
341 			}
342 			return(0);
343 		}
344 		/*
345 		 * Check for a "?" as the next pattern character.  It matches
346 		 * any single character.
347 		 */
348 		if (*pattern == '?')
349 			goto thisCharOK;
350 		/*
351 		 * Check for a "[" as the next pattern character.  It is
352 		 * followed by a list of characters that are acceptable, or
353 		 * by a range (two characters separated by "-").
354 		 */
355 		if (*pattern == '[') {
356 			++pattern;
357 			for (;;) {
358 				if ((*pattern == ']') || (*pattern == 0))
359 					return(0);
360 				if (*pattern == *string)
361 					break;
362 				if (pattern[1] == '-') {
363 					c2 = pattern[2];
364 					if (c2 == 0)
365 						return(0);
366 					if ((*pattern <= *string) &&
367 					    (c2 >= *string))
368 						break;
369 					if ((*pattern >= *string) &&
370 					    (c2 <= *string))
371 						break;
372 					pattern += 2;
373 				}
374 				++pattern;
375 			}
376 			while ((*pattern != ']') && (*pattern != 0))
377 				++pattern;
378 			goto thisCharOK;
379 		}
380 		/*
381 		 * If the next pattern character is '/', just strip off the
382 		 * '/' so we do exact matching on the character that follows.
383 		 */
384 		if (*pattern == '\\') {
385 			++pattern;
386 			if (*pattern == 0)
387 				return(0);
388 		}
389 		/*
390 		 * There's no special character.  Just make sure that the
391 		 * next characters of each string match.
392 		 */
393 		if (*pattern != *string)
394 			return(0);
395 thisCharOK:	++pattern;
396 		++string;
397 	}
398 }
399 
400 
401 /*-
402  *-----------------------------------------------------------------------
403  * Str_SYSVMatch --
404  *	Check word against pattern for a match (% is wild),
405  *
406  * Results:
407  *	Returns the beginning position of a match or null. The number
408  *	of characters matched is returned in len.
409  *
410  * Side Effects:
411  *	None
412  *
413  *-----------------------------------------------------------------------
414  */
415 char *
416 Str_SYSVMatch(word, pattern, len)
417     char	*word;		/* Word to examine */
418     char	*pattern;	/* Pattern to examine against */
419     int		*len;		/* Number of characters to substitute */
420 {
421     char *p = pattern;
422     char *w = word;
423     char *m;
424 
425     if (*p == '\0') {
426 	/* Null pattern is the whole string */
427 	*len = strlen(w);
428 	return w;
429     }
430 
431     if ((m = strchr(p, '%')) != NULL) {
432 	/* check that the prefix matches */
433 	for (; p != m && *w && *w == *p; w++, p++)
434 	     continue;
435 
436 	if (p != m)
437 	    return NULL;	/* No match */
438 
439 	if (*++p == '\0') {
440 	    /* No more pattern, return the rest of the string */
441 	    *len = strlen(w);
442 	    return w;
443 	}
444     }
445 
446     m = w;
447 
448     /* Find a matching tail */
449     do
450 	if (strcmp(p, w) == 0) {
451 	    *len = w - m;
452 	    return m;
453 	}
454     while (*w++ != '\0');
455 
456     return NULL;
457 }
458 
459 
460 /*-
461  *-----------------------------------------------------------------------
462  * Str_SYSVSubst --
463  *	Substitute '%' on the pattern with len characters from src.
464  *	If the pattern does not contain a '%' prepend len characters
465  *	from src.
466  *
467  * Results:
468  *	None
469  *
470  * Side Effects:
471  *	Places result on buf
472  *
473  *-----------------------------------------------------------------------
474  */
475 void
476 Str_SYSVSubst(buf, pat, src, len)
477     Buffer buf;
478     char *pat;
479     char *src;
480     int   len;
481 {
482     char *m;
483 
484     if ((m = strchr(pat, '%')) != NULL) {
485 	/* Copy the prefix */
486 	Buf_AddBytes(buf, m - pat, (Byte *) pat);
487 	/* skip the % */
488 	pat = m + 1;
489     }
490 
491     /* Copy the pattern */
492     Buf_AddBytes(buf, len, (Byte *) src);
493 
494     /* append the rest */
495     Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
496 }
497