xref: /netbsd-src/usr.bin/make/str.c (revision d37a996a49db27ca1b1e656cc3e261bfe3295a71)
1 /*	$NetBSD: str.c,v 1.19 2003/07/14 18:19:13 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 #ifdef MAKE_BOOTSTRAP
42 static char rcsid[] = "$NetBSD: str.c,v 1.19 2003/07/14 18:19:13 christos Exp $";
43 #else
44 #include <sys/cdefs.h>
45 #ifndef lint
46 #if 0
47 static char     sccsid[] = "@(#)str.c	5.8 (Berkeley) 6/1/90";
48 #else
49 __RCSID("$NetBSD: str.c,v 1.19 2003/07/14 18:19:13 christos Exp $");
50 #endif
51 #endif				/* not lint */
52 #endif
53 
54 #include "make.h"
55 
56 /*-
57  * str_concat --
58  *	concatenate the two strings, inserting a space or slash between them,
59  *	freeing them if requested.
60  *
61  * returns --
62  *	the resulting string in allocated space.
63  */
64 char *
65 str_concat(const char *s1, const char *s2, int flags)
66 {
67 	int len1, len2;
68 	char *result;
69 
70 	/* get the length of both strings */
71 	len1 = strlen(s1);
72 	len2 = strlen(s2);
73 
74 	/* allocate length plus separator plus EOS */
75 	result = emalloc((u_int)(len1 + len2 + 2));
76 
77 	/* copy first string into place */
78 	memcpy(result, s1, len1);
79 
80 	/* add separator character */
81 	if (flags & STR_ADDSPACE) {
82 		result[len1] = ' ';
83 		++len1;
84 	} else if (flags & STR_ADDSLASH) {
85 		result[len1] = '/';
86 		++len1;
87 	}
88 
89 	/* copy second string plus EOS into place */
90 	memcpy(result + len1, s2, len2 + 1);
91 
92 	return(result);
93 }
94 
95 /*-
96  * brk_string --
97  *	Fracture a string into an array of words (as delineated by tabs or
98  *	spaces) taking quotation marks into account.  Leading tabs/spaces
99  *	are ignored.
100  *
101  * returns --
102  *	Pointer to the array of pointers to the words.  To make life easier,
103  *	the first word is always the value of the .MAKE variable.
104  */
105 char **
106 brk_string(const char *str, int *store_argc, Boolean expand, char **buffer)
107 {
108 	int argc, ch;
109 	char inquote, *start, *t;
110 	const char *p;
111 	int len;
112 	int argmax = 50, curlen = 0;
113     	char **argv = (char **)emalloc((argmax + 1) * sizeof(char *));
114 
115 	/* skip leading space chars. */
116 	for (; *str == ' ' || *str == '\t'; ++str)
117 		continue;
118 
119 	/* allocate room for a copy of the string */
120 	if ((len = strlen(str) + 1) > curlen)
121 		*buffer = emalloc(curlen = len);
122 
123 	/*
124 	 * copy the string; at the same time, parse backslashes,
125 	 * quotes and build the argument list.
126 	 */
127 	argc = 0;
128 	inquote = '\0';
129 	for (p = str, start = t = *buffer;; ++p) {
130 		switch(ch = *p) {
131 		case '"':
132 		case '\'':
133 			if (inquote) {
134 				if (inquote == ch)
135 					inquote = '\0';
136 				else
137 					break;
138 			}
139 			else {
140 				inquote = (char) ch;
141 				/* Don't miss "" or '' */
142 				if (start == NULL && p[1] == inquote) {
143 					start = t + 1;
144 					break;
145 				}
146 			}
147 			if (!expand) {
148 				if (!start)
149 					start = t;
150 				*t++ = ch;
151 			}
152 			continue;
153 		case ' ':
154 		case '\t':
155 		case '\n':
156 			if (inquote)
157 				break;
158 			if (!start)
159 				continue;
160 			/* FALLTHROUGH */
161 		case '\0':
162 			/*
163 			 * end of a token -- make sure there's enough argv
164 			 * space and save off a pointer.
165 			 */
166 			if (!start)
167 			    goto done;
168 
169 			*t++ = '\0';
170 			if (argc == argmax) {
171 				argmax *= 2;		/* ramp up fast */
172 				argv = (char **)erealloc(argv,
173 				    (argmax + 1) * sizeof(char *));
174 			}
175 			argv[argc++] = start;
176 			start = (char *)NULL;
177 			if (ch == '\n' || ch == '\0')
178 				goto done;
179 			continue;
180 		case '\\':
181 			if (!expand) {
182 				if (!start)
183 					start = t;
184 				*t++ = '\\';
185 				ch = *++p;
186 				break;
187 			}
188 
189 			switch (ch = *++p) {
190 			case '\0':
191 			case '\n':
192 				/* hmmm; fix it up as best we can */
193 				ch = '\\';
194 				--p;
195 				break;
196 			case 'b':
197 				ch = '\b';
198 				break;
199 			case 'f':
200 				ch = '\f';
201 				break;
202 			case 'n':
203 				ch = '\n';
204 				break;
205 			case 'r':
206 				ch = '\r';
207 				break;
208 			case 't':
209 				ch = '\t';
210 				break;
211 			}
212 			break;
213 		}
214 		if (!start)
215 			start = t;
216 		*t++ = (char) ch;
217 	}
218 done:	argv[argc] = (char *)NULL;
219 	*store_argc = argc;
220 	return(argv);
221 }
222 
223 /*
224  * Str_FindSubstring -- See if a string contains a particular substring.
225  *
226  * Input:
227  *	string		String to search.
228  *	substring	Substring to find in string.
229  *
230  * Results: If string contains substring, the return value is the location of
231  * the first matching instance of substring in string.  If string doesn't
232  * contain substring, the return value is NULL.  Matching is done on an exact
233  * character-for-character basis with no wildcards or special characters.
234  *
235  * Side effects: None.
236  */
237 char *
238 Str_FindSubstring(const char *string, const char *substring)
239 {
240 	const char *a, *b;
241 
242 	/*
243 	 * First scan quickly through the two strings looking for a single-
244 	 * character match.  When it's found, then compare the rest of the
245 	 * substring.
246 	 */
247 
248 	for (b = substring; *string != 0; string += 1) {
249 		if (*string != *b)
250 			continue;
251 		a = string;
252 		for (;;) {
253 			if (*b == 0)
254 				return UNCONST(string);
255 			if (*a++ != *b++)
256 				break;
257 		}
258 		b = substring;
259 	}
260 	return NULL;
261 }
262 
263 /*
264  * Str_Match --
265  *
266  * See if a particular string matches a particular pattern.
267  *
268  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
269  * matching operation permits the following special characters in the
270  * pattern: *?\[] (see the man page for details on what these mean).
271  *
272  * Side effects: None.
273  */
274 int
275 Str_Match(const char *string, const char *pattern)
276 {
277 	char c2;
278 
279 	for (;;) {
280 		/*
281 		 * See if we're at the end of both the pattern and the
282 		 * string. If, we succeeded.  If we're at the end of the
283 		 * pattern but not at the end of the string, we failed.
284 		 */
285 		if (*pattern == 0)
286 			return(!*string);
287 		if (*string == 0 && *pattern != '*')
288 			return(0);
289 		/*
290 		 * Check for a "*" as the next pattern character.  It matches
291 		 * any substring.  We handle this by calling ourselves
292 		 * recursively for each postfix of string, until either we
293 		 * match or we reach the end of the string.
294 		 */
295 		if (*pattern == '*') {
296 			pattern += 1;
297 			if (*pattern == 0)
298 				return(1);
299 			while (*string != 0) {
300 				if (Str_Match(string, pattern))
301 					return(1);
302 				++string;
303 			}
304 			return(0);
305 		}
306 		/*
307 		 * Check for a "?" as the next pattern character.  It matches
308 		 * any single character.
309 		 */
310 		if (*pattern == '?')
311 			goto thisCharOK;
312 		/*
313 		 * Check for a "[" as the next pattern character.  It is
314 		 * followed by a list of characters that are acceptable, or
315 		 * by a range (two characters separated by "-").
316 		 */
317 		if (*pattern == '[') {
318 			++pattern;
319 			for (;;) {
320 				if ((*pattern == ']') || (*pattern == 0))
321 					return(0);
322 				if (*pattern == *string)
323 					break;
324 				if (pattern[1] == '-') {
325 					c2 = pattern[2];
326 					if (c2 == 0)
327 						return(0);
328 					if ((*pattern <= *string) &&
329 					    (c2 >= *string))
330 						break;
331 					if ((*pattern >= *string) &&
332 					    (c2 <= *string))
333 						break;
334 					pattern += 2;
335 				}
336 				++pattern;
337 			}
338 			while ((*pattern != ']') && (*pattern != 0))
339 				++pattern;
340 			goto thisCharOK;
341 		}
342 		/*
343 		 * If the next pattern character is '/', just strip off the
344 		 * '/' so we do exact matching on the character that follows.
345 		 */
346 		if (*pattern == '\\') {
347 			++pattern;
348 			if (*pattern == 0)
349 				return(0);
350 		}
351 		/*
352 		 * There's no special character.  Just make sure that the
353 		 * next characters of each string match.
354 		 */
355 		if (*pattern != *string)
356 			return(0);
357 thisCharOK:	++pattern;
358 		++string;
359 	}
360 }
361 
362 
363 /*-
364  *-----------------------------------------------------------------------
365  * Str_SYSVMatch --
366  *	Check word against pattern for a match (% is wild),
367  *
368  * Input:
369  *	word		Word to examine
370  *	pattern		Pattern to examine against
371  *	len		Number of characters to substitute
372  *
373  * Results:
374  *	Returns the beginning position of a match or null. The number
375  *	of characters matched is returned in len.
376  *
377  * Side Effects:
378  *	None
379  *
380  *-----------------------------------------------------------------------
381  */
382 char *
383 Str_SYSVMatch(const char *word, const char *pattern, int *len)
384 {
385     const char *p = pattern;
386     const char *w = word;
387     const char *m;
388 
389     if (*p == '\0') {
390 	/* Null pattern is the whole string */
391 	*len = strlen(w);
392 	return UNCONST(w);
393     }
394 
395     if ((m = strchr(p, '%')) != NULL) {
396 	/* check that the prefix matches */
397 	for (; p != m && *w && *w == *p; w++, p++)
398 	     continue;
399 
400 	if (p != m)
401 	    return NULL;	/* No match */
402 
403 	if (*++p == '\0') {
404 	    /* No more pattern, return the rest of the string */
405 	    *len = strlen(w);
406 	    return UNCONST(w);
407 	}
408     }
409 
410     m = w;
411 
412     /* Find a matching tail */
413     do
414 	if (strcmp(p, w) == 0) {
415 	    *len = w - m;
416 	    return UNCONST(m);
417 	}
418     while (*w++ != '\0');
419 
420     return NULL;
421 }
422 
423 
424 /*-
425  *-----------------------------------------------------------------------
426  * Str_SYSVSubst --
427  *	Substitute '%' on the pattern with len characters from src.
428  *	If the pattern does not contain a '%' prepend len characters
429  *	from src.
430  *
431  * Results:
432  *	None
433  *
434  * Side Effects:
435  *	Places result on buf
436  *
437  *-----------------------------------------------------------------------
438  */
439 void
440 Str_SYSVSubst(Buffer buf, char *pat, char *src, int len)
441 {
442     char *m;
443 
444     if ((m = strchr(pat, '%')) != NULL) {
445 	/* Copy the prefix */
446 	Buf_AddBytes(buf, m - pat, (Byte *) pat);
447 	/* skip the % */
448 	pat = m + 1;
449     }
450 
451     /* Copy the pattern */
452     Buf_AddBytes(buf, len, (Byte *) src);
453 
454     /* append the rest */
455     Buf_AddBytes(buf, strlen(pat), (Byte *) pat);
456 }
457