xref: /netbsd-src/usr.bin/make/str.c (revision 5f7096188587a2c7c95fa3c69b78e1ec9c7923d0)
1 /*-
2  * Copyright (c) 1988, 1989, 1990 The Regents of the University of California.
3  * Copyright (c) 1988, 1989 by Adam de Boor
4  * Copyright (c) 1989 by Berkeley Softworks
5  * 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. All advertising materials mentioning features or use of this software
19  *    must display the following acknowledgement:
20  *	This product includes software developed by the University of
21  *	California, Berkeley and its contributors.
22  * 4. Neither the name of the University nor the names of its contributors
23  *    may be used to endorse or promote products derived from this software
24  *    without specific prior written permission.
25  *
26  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
27  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
30  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
31  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
32  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
33  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
34  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
35  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
36  * SUCH DAMAGE.
37  */
38 
39 #ifndef lint
40 /*static char     sccsid[] = "from: @(#)str.c	5.8 (Berkeley) 6/1/90";*/
41 static char rcsid[] = "$Id: str.c,v 1.2 1993/08/01 18:11:43 mycroft Exp $";
42 #endif				/* not lint */
43 
44 #include "make.h"
45 
46 /*-
47  * str_concat --
48  *	concatenate the two strings, inserting a space or slash between them,
49  *	freeing them if requested.
50  *
51  * returns --
52  *	the resulting string in allocated space.
53  */
54 char *
55 str_concat(s1, s2, flags)
56 	char *s1, *s2;
57 	int flags;
58 {
59 	register int len1, len2;
60 	register char *result;
61 
62 	/* get the length of both strings */
63 	len1 = strlen(s1);
64 	len2 = strlen(s2);
65 
66 	/* allocate length plus separator plus EOS */
67 	result = emalloc((u_int)(len1 + len2 + 2));
68 
69 	/* copy first string into place */
70 	bcopy(s1, result, len1);
71 
72 	/* add separator character */
73 	if (flags & STR_ADDSPACE) {
74 		result[len1] = ' ';
75 		++len1;
76 	} else if (flags & STR_ADDSLASH) {
77 		result[len1] = '/';
78 		++len1;
79 	}
80 
81 	/* copy second string plus EOS into place */
82 	bcopy(s2, result + len1, len2 + 1);
83 
84 	/* free original strings */
85 	if (flags & STR_DOFREE) {
86 		(void)free(s1);
87 		(void)free(s2);
88 	}
89 	return(result);
90 }
91 
92 /*-
93  * brk_string --
94  *	Fracture a string into an array of words (as delineated by tabs or
95  *	spaces) taking quotation marks into account.  Leading tabs/spaces
96  *	are ignored.
97  *
98  * returns --
99  *	Pointer to the array of pointers to the words.  To make life easier,
100  *	the first word is always the value of the .MAKE variable.
101  */
102 char **
103 brk_string(str, store_argc)
104 	register char *str;
105 	int *store_argc;
106 {
107 	static int argmax, curlen;
108 	static char **argv, *buf;
109 	register int argc, ch;
110 	register char inquote, *p, *start, *t;
111 	int len;
112 
113 	/* save off pmake variable */
114 	if (!argv) {
115 		argv = (char **)emalloc((argmax = 50) * sizeof(char *));
116 		argv[0] = Var_Value(".MAKE", VAR_GLOBAL);
117 	}
118 
119 	/* skip leading space chars.
120 	for (; *str == ' ' || *str == '\t'; ++str);
121 
122 	/* allocate room for a copy of the string */
123 	if ((len = strlen(str) + 1) > curlen)
124 		buf = emalloc(curlen = len);
125 
126 	/*
127 	 * copy the string; at the same time, parse backslashes,
128 	 * quotes and build the argument list.
129 	 */
130 	argc = 1;
131 	inquote = '\0';
132 	for (p = str, start = t = buf;; ++p) {
133 		switch(ch = *p) {
134 		case '"':
135 		case '\'':
136 			if (inquote)
137 				if (inquote == ch)
138 					inquote = NULL;
139 				else
140 					break;
141 			else
142 				inquote = ch;
143 			continue;
144 		case ' ':
145 		case '\t':
146 			if (inquote)
147 				break;
148 			if (!start)
149 				continue;
150 			/* FALLTHROUGH */
151 		case '\n':
152 		case '\0':
153 			/*
154 			 * end of a token -- make sure there's enough argv
155 			 * space and save off a pointer.
156 			 */
157 			*t++ = '\0';
158 			if (argc == argmax) {
159 				argmax *= 2;		/* ramp up fast */
160 				if (!(argv = (char **)realloc(argv,
161 				    argmax * sizeof(char *))))
162 				enomem();
163 			}
164 			argv[argc++] = start;
165 			start = (char *)NULL;
166 			if (ch == '\n' || ch == '\0')
167 				goto done;
168 			continue;
169 		case '\\':
170 			switch (ch = *++p) {
171 			case '\0':
172 			case '\n':
173 				/* hmmm; fix it up as best we can */
174 				ch = '\\';
175 				--p;
176 				break;
177 			case 'b':
178 				ch = '\b';
179 				break;
180 			case 'f':
181 				ch = '\f';
182 				break;
183 			case 'n':
184 				ch = '\n';
185 				break;
186 			case 'r':
187 				ch = '\r';
188 				break;
189 			case 't':
190 				ch = '\t';
191 				break;
192 			}
193 			break;
194 		}
195 		if (!start)
196 			start = t;
197 		*t++ = ch;
198 	}
199 done:	argv[argc] = (char *)NULL;
200 	*store_argc = argc;
201 	return(argv);
202 }
203 
204 /*
205  * Str_FindSubstring -- See if a string contains a particular substring.
206  *
207  * Results: If string contains substring, the return value is the location of
208  * the first matching instance of substring in string.  If string doesn't
209  * contain substring, the return value is NULL.  Matching is done on an exact
210  * character-for-character basis with no wildcards or special characters.
211  *
212  * Side effects: None.
213  */
214 char *
215 Str_FindSubstring(string, substring)
216 	register char *string;		/* String to search. */
217 	char *substring;		/* Substring to find in string */
218 {
219 	register char *a, *b;
220 
221 	/*
222 	 * First scan quickly through the two strings looking for a single-
223 	 * character match.  When it's found, then compare the rest of the
224 	 * substring.
225 	 */
226 
227 	for (b = substring; *string != 0; string += 1) {
228 		if (*string != *b)
229 			continue;
230 		a = string;
231 		for (;;) {
232 			if (*b == 0)
233 				return(string);
234 			if (*a++ != *b++)
235 				break;
236 		}
237 		b = substring;
238 	}
239 	return((char *) NULL);
240 }
241 
242 /*
243  * Str_Match --
244  *
245  * See if a particular string matches a particular pattern.
246  *
247  * Results: Non-zero is returned if string matches pattern, 0 otherwise. The
248  * matching operation permits the following special characters in the
249  * pattern: *?\[] (see the man page for details on what these mean).
250  *
251  * Side effects: None.
252  */
253 Str_Match(string, pattern)
254 	register char *string;		/* String */
255 	register char *pattern;		/* Pattern */
256 {
257 	char c2;
258 
259 	for (;;) {
260 		/*
261 		 * See if we're at the end of both the pattern and the
262 		 * string. If, we succeeded.  If we're at the end of the
263 		 * pattern but not at the end of the string, we failed.
264 		 */
265 		if (*pattern == 0)
266 			return(!*string);
267 		if (*string == 0 && *pattern != '*')
268 			return(0);
269 		/*
270 		 * Check for a "*" as the next pattern character.  It matches
271 		 * any substring.  We handle this by calling ourselves
272 		 * recursively for each postfix of string, until either we
273 		 * match or we reach the end of the string.
274 		 */
275 		if (*pattern == '*') {
276 			pattern += 1;
277 			if (*pattern == 0)
278 				return(1);
279 			while (*string != 0) {
280 				if (Str_Match(string, pattern))
281 					return(1);
282 				++string;
283 			}
284 			return(0);
285 		}
286 		/*
287 		 * Check for a "?" as the next pattern character.  It matches
288 		 * any single character.
289 		 */
290 		if (*pattern == '?')
291 			goto thisCharOK;
292 		/*
293 		 * Check for a "[" as the next pattern character.  It is
294 		 * followed by a list of characters that are acceptable, or
295 		 * by a range (two characters separated by "-").
296 		 */
297 		if (*pattern == '[') {
298 			++pattern;
299 			for (;;) {
300 				if ((*pattern == ']') || (*pattern == 0))
301 					return(0);
302 				if (*pattern == *string)
303 					break;
304 				if (pattern[1] == '-') {
305 					c2 = pattern[2];
306 					if (c2 == 0)
307 						return(0);
308 					if ((*pattern <= *string) &&
309 					    (c2 >= *string))
310 						break;
311 					if ((*pattern >= *string) &&
312 					    (c2 <= *string))
313 						break;
314 					pattern += 2;
315 				}
316 				++pattern;
317 			}
318 			while ((*pattern != ']') && (*pattern != 0))
319 				++pattern;
320 			goto thisCharOK;
321 		}
322 		/*
323 		 * If the next pattern character is '/', just strip off the
324 		 * '/' so we do exact matching on the character that follows.
325 		 */
326 		if (*pattern == '\\') {
327 			++pattern;
328 			if (*pattern == 0)
329 				return(0);
330 		}
331 		/*
332 		 * There's no special character.  Just make sure that the
333 		 * next characters of each string match.
334 		 */
335 		if (*pattern != *string)
336 			return(0);
337 thisCharOK:	++pattern;
338 		++string;
339 	}
340 }
341