1 /* $NetBSD: str.c,v 1.85 2021/05/30 21:16:54 rillig 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 #include "make.h" 72 73 /* "@(#)str.c 5.8 (Berkeley) 6/1/90" */ 74 MAKE_RCSID("$NetBSD: str.c,v 1.85 2021/05/30 21:16:54 rillig Exp $"); 75 76 /* Return the concatenation of s1 and s2, freshly allocated. */ 77 char * 78 str_concat2(const char *s1, const char *s2) 79 { 80 size_t len1 = strlen(s1); 81 size_t len2 = strlen(s2); 82 char *result = bmake_malloc(len1 + len2 + 1); 83 memcpy(result, s1, len1); 84 memcpy(result + len1, s2, len2 + 1); 85 return result; 86 } 87 88 /* Return the concatenation of s1, s2 and s3, freshly allocated. */ 89 char * 90 str_concat3(const char *s1, const char *s2, const char *s3) 91 { 92 size_t len1 = strlen(s1); 93 size_t len2 = strlen(s2); 94 size_t len3 = strlen(s3); 95 char *result = bmake_malloc(len1 + len2 + len3 + 1); 96 memcpy(result, s1, len1); 97 memcpy(result + len1, s2, len2); 98 memcpy(result + len1 + len2, s3, len3 + 1); 99 return result; 100 } 101 102 /* 103 * Fracture a string into an array of words (as delineated by tabs or spaces) 104 * taking quotation marks into account. 105 * 106 * If expand is true, quotes are removed and escape sequences such as \r, \t, 107 * etc... are expanded. In this case, return NULL on parse errors. 108 * 109 * Returns the fractured words, which must be freed later using Words_Free, 110 * unless the returned Words.words was NULL. 111 */ 112 SubstringWords 113 Substring_Words(const char *str, bool expand) 114 { 115 size_t str_len; 116 char *words_buf; 117 size_t words_cap; 118 Substring *words; 119 size_t words_len; 120 char inquote; 121 char *word_start; 122 char *word_end; 123 const char *str_p; 124 125 /* XXX: why only hspace, not whitespace? */ 126 cpp_skip_hspace(&str); /* skip leading space chars. */ 127 128 /* words_buf holds the words, separated by '\0'. */ 129 str_len = strlen(str); 130 words_buf = bmake_malloc(str_len + 1); 131 132 words_cap = str_len / 5 > 50 ? str_len / 5 : 50; 133 words = bmake_malloc((words_cap + 1) * sizeof(words[0])); 134 135 /* 136 * copy the string; at the same time, parse backslashes, 137 * quotes and build the word list. 138 */ 139 words_len = 0; 140 inquote = '\0'; 141 word_start = words_buf; 142 word_end = words_buf; 143 for (str_p = str;; str_p++) { 144 char ch = *str_p; 145 switch (ch) { 146 case '"': 147 case '\'': 148 if (inquote != '\0') { 149 if (inquote == ch) 150 inquote = '\0'; 151 else 152 break; 153 } else { 154 inquote = ch; 155 /* Don't miss "" or '' */ 156 if (word_start == NULL && str_p[1] == inquote) { 157 if (!expand) { 158 word_start = word_end; 159 *word_end++ = ch; 160 } else 161 word_start = word_end + 1; 162 str_p++; 163 inquote = '\0'; 164 break; 165 } 166 } 167 if (!expand) { 168 if (word_start == NULL) 169 word_start = word_end; 170 *word_end++ = ch; 171 } 172 continue; 173 case ' ': 174 case '\t': 175 case '\n': 176 if (inquote != '\0') 177 break; 178 if (word_start == NULL) 179 continue; 180 /* FALLTHROUGH */ 181 case '\0': 182 /* 183 * end of a token -- make sure there's enough words 184 * space and save off a pointer. 185 */ 186 if (word_start == NULL) 187 goto done; 188 189 *word_end++ = '\0'; 190 if (words_len == words_cap) { 191 size_t new_size; 192 words_cap *= 2; 193 new_size = (words_cap + 1) * sizeof(words[0]); 194 words = bmake_realloc(words, new_size); 195 } 196 words[words_len++] = 197 Substring_Init(word_start, word_end - 1); 198 word_start = NULL; 199 if (ch == '\n' || ch == '\0') { 200 if (expand && inquote != '\0') { 201 SubstringWords res; 202 203 free(words); 204 free(words_buf); 205 206 res.words = NULL; 207 res.len = 0; 208 res.freeIt = NULL; 209 return res; 210 } 211 goto done; 212 } 213 continue; 214 case '\\': 215 if (!expand) { 216 if (word_start == NULL) 217 word_start = word_end; 218 *word_end++ = '\\'; 219 /* catch '\' at end of line */ 220 if (str_p[1] == '\0') 221 continue; 222 ch = *++str_p; 223 break; 224 } 225 226 switch (ch = *++str_p) { 227 case '\0': 228 case '\n': 229 /* hmmm; fix it up as best we can */ 230 ch = '\\'; 231 str_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 (word_start == NULL) 252 word_start = word_end; 253 *word_end++ = ch; 254 } 255 done: 256 words[words_len] = Substring_Init(NULL, NULL); /* useful for argv */ 257 258 { 259 SubstringWords result; 260 261 result.words = words; 262 result.len = words_len; 263 result.freeIt = words_buf; 264 return result; 265 } 266 } 267 268 Words 269 Str_Words(const char *str, bool expand) 270 { 271 SubstringWords swords; 272 Words words; 273 size_t i; 274 275 swords = Substring_Words(str, expand); 276 if (swords.words == NULL) { 277 words.words = NULL; 278 words.len = 0; 279 words.freeIt = NULL; 280 return words; 281 } 282 283 words.words = bmake_malloc((swords.len + 1) * sizeof(words.words[0])); 284 words.len = swords.len; 285 words.freeIt = swords.freeIt; 286 for (i = 0; i < swords.len + 1; i++) 287 words.words[i] = UNCONST(swords.words[i].start); 288 free(swords.words); 289 return words; 290 } 291 292 /* 293 * Str_Match -- Test if a string matches a pattern like "*.[ch]". 294 * The following special characters are known *?\[] (as in fnmatch(3)). 295 * 296 * XXX: this function does not detect or report malformed patterns. 297 */ 298 bool 299 Str_Match(const char *str, const char *pat) 300 { 301 for (;;) { 302 /* 303 * See if we're at the end of both the pattern and the 304 * string. If so, we succeeded. If we're at the end of the 305 * pattern but not at the end of the string, we failed. 306 */ 307 if (*pat == '\0') 308 return *str == '\0'; 309 if (*str == '\0' && *pat != '*') 310 return false; 311 312 /* 313 * A '*' in the pattern matches any substring. We handle this 314 * by calling ourselves for each suffix of the string. 315 */ 316 if (*pat == '*') { 317 pat++; 318 while (*pat == '*') 319 pat++; 320 if (*pat == '\0') 321 return true; 322 while (*str != '\0') { 323 if (Str_Match(str, pat)) 324 return true; 325 str++; 326 } 327 return false; 328 } 329 330 /* A '?' in the pattern matches any single character. */ 331 if (*pat == '?') 332 goto thisCharOK; 333 334 /* 335 * A '[' in the pattern matches a character from a list. 336 * The '[' is followed by the list of acceptable characters, 337 * or by ranges (two characters separated by '-'). In these 338 * character lists, the backslash is an ordinary character. 339 */ 340 if (*pat == '[') { 341 bool neg = pat[1] == '^'; 342 pat += neg ? 2 : 1; 343 344 for (;;) { 345 if (*pat == ']' || *pat == '\0') { 346 if (neg) 347 break; 348 return false; 349 } 350 /* 351 * XXX: This naive comparison makes the 352 * control flow of the pattern parser 353 * dependent on the actual value of the 354 * string. This is unpredictable. It may be 355 * though that the code only looks wrong but 356 * actually all code paths result in the same 357 * behavior. This needs further tests. 358 */ 359 if (*pat == *str) 360 break; 361 if (pat[1] == '-') { 362 if (pat[2] == '\0') 363 return neg; 364 if (*pat <= *str && pat[2] >= *str) 365 break; 366 if (*pat >= *str && pat[2] <= *str) 367 break; 368 pat += 2; 369 } 370 pat++; 371 } 372 if (neg && *pat != ']' && *pat != '\0') 373 return false; 374 while (*pat != ']' && *pat != '\0') 375 pat++; 376 if (*pat == '\0') 377 pat--; 378 goto thisCharOK; 379 } 380 381 /* 382 * A backslash in the pattern matches the character following 383 * it exactly. 384 */ 385 if (*pat == '\\') { 386 pat++; 387 if (*pat == '\0') 388 return false; 389 } 390 391 if (*pat != *str) 392 return false; 393 394 thisCharOK: 395 pat++; 396 str++; 397 } 398 } 399