1 /* 2 * Copyright (c) 1980, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by the University of 16 * California, Berkeley and its contributors. 17 * 4. Neither the name of the University nor the names of its contributors 18 * may be used to endorse or promote products derived from this software 19 * without specific prior written permission. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 */ 33 34 #ifndef lint 35 static const char copyright[] = 36 "@(#) Copyright (c) 1980, 1993\n\ 37 The Regents of the University of California. All rights reserved.\n"; 38 #endif /* not lint */ 39 40 #ifndef lint 41 #if 0 42 static char sccsid[] = "@(#)xstr.c 8.1 (Berkeley) 6/9/93"; 43 #endif 44 static const char rcsid[] = 45 "$FreeBSD$"; 46 #endif /* not lint */ 47 48 #include <sys/types.h> 49 #include <ctype.h> 50 #include <err.h> 51 #include <stdio.h> 52 #include <stdlib.h> 53 #include <signal.h> 54 #include <string.h> 55 #include <unistd.h> 56 #include "pathnames.h" 57 58 /* 59 * xstr - extract and hash strings in a C program 60 * 61 * Bill Joy UCB 62 * November, 1978 63 */ 64 65 #define ignore(a) ((void) a) 66 67 off_t tellpt; 68 off_t hashit(); 69 void onintr(); 70 char *savestr(); 71 off_t yankstr(); 72 73 off_t mesgpt; 74 char *strings = "strings"; 75 76 int cflg; 77 int vflg; 78 int readstd; 79 80 static void usage __P((void)); 81 int istail __P((char *, char *)); 82 char lastchr __P((char *)); 83 void xsdotc __P((void)); 84 void prstr __P((char *)); 85 void found __P((int, off_t, char *)); 86 void flushsh __P((void)); 87 int xgetc __P((FILE *)); 88 int fgetNUL __P((char *, int, FILE *)); 89 void inithash __P((void)); 90 int octdigit __P((char)); 91 void process __P((char *)); 92 93 int 94 main(argc, argv) 95 int argc; 96 char *argv[]; 97 { 98 int c; 99 100 while ((c = getopt(argc, argv, "-cv")) != -1) 101 switch (c) { 102 case '-': 103 readstd++; 104 break; 105 case 'c': 106 cflg++; 107 break; 108 case 'v': 109 vflg++; 110 break; 111 default: 112 usage(); 113 } 114 argc -= optind; 115 argv += optind; 116 117 if (signal(SIGINT, SIG_IGN) == SIG_DFL) 118 signal(SIGINT, onintr); 119 if (cflg || (argc == 0 && !readstd)) 120 inithash(); 121 else 122 strings = mktemp(strdup(_PATH_TMP)); 123 while (readstd || argc > 0) { 124 if (freopen("x.c", "w", stdout) == NULL) 125 err(1, "x.c"); 126 if (!readstd && freopen(argv[0], "r", stdin) == NULL) 127 err(2, "%s", argv[0]); 128 process("x.c"); 129 if (readstd == 0) 130 argc--, argv++; 131 else 132 readstd = 0; 133 }; 134 flushsh(); 135 if (cflg == 0) 136 xsdotc(); 137 if (strings[0] == '/') 138 ignore(unlink(strings)); 139 exit(0); 140 } 141 142 static void 143 usage() 144 { 145 fprintf(stderr, "usage: xstr [-v] [-c] [-] [name ...]\n"); 146 exit (1); 147 } 148 149 char linebuf[BUFSIZ]; 150 151 void 152 process(name) 153 char *name; 154 { 155 char *cp; 156 register int c; 157 register int incomm = 0; 158 int ret; 159 160 printf("extern char\txstr[];\n"); 161 for (;;) { 162 if (fgets(linebuf, sizeof linebuf, stdin) == NULL) { 163 if (ferror(stdin)) 164 err(3, "%s", name); 165 break; 166 } 167 if (linebuf[0] == '#') { 168 if (linebuf[1] == ' ' && isdigit(linebuf[2])) 169 printf("#line%s", &linebuf[1]); 170 else 171 printf("%s", linebuf); 172 continue; 173 } 174 for (cp = linebuf; (c = *cp++);) switch (c) { 175 176 case '"': 177 if (incomm) 178 goto def; 179 if ((ret = (int) yankstr(&cp)) == -1) 180 goto out; 181 printf("(&xstr[%d])", ret); 182 break; 183 184 case '\'': 185 if (incomm) 186 goto def; 187 putchar(c); 188 if (*cp) 189 putchar(*cp++); 190 break; 191 192 case '/': 193 if (incomm || *cp != '*') 194 goto def; 195 incomm = 1; 196 cp++; 197 printf("/*"); 198 continue; 199 200 case '*': 201 if (incomm && *cp == '/') { 202 incomm = 0; 203 cp++; 204 printf("*/"); 205 continue; 206 } 207 goto def; 208 209 def: 210 default: 211 putchar(c); 212 break; 213 } 214 } 215 out: 216 if (ferror(stdout)) 217 warn("x.c"), onintr(); 218 } 219 220 off_t 221 yankstr(cpp) 222 register char **cpp; 223 { 224 register char *cp = *cpp; 225 register int c, ch; 226 char dbuf[BUFSIZ]; 227 register char *dp = dbuf; 228 register char *tp; 229 230 while ((c = *cp++)) { 231 switch (c) { 232 233 case '"': 234 cp++; 235 goto out; 236 237 case '\\': 238 c = *cp++; 239 if (c == 0) 240 break; 241 if (c == '\n') { 242 if (fgets(linebuf, sizeof linebuf, stdin) 243 == NULL) { 244 if (ferror(stdin)) 245 err(3, "x.c"); 246 return(-1); 247 } 248 cp = linebuf; 249 continue; 250 } 251 for (tp = "b\bt\tr\rn\nf\f\\\\\"\""; (ch = *tp++); tp++) 252 if (c == ch) { 253 c = *tp; 254 goto gotc; 255 } 256 if (!octdigit(c)) { 257 *dp++ = '\\'; 258 break; 259 } 260 c -= '0'; 261 if (!octdigit(*cp)) 262 break; 263 c <<= 3, c += *cp++ - '0'; 264 if (!octdigit(*cp)) 265 break; 266 c <<= 3, c += *cp++ - '0'; 267 break; 268 } 269 gotc: 270 *dp++ = c; 271 } 272 out: 273 *cpp = --cp; 274 *dp = 0; 275 return (hashit(dbuf, 1)); 276 } 277 278 int 279 octdigit(c) 280 char c; 281 { 282 return (isdigit(c) && c != '8' && c != '9'); 283 } 284 285 void 286 inithash() 287 { 288 char buf[BUFSIZ]; 289 register FILE *mesgread = fopen(strings, "r"); 290 291 if (mesgread == NULL) 292 return; 293 for (;;) { 294 mesgpt = tellpt; 295 if (fgetNUL(buf, sizeof buf, mesgread) == 0) 296 break; 297 ignore(hashit(buf, 0)); 298 } 299 ignore(fclose(mesgread)); 300 } 301 302 int 303 fgetNUL(obuf, rmdr, file) 304 char *obuf; 305 register int rmdr; 306 FILE *file; 307 { 308 register c; 309 register char *buf = obuf; 310 311 while (--rmdr > 0 && (c = xgetc(file)) != 0 && c != EOF) 312 *buf++ = c; 313 *buf++ = 0; 314 return ((feof(file) || ferror(file)) ? 0 : 1); 315 } 316 317 int 318 xgetc(file) 319 FILE *file; 320 { 321 322 tellpt++; 323 return (getc(file)); 324 } 325 326 #define BUCKETS 128 327 328 struct hash { 329 off_t hpt; 330 char *hstr; 331 struct hash *hnext; 332 short hnew; 333 } bucket[BUCKETS]; 334 335 off_t 336 hashit(str, new) 337 char *str; 338 int new; 339 { 340 int i; 341 register struct hash *hp, *hp0; 342 343 hp = hp0 = &bucket[lastchr(str) & 0177]; 344 while (hp->hnext) { 345 hp = hp->hnext; 346 i = istail(str, hp->hstr); 347 if (i >= 0) 348 return (hp->hpt + i); 349 } 350 if ((hp = (struct hash *) calloc(1, sizeof (*hp))) == NULL) 351 errx(8, "calloc"); 352 hp->hpt = mesgpt; 353 if (!(hp->hstr = strdup(str))) 354 err(1, NULL); 355 mesgpt += strlen(hp->hstr) + 1; 356 hp->hnext = hp0->hnext; 357 hp->hnew = new; 358 hp0->hnext = hp; 359 return (hp->hpt); 360 } 361 362 void 363 flushsh() 364 { 365 register int i; 366 register struct hash *hp; 367 register FILE *mesgwrit; 368 register int old = 0, new = 0; 369 370 for (i = 0; i < BUCKETS; i++) 371 for (hp = bucket[i].hnext; hp != NULL; hp = hp->hnext) 372 if (hp->hnew) 373 new++; 374 else 375 old++; 376 if (new == 0 && old != 0) 377 return; 378 mesgwrit = fopen(strings, old ? "r+" : "w"); 379 if (mesgwrit == NULL) 380 perror(strings), exit(4); 381 for (i = 0; i < BUCKETS; i++) 382 for (hp = bucket[i].hnext; hp != NULL; hp = hp->hnext) { 383 found(hp->hnew, hp->hpt, hp->hstr); 384 if (hp->hnew) { 385 fseek(mesgwrit, hp->hpt, 0); 386 ignore(fwrite(hp->hstr, strlen(hp->hstr) + 1, 1, mesgwrit)); 387 if (ferror(mesgwrit)) 388 err(4, "%s", strings); 389 } 390 } 391 if (fclose(mesgwrit) == EOF) 392 err(4, "%s", strings); 393 } 394 395 void 396 found(new, off, str) 397 int new; 398 off_t off; 399 char *str; 400 { 401 if (vflg == 0) 402 return; 403 if (!new) 404 fprintf(stderr, "found at %d:", (int) off); 405 else 406 fprintf(stderr, "new at %d:", (int) off); 407 prstr(str); 408 fprintf(stderr, "\n"); 409 } 410 411 void 412 prstr(cp) 413 register char *cp; 414 { 415 register int c; 416 417 while ((c = (*cp++ & 0377))) 418 if (c < ' ') 419 fprintf(stderr, "^%c", c + '`'); 420 else if (c == 0177) 421 fprintf(stderr, "^?"); 422 else if (c > 0200) 423 fprintf(stderr, "\\%03o", c); 424 else 425 fprintf(stderr, "%c", c); 426 } 427 428 void 429 xsdotc() 430 { 431 register FILE *strf = fopen(strings, "r"); 432 register FILE *xdotcf; 433 434 if (strf == NULL) 435 err(5, "%s", strings); 436 xdotcf = fopen("xs.c", "w"); 437 if (xdotcf == NULL) 438 err(6, "xs.c"); 439 fprintf(xdotcf, "char\txstr[] = {\n"); 440 for (;;) { 441 register int i, c; 442 443 for (i = 0; i < 8; i++) { 444 c = getc(strf); 445 if (ferror(strf)) { 446 warn("%s", strings); 447 onintr(); 448 } 449 if (feof(strf)) { 450 fprintf(xdotcf, "\n"); 451 goto out; 452 } 453 fprintf(xdotcf, "0x%02x,", c); 454 } 455 fprintf(xdotcf, "\n"); 456 } 457 out: 458 fprintf(xdotcf, "};\n"); 459 ignore(fclose(xdotcf)); 460 ignore(fclose(strf)); 461 } 462 463 char 464 lastchr(cp) 465 register char *cp; 466 { 467 468 while (cp[0] && cp[1]) 469 cp++; 470 return (*cp); 471 } 472 473 int 474 istail(str, of) 475 register char *str, *of; 476 { 477 register int d = strlen(of) - strlen(str); 478 479 if (d < 0 || strcmp(&of[d], str) != 0) 480 return (-1); 481 return (d); 482 } 483 484 void 485 onintr() 486 { 487 488 ignore(signal(SIGINT, SIG_IGN)); 489 if (strings[0] == '/') 490 ignore(unlink(strings)); 491 ignore(unlink("x.c")); 492 ignore(unlink("xs.c")); 493 exit(7); 494 } 495