1 /* 2 * Copyright (c) 1989, 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 * 4. Neither the name of the University nor the names of its contributors 14 * may be used to endorse or promote products derived from this software 15 * without specific prior written permission. 16 * 17 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 20 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 */ 29 30 #ifndef lint 31 #if 0 32 static char sccsid[] = "@(#)parse.c 8.1 (Berkeley) 6/6/93"; 33 #endif 34 #endif /* not lint */ 35 #include <sys/cdefs.h> 36 __FBSDID("$FreeBSD$"); 37 38 #include <sys/types.h> 39 40 #include <err.h> 41 #include <fcntl.h> 42 #include <stdio.h> 43 #include <stdlib.h> 44 #include <ctype.h> 45 #include <string.h> 46 #include "hexdump.h" 47 48 FU *endfu; /* format at end-of-data */ 49 50 void 51 addfile(char *name) 52 { 53 unsigned char *p; 54 FILE *fp; 55 int ch; 56 char buf[2048 + 1]; 57 58 if ((fp = fopen(name, "r")) == NULL) 59 err(1, "%s", name); 60 while (fgets(buf, sizeof(buf), fp)) { 61 if (!(p = strchr(buf, '\n'))) { 62 warnx("line too long"); 63 while ((ch = getchar()) != '\n' && ch != EOF); 64 continue; 65 } 66 *p = '\0'; 67 for (p = buf; *p && isspace(*p); ++p); 68 if (!*p || *p == '#') 69 continue; 70 add(p); 71 } 72 (void)fclose(fp); 73 } 74 75 void 76 add(const char *fmt) 77 { 78 unsigned const char *p, *savep; 79 static FS **nextfs; 80 FS *tfs; 81 FU *tfu, **nextfu; 82 83 /* start new linked list of format units */ 84 if ((tfs = calloc(1, sizeof(FS))) == NULL) 85 err(1, NULL); 86 if (!fshead) 87 fshead = tfs; 88 else 89 *nextfs = tfs; 90 nextfs = &tfs->nextfs; 91 nextfu = &tfs->nextfu; 92 93 /* take the format string and break it up into format units */ 94 for (p = fmt;;) { 95 /* skip leading white space */ 96 for (; isspace(*p); ++p); 97 if (!*p) 98 break; 99 100 /* allocate a new format unit and link it in */ 101 if ((tfu = calloc(1, sizeof(FU))) == NULL) 102 err(1, NULL); 103 *nextfu = tfu; 104 nextfu = &tfu->nextfu; 105 tfu->reps = 1; 106 107 /* if leading digit, repetition count */ 108 if (isdigit(*p)) { 109 for (savep = p; isdigit(*p); ++p); 110 if (!isspace(*p) && *p != '/') 111 badfmt(fmt); 112 /* may overwrite either white space or slash */ 113 tfu->reps = atoi(savep); 114 tfu->flags = F_SETREP; 115 /* skip trailing white space */ 116 for (++p; isspace(*p); ++p); 117 } 118 119 /* skip slash and trailing white space */ 120 if (*p == '/') 121 while (isspace(*++p)); 122 123 /* byte count */ 124 if (isdigit(*p)) { 125 for (savep = p; isdigit(*p); ++p); 126 if (!isspace(*p)) 127 badfmt(fmt); 128 tfu->bcnt = atoi(savep); 129 /* skip trailing white space */ 130 for (++p; isspace(*p); ++p); 131 } 132 133 /* format */ 134 if (*p != '"') 135 badfmt(fmt); 136 for (savep = ++p; *p != '"';) 137 if (*p++ == 0) 138 badfmt(fmt); 139 if (!(tfu->fmt = malloc(p - savep + 1))) 140 err(1, NULL); 141 (void) strlcpy(tfu->fmt, savep, p - savep + 1); 142 escape(tfu->fmt); 143 p++; 144 } 145 } 146 147 static const char *spec = ".#-+ 0123456789"; 148 149 int 150 size(FS *fs) 151 { 152 FU *fu; 153 int bcnt, cursize; 154 unsigned char *fmt; 155 int prec; 156 157 /* figure out the data block size needed for each format unit */ 158 for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) { 159 if (fu->bcnt) { 160 cursize += fu->bcnt * fu->reps; 161 continue; 162 } 163 for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) { 164 if (*fmt != '%') 165 continue; 166 /* 167 * skip any special chars -- save precision in 168 * case it's a %s format. 169 */ 170 while (strchr(spec + 1, *++fmt)); 171 if (*fmt == '.' && isdigit(*++fmt)) { 172 prec = atoi(fmt); 173 while (isdigit(*++fmt)); 174 } 175 switch(*fmt) { 176 case 'c': 177 bcnt += 1; 178 break; 179 case 'd': case 'i': case 'o': case 'u': 180 case 'x': case 'X': 181 bcnt += 4; 182 break; 183 case 'e': case 'E': case 'f': case 'g': case 'G': 184 bcnt += 8; 185 break; 186 case 's': 187 bcnt += prec; 188 break; 189 case '_': 190 switch(*++fmt) { 191 case 'c': case 'p': case 'u': 192 bcnt += 1; 193 break; 194 } 195 } 196 } 197 cursize += bcnt * fu->reps; 198 } 199 return (cursize); 200 } 201 202 void 203 rewrite(FS *fs) 204 { 205 enum { NOTOKAY, USEBCNT, USEPREC } sokay; 206 PR *pr, **nextpr; 207 FU *fu; 208 unsigned char *p1, *p2, *fmtp; 209 char savech, cs[3]; 210 int nconv, prec; 211 size_t len; 212 213 nextpr = NULL; 214 prec = 0; 215 216 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 217 /* 218 * Break each format unit into print units; each conversion 219 * character gets its own. 220 */ 221 for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) { 222 if ((pr = calloc(1, sizeof(PR))) == NULL) 223 err(1, NULL); 224 if (!fu->nextpr) 225 fu->nextpr = pr; 226 else 227 *nextpr = pr; 228 229 /* Skip preceding text and up to the next % sign. */ 230 for (p1 = fmtp; *p1 && *p1 != '%'; ++p1); 231 232 /* Only text in the string. */ 233 if (!*p1) { 234 pr->fmt = fmtp; 235 pr->flags = F_TEXT; 236 break; 237 } 238 239 /* 240 * Get precision for %s -- if have a byte count, don't 241 * need it. 242 */ 243 if (fu->bcnt) { 244 sokay = USEBCNT; 245 /* Skip to conversion character. */ 246 for (++p1; strchr(spec, *p1); ++p1); 247 } else { 248 /* Skip any special chars, field width. */ 249 while (strchr(spec + 1, *++p1)); 250 if (*p1 == '.' && isdigit(*++p1)) { 251 sokay = USEPREC; 252 prec = atoi(p1); 253 while (isdigit(*++p1)); 254 } else 255 sokay = NOTOKAY; 256 } 257 258 p2 = *p1 ? p1 + 1 : p1; /* Set end pointer -- make sure 259 * that it's non-NUL/-NULL first 260 * though. */ 261 cs[0] = *p1; /* Set conversion string. */ 262 cs[1] = '\0'; 263 264 /* 265 * Figure out the byte count for each conversion; 266 * rewrite the format as necessary, set up blank- 267 * padding for end of data. 268 */ 269 switch(cs[0]) { 270 case 'c': 271 pr->flags = F_CHAR; 272 switch(fu->bcnt) { 273 case 0: case 1: 274 pr->bcnt = 1; 275 break; 276 default: 277 p1[1] = '\0'; 278 badcnt(p1); 279 } 280 break; 281 case 'd': case 'i': 282 pr->flags = F_INT; 283 goto isint; 284 case 'o': case 'u': case 'x': case 'X': 285 pr->flags = F_UINT; 286 isint: cs[2] = '\0'; 287 cs[1] = cs[0]; 288 cs[0] = 'q'; 289 switch(fu->bcnt) { 290 case 0: case 4: 291 pr->bcnt = 4; 292 break; 293 case 1: 294 pr->bcnt = 1; 295 break; 296 case 2: 297 pr->bcnt = 2; 298 break; 299 default: 300 p1[1] = '\0'; 301 badcnt(p1); 302 } 303 break; 304 case 'e': case 'E': case 'f': case 'g': case 'G': 305 pr->flags = F_DBL; 306 switch(fu->bcnt) { 307 case 0: case 8: 308 pr->bcnt = 8; 309 break; 310 case 4: 311 pr->bcnt = 4; 312 break; 313 default: 314 if (fu->bcnt == sizeof(long double)) { 315 cs[2] = '\0'; 316 cs[1] = cs[0]; 317 cs[0] = 'L'; 318 pr->bcnt = sizeof(long double); 319 } else { 320 p1[1] = '\0'; 321 badcnt(p1); 322 } 323 } 324 break; 325 case 's': 326 pr->flags = F_STR; 327 switch(sokay) { 328 case NOTOKAY: 329 badsfmt(); 330 case USEBCNT: 331 pr->bcnt = fu->bcnt; 332 break; 333 case USEPREC: 334 pr->bcnt = prec; 335 break; 336 } 337 break; 338 case '_': 339 ++p2; 340 switch(p1[1]) { 341 case 'A': 342 endfu = fu; 343 fu->flags |= F_IGNORE; 344 /* FALLTHROUGH */ 345 case 'a': 346 pr->flags = F_ADDRESS; 347 ++p2; 348 switch(p1[2]) { 349 case 'd': case 'o': case'x': 350 cs[0] = 'q'; 351 cs[1] = p1[2]; 352 cs[2] = '\0'; 353 break; 354 default: 355 p1[3] = '\0'; 356 badconv(p1); 357 } 358 break; 359 case 'c': 360 pr->flags = F_C; 361 /* cs[0] = 'c'; set in conv_c */ 362 goto isint2; 363 case 'p': 364 pr->flags = F_P; 365 cs[0] = 'c'; 366 goto isint2; 367 case 'u': 368 pr->flags = F_U; 369 /* cs[0] = 'c'; set in conv_u */ 370 isint2: switch(fu->bcnt) { 371 case 0: case 1: 372 pr->bcnt = 1; 373 break; 374 default: 375 p1[2] = '\0'; 376 badcnt(p1); 377 } 378 break; 379 default: 380 p1[2] = '\0'; 381 badconv(p1); 382 } 383 break; 384 default: 385 p1[1] = '\0'; 386 badconv(p1); 387 } 388 389 /* 390 * Copy to PR format string, set conversion character 391 * pointer, update original. 392 */ 393 savech = *p2; 394 p1[0] = '\0'; 395 len = strlen(fmtp) + strlen(cs) + 1; 396 if ((pr->fmt = calloc(1, len)) == NULL) 397 err(1, NULL); 398 snprintf(pr->fmt, len, "%s%s", fmtp, cs); 399 *p2 = savech; 400 pr->cchar = pr->fmt + (p1 - fmtp); 401 fmtp = p2; 402 403 /* Only one conversion character if byte count. */ 404 if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++) 405 errx(1, "byte count with multiple conversion characters"); 406 } 407 /* 408 * If format unit byte count not specified, figure it out 409 * so can adjust rep count later. 410 */ 411 if (!fu->bcnt) 412 for (pr = fu->nextpr; pr; pr = pr->nextpr) 413 fu->bcnt += pr->bcnt; 414 } 415 /* 416 * If the format string interprets any data at all, and it's 417 * not the same as the blocksize, and its last format unit 418 * interprets any data at all, and has no iteration count, 419 * repeat it as necessary. 420 * 421 * If, rep count is greater than 1, no trailing whitespace 422 * gets output from the last iteration of the format unit. 423 */ 424 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 425 if (!fu->nextfu && fs->bcnt < blocksize && 426 !(fu->flags&F_SETREP) && fu->bcnt) 427 fu->reps += (blocksize - fs->bcnt) / fu->bcnt; 428 if (fu->reps > 1) { 429 for (pr = fu->nextpr;; pr = pr->nextpr) 430 if (!pr->nextpr) 431 break; 432 for (p1 = pr->fmt, p2 = NULL; *p1; ++p1) 433 p2 = isspace(*p1) ? p1 : NULL; 434 if (p2) 435 pr->nospace = p2; 436 } 437 } 438 #ifdef DEBUG 439 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 440 (void)printf("fmt:"); 441 for (pr = fu->nextpr; pr; pr = pr->nextpr) 442 (void)printf(" {%s}", pr->fmt); 443 (void)printf("\n"); 444 } 445 #endif 446 } 447 448 void 449 escape(char *p1) 450 { 451 char *p2; 452 453 /* alphabetic escape sequences have to be done in place */ 454 for (p2 = p1;; p1++, p2++) { 455 if (*p1 == '\\') { 456 p1++; 457 switch(*p1) { 458 case '\0': 459 *p2 = '\\'; 460 *++p2 = '\0'; 461 return; 462 case 'a': 463 /* *p2 = '\a'; */ 464 *p2 = '\007'; 465 break; 466 case 'b': 467 *p2 = '\b'; 468 break; 469 case 'f': 470 *p2 = '\f'; 471 break; 472 case 'n': 473 *p2 = '\n'; 474 break; 475 case 'r': 476 *p2 = '\r'; 477 break; 478 case 't': 479 *p2 = '\t'; 480 break; 481 case 'v': 482 *p2 = '\v'; 483 break; 484 default: 485 *p2 = *p1; 486 break; 487 } 488 } else { 489 *p2 = *p1; 490 if (*p1 == '\0') 491 return; 492 } 493 } 494 } 495 496 void 497 badcnt(char *s) 498 { 499 errx(1, "%s: bad byte count", s); 500 } 501 502 void 503 badsfmt(void) 504 { 505 errx(1, "%%s: requires a precision or a byte count"); 506 } 507 508 void 509 badfmt(const char *fmt) 510 { 511 errx(1, "\"%s\": bad format", fmt); 512 } 513 514 void 515 badconv(char *ch) 516 { 517 errx(1, "%%%s: bad conversion character", ch); 518 } 519