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 * 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 * @(#)parse.c 8.1 (Berkeley) 6/6/93 34 * $FreeBSD: src/usr.bin/hexdump/parse.c,v 1.4.2.1 2002/07/23 14:27:06 tjr Exp $ 35 * $DragonFly: src/usr.bin/hexdump/parse.c,v 1.2 2003/06/17 04:29:27 dillon Exp $ 36 */ 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(name) 52 char *name; 53 { 54 register unsigned char *p; 55 FILE *fp; 56 int ch; 57 char buf[2048 + 1]; 58 59 if ((fp = fopen(name, "r")) == NULL) 60 err(1, "%s", name); 61 while (fgets(buf, sizeof(buf), fp)) { 62 if (!(p = index(buf, '\n'))) { 63 warnx("line too long"); 64 while ((ch = getchar()) != '\n' && ch != EOF); 65 continue; 66 } 67 *p = '\0'; 68 for (p = buf; *p && isspace(*p); ++p); 69 if (!*p || *p == '#') 70 continue; 71 add(p); 72 } 73 (void)fclose(fp); 74 } 75 76 void 77 add(fmt) 78 const char *fmt; 79 { 80 unsigned const char *p, *savep; 81 static FS **nextfs; 82 FS *tfs; 83 FU *tfu, **nextfu; 84 85 /* start new linked list of format units */ 86 if ((tfs = calloc(1, sizeof(FS))) == NULL) 87 err(1, NULL); 88 if (!fshead) 89 fshead = tfs; 90 else 91 *nextfs = tfs; 92 nextfs = &tfs->nextfs; 93 nextfu = &tfs->nextfu; 94 95 /* take the format string and break it up into format units */ 96 for (p = fmt;;) { 97 /* skip leading white space */ 98 for (; isspace(*p); ++p); 99 if (!*p) 100 break; 101 102 /* allocate a new format unit and link it in */ 103 if ((tfu = calloc(1, sizeof(FU))) == NULL) 104 err(1, NULL); 105 *nextfu = tfu; 106 nextfu = &tfu->nextfu; 107 tfu->reps = 1; 108 109 /* if leading digit, repetition count */ 110 if (isdigit(*p)) { 111 for (savep = p; isdigit(*p); ++p); 112 if (!isspace(*p) && *p != '/') 113 badfmt(fmt); 114 /* may overwrite either white space or slash */ 115 tfu->reps = atoi(savep); 116 tfu->flags = F_SETREP; 117 /* skip trailing white space */ 118 for (++p; isspace(*p); ++p); 119 } 120 121 /* skip slash and trailing white space */ 122 if (*p == '/') 123 while (isspace(*++p)); 124 125 /* byte count */ 126 if (isdigit(*p)) { 127 for (savep = p; isdigit(*p); ++p); 128 if (!isspace(*p)) 129 badfmt(fmt); 130 tfu->bcnt = atoi(savep); 131 /* skip trailing white space */ 132 for (++p; isspace(*p); ++p); 133 } 134 135 /* format */ 136 if (*p != '"') 137 badfmt(fmt); 138 for (savep = ++p; *p != '"';) 139 if (*p++ == 0) 140 badfmt(fmt); 141 if (!(tfu->fmt = malloc(p - savep + 1))) 142 err(1, NULL); 143 (void) strncpy(tfu->fmt, savep, p - savep); 144 tfu->fmt[p - savep] = '\0'; 145 escape(tfu->fmt); 146 p++; 147 } 148 } 149 150 static const char *spec = ".#-+ 0123456789"; 151 152 int 153 size(fs) 154 FS *fs; 155 { 156 register FU *fu; 157 register int bcnt, cursize; 158 register unsigned char *fmt; 159 int prec; 160 161 /* figure out the data block size needed for each format unit */ 162 for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) { 163 if (fu->bcnt) { 164 cursize += fu->bcnt * fu->reps; 165 continue; 166 } 167 for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) { 168 if (*fmt != '%') 169 continue; 170 /* 171 * skip any special chars -- save precision in 172 * case it's a %s format. 173 */ 174 while (index(spec + 1, *++fmt)); 175 if (*fmt == '.' && isdigit(*++fmt)) { 176 prec = atoi(fmt); 177 while (isdigit(*++fmt)); 178 } 179 switch(*fmt) { 180 case 'c': 181 bcnt += 1; 182 break; 183 case 'd': case 'i': case 'o': case 'u': 184 case 'x': case 'X': 185 bcnt += 4; 186 break; 187 case 'e': case 'E': case 'f': case 'g': case 'G': 188 bcnt += 8; 189 break; 190 case 's': 191 bcnt += prec; 192 break; 193 case '_': 194 switch(*++fmt) { 195 case 'c': case 'p': case 'u': 196 bcnt += 1; 197 break; 198 } 199 } 200 } 201 cursize += bcnt * fu->reps; 202 } 203 return (cursize); 204 } 205 206 void 207 rewrite(fs) 208 FS *fs; 209 { 210 enum { NOTOKAY, USEBCNT, USEPREC } sokay; 211 register PR *pr, **nextpr; 212 register FU *fu; 213 unsigned char *p1, *p2, *fmtp; 214 char savech, cs[3]; 215 int nconv, prec; 216 217 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 218 /* 219 * Break each format unit into print units; each conversion 220 * character gets its own. 221 */ 222 for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) { 223 if ((pr = calloc(1, sizeof(PR))) == NULL) 224 err(1, NULL); 225 if (!fu->nextpr) 226 fu->nextpr = pr; 227 else 228 *nextpr = pr; 229 230 /* Skip preceding text and up to the next % sign. */ 231 for (p1 = fmtp; *p1 && *p1 != '%'; ++p1); 232 233 /* Only text in the string. */ 234 if (!*p1) { 235 pr->fmt = fmtp; 236 pr->flags = F_TEXT; 237 break; 238 } 239 240 /* 241 * Get precision for %s -- if have a byte count, don't 242 * need it. 243 */ 244 if (fu->bcnt) { 245 sokay = USEBCNT; 246 /* Skip to conversion character. */ 247 for (++p1; index(spec, *p1); ++p1); 248 } else { 249 /* Skip any special chars, field width. */ 250 while (index(spec + 1, *++p1)); 251 if (*p1 == '.' && isdigit(*++p1)) { 252 sokay = USEPREC; 253 prec = atoi(p1); 254 while (isdigit(*++p1)); 255 } else 256 sokay = NOTOKAY; 257 } 258 259 p2 = p1 + 1; /* Set end pointer. */ 260 cs[0] = *p1; /* Set conversion string. */ 261 cs[1] = '\0'; 262 263 /* 264 * Figure out the byte count for each conversion; 265 * rewrite the format as necessary, set up blank- 266 * padding for end of data. 267 */ 268 switch(cs[0]) { 269 case 'c': 270 pr->flags = F_CHAR; 271 switch(fu->bcnt) { 272 case 0: case 1: 273 pr->bcnt = 1; 274 break; 275 default: 276 p1[1] = '\0'; 277 badcnt(p1); 278 } 279 break; 280 case 'd': case 'i': 281 pr->flags = F_INT; 282 goto isint; 283 case 'o': case 'u': case 'x': case 'X': 284 pr->flags = F_UINT; 285 isint: cs[2] = '\0'; 286 cs[1] = cs[0]; 287 cs[0] = 'q'; 288 switch(fu->bcnt) { 289 case 0: case 4: 290 pr->bcnt = 4; 291 break; 292 case 1: 293 pr->bcnt = 1; 294 break; 295 case 2: 296 pr->bcnt = 2; 297 break; 298 default: 299 p1[1] = '\0'; 300 badcnt(p1); 301 } 302 break; 303 case 'e': case 'E': case 'f': case 'g': case 'G': 304 pr->flags = F_DBL; 305 switch(fu->bcnt) { 306 case 0: case 8: 307 pr->bcnt = 8; 308 break; 309 case 4: 310 pr->bcnt = 4; 311 break; 312 default: 313 if (fu->bcnt == sizeof(long double)) { 314 cs[2] = '\0'; 315 cs[1] = cs[0]; 316 cs[0] = 'L'; 317 pr->bcnt = sizeof(long double); 318 } else { 319 p1[1] = '\0'; 320 badcnt(p1); 321 } 322 } 323 break; 324 case 's': 325 pr->flags = F_STR; 326 switch(sokay) { 327 case NOTOKAY: 328 badsfmt(); 329 case USEBCNT: 330 pr->bcnt = fu->bcnt; 331 break; 332 case USEPREC: 333 pr->bcnt = prec; 334 break; 335 } 336 break; 337 case '_': 338 ++p2; 339 switch(p1[1]) { 340 case 'A': 341 endfu = fu; 342 fu->flags |= F_IGNORE; 343 /* FALLTHROUGH */ 344 case 'a': 345 pr->flags = F_ADDRESS; 346 ++p2; 347 switch(p1[2]) { 348 case 'd': case 'o': case'x': 349 cs[0] = 'q'; 350 cs[1] = p1[2]; 351 cs[2] = '\0'; 352 break; 353 default: 354 p1[3] = '\0'; 355 badconv(p1); 356 } 357 break; 358 case 'c': 359 pr->flags = F_C; 360 /* cs[0] = 'c'; set in conv_c */ 361 goto isint2; 362 case 'p': 363 pr->flags = F_P; 364 cs[0] = 'c'; 365 goto isint2; 366 case 'u': 367 pr->flags = F_U; 368 /* cs[0] = 'c'; set in conv_u */ 369 isint2: switch(fu->bcnt) { 370 case 0: case 1: 371 pr->bcnt = 1; 372 break; 373 default: 374 p1[2] = '\0'; 375 badcnt(p1); 376 } 377 break; 378 default: 379 p1[2] = '\0'; 380 badconv(p1); 381 } 382 break; 383 default: 384 p1[1] = '\0'; 385 badconv(p1); 386 } 387 388 /* 389 * Copy to PR format string, set conversion character 390 * pointer, update original. 391 */ 392 savech = *p2; 393 p1[0] = '\0'; 394 if ((pr->fmt = calloc(1, strlen(fmtp) + 2)) == NULL) 395 err(1, NULL); 396 (void)strcpy(pr->fmt, fmtp); 397 (void)strcat(pr->fmt, cs); 398 *p2 = savech; 399 pr->cchar = pr->fmt + (p1 - fmtp); 400 fmtp = p2; 401 402 /* Only one conversion character if byte count. */ 403 if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++) 404 errx(1, "byte count with multiple conversion characters"); 405 } 406 /* 407 * If format unit byte count not specified, figure it out 408 * so can adjust rep count later. 409 */ 410 if (!fu->bcnt) 411 for (pr = fu->nextpr; pr; pr = pr->nextpr) 412 fu->bcnt += pr->bcnt; 413 } 414 /* 415 * If the format string interprets any data at all, and it's 416 * not the same as the blocksize, and its last format unit 417 * interprets any data at all, and has no iteration count, 418 * repeat it as necessary. 419 * 420 * If, rep count is greater than 1, no trailing whitespace 421 * gets output from the last iteration of the format unit. 422 */ 423 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 424 if (!fu->nextfu && fs->bcnt < blocksize && 425 !(fu->flags&F_SETREP) && fu->bcnt) 426 fu->reps += (blocksize - fs->bcnt) / fu->bcnt; 427 if (fu->reps > 1) { 428 for (pr = fu->nextpr;; pr = pr->nextpr) 429 if (!pr->nextpr) 430 break; 431 for (p1 = pr->fmt, p2 = NULL; *p1; ++p1) 432 p2 = isspace(*p1) ? p1 : NULL; 433 if (p2) 434 pr->nospace = p2; 435 } 436 } 437 #ifdef DEBUG 438 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 439 (void)printf("fmt:"); 440 for (pr = fu->nextpr; pr; pr = pr->nextpr) 441 (void)printf(" {%s}", pr->fmt); 442 (void)printf("\n"); 443 } 444 #endif 445 } 446 447 void 448 escape(p1) 449 register char *p1; 450 { 451 register char *p2; 452 453 /* alphabetic escape sequences have to be done in place */ 454 for (p2 = p1;; ++p1, ++p2) { 455 if (!*p1) { 456 *p2 = *p1; 457 break; 458 } 459 if (*p1 == '\\') 460 switch(*++p1) { 461 case 'a': 462 /* *p2 = '\a'; */ 463 *p2 = '\007'; 464 break; 465 case 'b': 466 *p2 = '\b'; 467 break; 468 case 'f': 469 *p2 = '\f'; 470 break; 471 case 'n': 472 *p2 = '\n'; 473 break; 474 case 'r': 475 *p2 = '\r'; 476 break; 477 case 't': 478 *p2 = '\t'; 479 break; 480 case 'v': 481 *p2 = '\v'; 482 break; 483 default: 484 *p2 = *p1; 485 break; 486 } 487 } 488 } 489 490 void 491 badcnt(s) 492 char *s; 493 { 494 errx(1, "%s: bad byte count", s); 495 } 496 497 void 498 badsfmt() 499 { 500 errx(1, "%%s: requires a precision or a byte count"); 501 } 502 503 void 504 badfmt(fmt) 505 const char *fmt; 506 { 507 errx(1, "\"%s\": bad format", fmt); 508 } 509 510 void 511 badconv(ch) 512 char *ch; 513 { 514 errx(1, "%%%s: bad conversion character", ch); 515 } 516