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 = index(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 (index(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; index(spec, *p1); ++p1); 247 } else { 248 /* Skip any special chars, field width. */ 249 while (index(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 + 1; /* Set end pointer. */ 259 cs[0] = *p1; /* Set conversion string. */ 260 cs[1] = '\0'; 261 262 /* 263 * Figure out the byte count for each conversion; 264 * rewrite the format as necessary, set up blank- 265 * padding for end of data. 266 */ 267 switch(cs[0]) { 268 case 'c': 269 pr->flags = F_CHAR; 270 switch(fu->bcnt) { 271 case 0: case 1: 272 pr->bcnt = 1; 273 break; 274 default: 275 p1[1] = '\0'; 276 badcnt(p1); 277 } 278 break; 279 case 'd': case 'i': 280 pr->flags = F_INT; 281 goto isint; 282 case 'o': case 'u': case 'x': case 'X': 283 pr->flags = F_UINT; 284 isint: cs[2] = '\0'; 285 cs[1] = cs[0]; 286 cs[0] = 'q'; 287 switch(fu->bcnt) { 288 case 0: case 4: 289 pr->bcnt = 4; 290 break; 291 case 1: 292 pr->bcnt = 1; 293 break; 294 case 2: 295 pr->bcnt = 2; 296 break; 297 default: 298 p1[1] = '\0'; 299 badcnt(p1); 300 } 301 break; 302 case 'e': case 'E': case 'f': case 'g': case 'G': 303 pr->flags = F_DBL; 304 switch(fu->bcnt) { 305 case 0: case 8: 306 pr->bcnt = 8; 307 break; 308 case 4: 309 pr->bcnt = 4; 310 break; 311 default: 312 if (fu->bcnt == sizeof(long double)) { 313 cs[2] = '\0'; 314 cs[1] = cs[0]; 315 cs[0] = 'L'; 316 pr->bcnt = sizeof(long double); 317 } else { 318 p1[1] = '\0'; 319 badcnt(p1); 320 } 321 } 322 break; 323 case 's': 324 pr->flags = F_STR; 325 switch(sokay) { 326 case NOTOKAY: 327 badsfmt(); 328 case USEBCNT: 329 pr->bcnt = fu->bcnt; 330 break; 331 case USEPREC: 332 pr->bcnt = prec; 333 break; 334 } 335 break; 336 case '_': 337 ++p2; 338 switch(p1[1]) { 339 case 'A': 340 endfu = fu; 341 fu->flags |= F_IGNORE; 342 /* FALLTHROUGH */ 343 case 'a': 344 pr->flags = F_ADDRESS; 345 ++p2; 346 switch(p1[2]) { 347 case 'd': case 'o': case'x': 348 cs[0] = 'q'; 349 cs[1] = p1[2]; 350 cs[2] = '\0'; 351 break; 352 default: 353 p1[3] = '\0'; 354 badconv(p1); 355 } 356 break; 357 case 'c': 358 pr->flags = F_C; 359 /* cs[0] = 'c'; set in conv_c */ 360 goto isint2; 361 case 'p': 362 pr->flags = F_P; 363 cs[0] = 'c'; 364 goto isint2; 365 case 'u': 366 pr->flags = F_U; 367 /* cs[0] = 'c'; set in conv_u */ 368 isint2: switch(fu->bcnt) { 369 case 0: case 1: 370 pr->bcnt = 1; 371 break; 372 default: 373 p1[2] = '\0'; 374 badcnt(p1); 375 } 376 break; 377 default: 378 p1[2] = '\0'; 379 badconv(p1); 380 } 381 break; 382 default: 383 p1[1] = '\0'; 384 badconv(p1); 385 } 386 387 /* 388 * Copy to PR format string, set conversion character 389 * pointer, update original. 390 */ 391 savech = *p2; 392 p1[0] = '\0'; 393 len = strlen(fmtp) + strlen(cs) + 1; 394 if ((pr->fmt = calloc(1, len)) == NULL) 395 err(1, NULL); 396 snprintf(pr->fmt, len, "%s%s", fmtp, cs); 397 *p2 = savech; 398 pr->cchar = pr->fmt + (p1 - fmtp); 399 fmtp = p2; 400 401 /* Only one conversion character if byte count. */ 402 if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++) 403 errx(1, "byte count with multiple conversion characters"); 404 } 405 /* 406 * If format unit byte count not specified, figure it out 407 * so can adjust rep count later. 408 */ 409 if (!fu->bcnt) 410 for (pr = fu->nextpr; pr; pr = pr->nextpr) 411 fu->bcnt += pr->bcnt; 412 } 413 /* 414 * If the format string interprets any data at all, and it's 415 * not the same as the blocksize, and its last format unit 416 * interprets any data at all, and has no iteration count, 417 * repeat it as necessary. 418 * 419 * If, rep count is greater than 1, no trailing whitespace 420 * gets output from the last iteration of the format unit. 421 */ 422 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 423 if (!fu->nextfu && fs->bcnt < blocksize && 424 !(fu->flags&F_SETREP) && fu->bcnt) 425 fu->reps += (blocksize - fs->bcnt) / fu->bcnt; 426 if (fu->reps > 1) { 427 for (pr = fu->nextpr;; pr = pr->nextpr) 428 if (!pr->nextpr) 429 break; 430 for (p1 = pr->fmt, p2 = NULL; *p1; ++p1) 431 p2 = isspace(*p1) ? p1 : NULL; 432 if (p2) 433 pr->nospace = p2; 434 } 435 } 436 #ifdef DEBUG 437 for (fu = fs->nextfu; fu; fu = fu->nextfu) { 438 (void)printf("fmt:"); 439 for (pr = fu->nextpr; pr; pr = pr->nextpr) 440 (void)printf(" {%s}", pr->fmt); 441 (void)printf("\n"); 442 } 443 #endif 444 } 445 446 void 447 escape(char *p1) 448 { 449 char *p2; 450 451 /* alphabetic escape sequences have to be done in place */ 452 for (p2 = p1;; ++p1, ++p2) { 453 if (!*p1) { 454 *p2 = *p1; 455 break; 456 } 457 if (*p1 == '\\') 458 switch(*++p1) { 459 case 'a': 460 /* *p2 = '\a'; */ 461 *p2 = '\007'; 462 break; 463 case 'b': 464 *p2 = '\b'; 465 break; 466 case 'f': 467 *p2 = '\f'; 468 break; 469 case 'n': 470 *p2 = '\n'; 471 break; 472 case 'r': 473 *p2 = '\r'; 474 break; 475 case 't': 476 *p2 = '\t'; 477 break; 478 case 'v': 479 *p2 = '\v'; 480 break; 481 default: 482 *p2 = *p1; 483 break; 484 } 485 } 486 } 487 488 void 489 badcnt(char *s) 490 { 491 errx(1, "%s: bad byte count", s); 492 } 493 494 void 495 badsfmt(void) 496 { 497 errx(1, "%%s: requires a precision or a byte count"); 498 } 499 500 void 501 badfmt(const char *fmt) 502 { 503 errx(1, "\"%s\": bad format", fmt); 504 } 505 506 void 507 badconv(char *ch) 508 { 509 errx(1, "%%%s: bad conversion character", ch); 510 } 511