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