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. 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 * @(#)parse.c 8.1 (Berkeley) 6/6/93
30 * $FreeBSD: src/usr.bin/hexdump/parse.c,v 1.4.2.1 2002/07/23 14:27:06 tjr Exp $
31 */
32
33 #include <sys/types.h>
34
35 #include <err.h>
36 #include <fcntl.h>
37 #include <stdio.h>
38 #include <stdlib.h>
39 #include <ctype.h>
40 #include <string.h>
41 #include "hexdump.h"
42
43 FU *endfu; /* format at end-of-data */
44
45 void
addfile(char * name)46 addfile(char *name)
47 {
48 unsigned char *p;
49 FILE *fp;
50 int ch;
51 char buf[2048 + 1];
52
53 if ((fp = fopen(name, "r")) == NULL)
54 err(1, "%s", name);
55 while (fgets(buf, sizeof(buf), fp)) {
56 if (!(p = strchr(buf, '\n'))) {
57 warnx("line too long");
58 while ((ch = getchar()) != '\n' && ch != EOF);
59 continue;
60 }
61 *p = '\0';
62 for (p = buf; *p && isspace(*p); ++p);
63 if (!*p || *p == '#')
64 continue;
65 add(p);
66 }
67 (void)fclose(fp);
68 }
69
70 void
add(const char * fmt)71 add(const char *fmt)
72 {
73 unsigned const char *p, *savep;
74 static FS **nextfs;
75 FS *tfs;
76 FU *tfu, **nextfu;
77
78 /* start new linked list of format units */
79 if ((tfs = calloc(1, sizeof(FS))) == NULL)
80 err(1, NULL);
81 if (!fshead)
82 fshead = tfs;
83 else
84 *nextfs = tfs;
85 nextfs = &tfs->nextfs;
86 nextfu = &tfs->nextfu;
87
88 /* take the format string and break it up into format units */
89 for (p = fmt;;) {
90 /* skip leading white space */
91 for (; isspace(*p); ++p);
92 if (!*p)
93 break;
94
95 /* allocate a new format unit and link it in */
96 if ((tfu = calloc(1, sizeof(FU))) == NULL)
97 err(1, NULL);
98 *nextfu = tfu;
99 nextfu = &tfu->nextfu;
100 tfu->reps = 1;
101
102 /* if leading digit, repetition count */
103 if (isdigit(*p)) {
104 for (savep = p; isdigit(*p); ++p);
105 if (!isspace(*p) && *p != '/')
106 badfmt(fmt);
107 /* may overwrite either white space or slash */
108 tfu->reps = atoi(savep);
109 tfu->flags = F_SETREP;
110 /* skip trailing white space */
111 for (++p; isspace(*p); ++p);
112 }
113
114 /* skip slash and trailing white space */
115 if (*p == '/')
116 while (isspace(*++p));
117
118 /* byte count */
119 if (isdigit(*p)) {
120 for (savep = p; isdigit(*p); ++p);
121 if (!isspace(*p))
122 badfmt(fmt);
123 tfu->bcnt = atoi(savep);
124 /* skip trailing white space */
125 for (++p; isspace(*p); ++p);
126 }
127
128 /* format */
129 if (*p != '"')
130 badfmt(fmt);
131 for (savep = ++p; *p != '"';)
132 if (*p++ == 0)
133 badfmt(fmt);
134 if (!(tfu->fmt = malloc(p - savep + 1)))
135 err(1, NULL);
136 (void) strncpy(tfu->fmt, savep, p - savep);
137 tfu->fmt[p - savep] = '\0';
138 escape(tfu->fmt);
139 p++;
140 }
141 }
142
143 static const char *spec = ".#-+ 0123456789";
144
145 int
size(FS * fs)146 size(FS *fs)
147 {
148 FU *fu;
149 int bcnt, cursize;
150 unsigned char *fmt;
151 int prec;
152
153 /* figure out the data block size needed for each format unit */
154 for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) {
155 if (fu->bcnt) {
156 cursize += fu->bcnt * fu->reps;
157 continue;
158 }
159 for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) {
160 if (*fmt != '%')
161 continue;
162 /*
163 * skip any special chars -- save precision in
164 * case it's a %s format.
165 */
166 while (strchr(spec + 1, *++fmt));
167 if (*fmt == '.' && isdigit(*++fmt)) {
168 prec = atoi(fmt);
169 while (isdigit(*++fmt));
170 }
171 switch(*fmt) {
172 case 'c':
173 bcnt += 1;
174 break;
175 case 'd': case 'i': case 'o': case 'u':
176 case 'x': case 'X':
177 bcnt += 4;
178 break;
179 case 'e': case 'E': case 'f': case 'g': case 'G':
180 bcnt += 8;
181 break;
182 case 's':
183 bcnt += prec;
184 break;
185 case '_':
186 switch(*++fmt) {
187 case 'c': case 'p': case 'u':
188 bcnt += 1;
189 break;
190 }
191 }
192 }
193 cursize += bcnt * fu->reps;
194 }
195 return (cursize);
196 }
197
198 void
rewrite(FS * fs)199 rewrite(FS *fs)
200 {
201 enum { NOTOKAY, USEBCNT, USEPREC } sokay;
202 PR *pr, **nextpr = NULL;
203 FU *fu;
204 unsigned char *p1, *p2, *fmtp;
205 char savech, cs[3];
206 int nconv, prec = 0;
207
208 for (fu = fs->nextfu; fu; fu = fu->nextfu) {
209 /*
210 * Break each format unit into print units; each conversion
211 * character gets its own.
212 */
213 for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) {
214 if ((pr = calloc(1, sizeof(PR))) == NULL)
215 err(1, NULL);
216 if (!fu->nextpr)
217 fu->nextpr = pr;
218 else
219 *nextpr = pr;
220
221 /* Skip preceding text and up to the next % sign. */
222 for (p1 = fmtp; *p1 && *p1 != '%'; ++p1);
223
224 /* Only text in the string. */
225 if (!*p1) {
226 pr->fmt = fmtp;
227 pr->flags = F_TEXT;
228 break;
229 }
230
231 /*
232 * Get precision for %s -- if have a byte count, don't
233 * need it.
234 */
235 if (fu->bcnt) {
236 sokay = USEBCNT;
237 /* Skip to conversion character. */
238 for (++p1; strchr(spec, *p1); ++p1);
239 } else {
240 /* Skip any special chars, field width. */
241 while (strchr(spec + 1, *++p1));
242 if (*p1 == '.' && isdigit(*++p1)) {
243 sokay = USEPREC;
244 prec = atoi(p1);
245 while (isdigit(*++p1));
246 } else
247 sokay = NOTOKAY;
248 }
249
250 p2 = p1 + 1; /* Set end pointer. */
251 cs[0] = *p1; /* Set conversion string. */
252 cs[1] = '\0';
253
254 /*
255 * Figure out the byte count for each conversion;
256 * rewrite the format as necessary, set up blank-
257 * padding for end of data.
258 */
259 switch(cs[0]) {
260 case 'c':
261 pr->flags = F_CHAR;
262 switch(fu->bcnt) {
263 case 0: case 1:
264 pr->bcnt = 1;
265 break;
266 default:
267 p1[1] = '\0';
268 badcnt(p1);
269 }
270 break;
271 case 'd': case 'i':
272 pr->flags = F_INT;
273 goto isint;
274 case 'o': case 'u': case 'x': case 'X':
275 pr->flags = F_UINT;
276 isint: cs[2] = '\0';
277 cs[1] = cs[0];
278 cs[0] = 'q';
279 switch(fu->bcnt) {
280 case 0:
281 pr->bcnt = 4;
282 break;
283 case 1:
284 case 2:
285 case 4:
286 case 8:
287 pr->bcnt = fu->bcnt;
288 break;
289 default:
290 p1[1] = '\0';
291 badcnt(p1);
292 }
293 break;
294 case 'e': case 'E': case 'f': case 'g': case 'G':
295 pr->flags = F_DBL;
296 switch(fu->bcnt) {
297 case 0: case 8:
298 pr->bcnt = 8;
299 break;
300 case 4:
301 pr->bcnt = 4;
302 break;
303 default:
304 if (fu->bcnt == sizeof(long double)) {
305 cs[2] = '\0';
306 cs[1] = cs[0];
307 cs[0] = 'L';
308 pr->bcnt = sizeof(long double);
309 } else {
310 p1[1] = '\0';
311 badcnt(p1);
312 }
313 }
314 break;
315 case 's':
316 pr->flags = F_STR;
317 switch(sokay) {
318 case NOTOKAY:
319 badsfmt();
320 case USEBCNT:
321 pr->bcnt = fu->bcnt;
322 break;
323 case USEPREC:
324 pr->bcnt = prec;
325 break;
326 }
327 break;
328 case '_':
329 ++p2;
330 switch(p1[1]) {
331 case 'A':
332 endfu = fu;
333 fu->flags |= F_IGNORE;
334 /* FALLTHROUGH */
335 case 'a':
336 pr->flags = F_ADDRESS;
337 ++p2;
338 switch(p1[2]) {
339 case 'd': case 'o': case'x':
340 cs[0] = 'q';
341 cs[1] = p1[2];
342 cs[2] = '\0';
343 break;
344 default:
345 p1[3] = '\0';
346 badconv(p1);
347 }
348 break;
349 case 'c':
350 pr->flags = F_C;
351 /* cs[0] = 'c'; set in conv_c */
352 goto isint2;
353 case 'p':
354 pr->flags = F_P;
355 cs[0] = 'c';
356 goto isint2;
357 case 'u':
358 pr->flags = F_U;
359 /* cs[0] = 'c'; set in conv_u */
360 isint2: switch(fu->bcnt) {
361 case 0: case 1:
362 pr->bcnt = 1;
363 break;
364 default:
365 p1[2] = '\0';
366 badcnt(p1);
367 }
368 break;
369 default:
370 p1[2] = '\0';
371 badconv(p1);
372 }
373 break;
374 default:
375 p1[1] = '\0';
376 badconv(p1);
377 }
378
379 /*
380 * Copy to PR format string, set conversion character
381 * pointer, update original.
382 */
383 savech = *p2;
384 p1[0] = '\0';
385 if ((pr->fmt = calloc(1, strlen(fmtp) + 2)) == NULL)
386 err(1, NULL);
387 (void)strcpy(pr->fmt, fmtp);
388 (void)strcat(pr->fmt, cs);
389 *p2 = savech;
390 pr->cchar = pr->fmt + (p1 - fmtp);
391 fmtp = p2;
392
393 /* Only one conversion character if byte count. */
394 if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++) {
395 errx(1, "byte count with multiple "
396 "conversion characters");
397 }
398 }
399 /*
400 * If format unit byte count not specified, figure it out
401 * so can adjust rep count later.
402 */
403 if (!fu->bcnt)
404 for (pr = fu->nextpr; pr; pr = pr->nextpr)
405 fu->bcnt += pr->bcnt;
406 }
407 /*
408 * If the format string interprets any data at all, and it's
409 * not the same as the blocksize, and its last format unit
410 * interprets any data at all, and has no iteration count,
411 * repeat it as necessary.
412 *
413 * If, rep count is greater than 1, no trailing whitespace
414 * gets output from the last iteration of the format unit.
415 */
416 for (fu = fs->nextfu; fu; fu = fu->nextfu) {
417 if (!fu->nextfu && fs->bcnt < blocksize &&
418 !(fu->flags&F_SETREP) && fu->bcnt)
419 fu->reps += (blocksize - fs->bcnt) / fu->bcnt;
420 if (fu->reps > 1) {
421 for (pr = fu->nextpr;; pr = pr->nextpr)
422 if (!pr->nextpr)
423 break;
424 for (p1 = pr->fmt, p2 = NULL; *p1; ++p1)
425 p2 = isspace(*p1) ? p1 : NULL;
426 if (p2)
427 pr->nospace = p2;
428 }
429 }
430 #ifdef DEBUG
431 for (fu = fs->nextfu; fu; fu = fu->nextfu) {
432 (void)printf("fmt:");
433 for (pr = fu->nextpr; pr; pr = pr->nextpr)
434 (void)printf(" {%s}", pr->fmt);
435 (void)printf("\n");
436 }
437 #endif
438 }
439
440 void
escape(char * p1)441 escape(char *p1)
442 {
443 char *p2;
444
445 /* alphabetic escape sequences have to be done in place */
446 for (p2 = p1;; ++p1, ++p2) {
447 if (!*p1) {
448 *p2 = *p1;
449 break;
450 }
451 if (*p1 == '\\')
452 switch(*++p1) {
453 case 'a':
454 /* *p2 = '\a'; */
455 *p2 = '\007';
456 break;
457 case 'b':
458 *p2 = '\b';
459 break;
460 case 'f':
461 *p2 = '\f';
462 break;
463 case 'n':
464 *p2 = '\n';
465 break;
466 case 'r':
467 *p2 = '\r';
468 break;
469 case 't':
470 *p2 = '\t';
471 break;
472 case 'v':
473 *p2 = '\v';
474 break;
475 default:
476 *p2 = *p1;
477 break;
478 }
479 }
480 }
481
482 void
badcnt(char * s)483 badcnt(char *s)
484 {
485 errx(1, "%s: bad byte count", s);
486 }
487
488 void
badsfmt(void)489 badsfmt(void)
490 {
491 errx(1, "%%s: requires a precision or a byte count");
492 }
493
494 void
badfmt(const char * fmt)495 badfmt(const char *fmt)
496 {
497 errx(1, "\"%s\": bad format", fmt);
498 }
499
500 void
badconv(char * ch)501 badconv(char *ch)
502 {
503 errx(1, "%%%s: bad conversion character", ch);
504 }
505