xref: /freebsd/usr.bin/hexdump/parse.c (revision d37ea99837e6ad50837fd9fe1771ddf1c3ba6002)
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 
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(char *p1)
449 {
450 	char *p2;
451 
452 	/* alphabetic escape sequences have to be done in place */
453 	for (p2 = p1;; ++p1, ++p2) {
454 		if (!*p1) {
455 			*p2 = *p1;
456 			break;
457 		}
458 		if (*p1 == '\\')
459 			switch(*++p1) {
460 			case 'a':
461 			     /* *p2 = '\a'; */
462 				*p2 = '\007';
463 				break;
464 			case 'b':
465 				*p2 = '\b';
466 				break;
467 			case 'f':
468 				*p2 = '\f';
469 				break;
470 			case 'n':
471 				*p2 = '\n';
472 				break;
473 			case 'r':
474 				*p2 = '\r';
475 				break;
476 			case 't':
477 				*p2 = '\t';
478 				break;
479 			case 'v':
480 				*p2 = '\v';
481 				break;
482 			default:
483 				*p2 = *p1;
484 				break;
485 			}
486 	}
487 }
488 
489 void
490 badcnt(char *s)
491 {
492 	errx(1, "%s: bad byte count", s);
493 }
494 
495 void
496 badsfmt(void)
497 {
498 	errx(1, "%%s: requires a precision or a byte count");
499 }
500 
501 void
502 badfmt(const char *fmt)
503 {
504 	errx(1, "\"%s\": bad format", fmt);
505 }
506 
507 void
508 badconv(char *ch)
509 {
510 	errx(1, "%%%s: bad conversion character", ch);
511 }
512