xref: /freebsd/usr.bin/hexdump/parse.c (revision a8445737e740901f5f2c8d24c12ef7fc8b00134e)
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 static const char rcsid[] =
39 	"$Id: parse.c,v 1.2 1997/07/10 06:48:24 charnier Exp $";
40 #endif /* not lint */
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(name)
56 	char *name;
57 {
58 	register unsigned char *p;
59 	FILE *fp;
60 	int ch;
61 	char buf[2048 + 1];
62 
63 	if ((fp = fopen(name, "r")) == NULL)
64 		err(1, "%s", name);
65 	while (fgets(buf, sizeof(buf), fp)) {
66 		if (!(p = index(buf, '\n'))) {
67 			warnx("line too long");
68 			while ((ch = getchar()) != '\n' && ch != EOF);
69 			continue;
70 		}
71 		*p = '\0';
72 		for (p = buf; *p && isspace(*p); ++p);
73 		if (!*p || *p == '#')
74 			continue;
75 		add(p);
76 	}
77 	(void)fclose(fp);
78 }
79 
80 void
81 add(fmt)
82 	char *fmt;
83 {
84 	unsigned char *p, *savep;
85 	static FS **nextfs;
86 	FS *tfs;
87 	FU *tfu, **nextfu;
88 
89 	/* start new linked list of format units */
90 	tfs = emalloc(sizeof(FS));
91 	if (!fshead)
92 		fshead = tfs;
93 	else
94 		*nextfs = tfs;
95 	nextfs = &tfs->nextfs;
96 	nextfu = &tfs->nextfu;
97 
98 	/* take the format string and break it up into format units */
99 	for (p = fmt;;) {
100 		/* skip leading white space */
101 		for (; isspace(*p); ++p);
102 		if (!*p)
103 			break;
104 
105 		/* allocate a new format unit and link it in */
106 		tfu = emalloc(sizeof(FU));
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 			nomem();
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 char *spec = ".#-+ 0123456789";
153 
154 int
155 size(fs)
156 	FS *fs;
157 {
158 	register FU *fu;
159 	register int bcnt, cursize;
160 	register unsigned char *fmt;
161 	int prec;
162 
163 	/* figure out the data block size needed for each format unit */
164 	for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) {
165 		if (fu->bcnt) {
166 			cursize += fu->bcnt * fu->reps;
167 			continue;
168 		}
169 		for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) {
170 			if (*fmt != '%')
171 				continue;
172 			/*
173 			 * skip any special chars -- save precision in
174 			 * case it's a %s format.
175 			 */
176 			while (index(spec + 1, *++fmt));
177 			if (*fmt == '.' && isdigit(*++fmt)) {
178 				prec = atoi(fmt);
179 				while (isdigit(*++fmt));
180 			}
181 			switch(*fmt) {
182 			case 'c':
183 				bcnt += 1;
184 				break;
185 			case 'd': case 'i': case 'o': case 'u':
186 			case 'x': case 'X':
187 				bcnt += 4;
188 				break;
189 			case 'e': case 'E': case 'f': case 'g': case 'G':
190 				bcnt += 8;
191 				break;
192 			case 's':
193 				bcnt += prec;
194 				break;
195 			case '_':
196 				switch(*++fmt) {
197 				case 'c': case 'p': case 'u':
198 					bcnt += 1;
199 					break;
200 				}
201 			}
202 		}
203 		cursize += bcnt * fu->reps;
204 	}
205 	return (cursize);
206 }
207 
208 void
209 rewrite(fs)
210 	FS *fs;
211 {
212 	enum { NOTOKAY, USEBCNT, USEPREC } sokay;
213 	register PR *pr, **nextpr;
214 	register FU *fu;
215 	unsigned char *p1, *p2, *fmtp;
216 	char savech, cs[3];
217 	int nconv, prec;
218 
219 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
220 		/*
221 		 * Break each format unit into print units; each conversion
222 		 * character gets its own.
223 		 */
224 		for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) {
225 			pr = emalloc(sizeof(PR));
226 			if (!fu->nextpr)
227 				fu->nextpr = pr;
228 			else
229 				*nextpr = pr;
230 
231 			/* Skip preceding text and up to the next % sign. */
232 			for (p1 = fmtp; *p1 && *p1 != '%'; ++p1);
233 
234 			/* Only text in the string. */
235 			if (!*p1) {
236 				pr->fmt = fmtp;
237 				pr->flags = F_TEXT;
238 				break;
239 			}
240 
241 			/*
242 			 * Get precision for %s -- if have a byte count, don't
243 			 * need it.
244 			 */
245 			if (fu->bcnt) {
246 				sokay = USEBCNT;
247 				/* Skip to conversion character. */
248 				for (++p1; index(spec, *p1); ++p1);
249 			} else {
250 				/* Skip any special chars, field width. */
251 				while (index(spec + 1, *++p1));
252 				if (*p1 == '.' && isdigit(*++p1)) {
253 					sokay = USEPREC;
254 					prec = atoi(p1);
255 					while (isdigit(*++p1));
256 				} else
257 					sokay = NOTOKAY;
258 			}
259 
260 			p2 = p1 + 1;		/* Set end pointer. */
261 			cs[0] = *p1;		/* Set conversion string. */
262 			cs[1] = '\0';
263 
264 			/*
265 			 * Figure out the byte count for each conversion;
266 			 * rewrite the format as necessary, set up blank-
267 			 * padding for end of data.
268 			 */
269 			switch(cs[0]) {
270 			case 'c':
271 				pr->flags = F_CHAR;
272 				switch(fu->bcnt) {
273 				case 0: case 1:
274 					pr->bcnt = 1;
275 					break;
276 				default:
277 					p1[1] = '\0';
278 					badcnt(p1);
279 				}
280 				break;
281 			case 'd': case 'i':
282 				pr->flags = F_INT;
283 				goto isint;
284 			case 'o': case 'u': case 'x': case 'X':
285 				pr->flags = F_UINT;
286 isint:				cs[2] = '\0';
287 				cs[1] = cs[0];
288 				cs[0] = 'q';
289 				switch(fu->bcnt) {
290 				case 0: case 4:
291 					pr->bcnt = 4;
292 					break;
293 				case 1:
294 					pr->bcnt = 1;
295 					break;
296 				case 2:
297 					pr->bcnt = 2;
298 					break;
299 				default:
300 					p1[1] = '\0';
301 					badcnt(p1);
302 				}
303 				break;
304 			case 'e': case 'E': case 'f': case 'g': case 'G':
305 				pr->flags = F_DBL;
306 				switch(fu->bcnt) {
307 				case 0: case 8:
308 					pr->bcnt = 8;
309 					break;
310 				case 4:
311 					pr->bcnt = 4;
312 					break;
313 				default:
314 					p1[1] = '\0';
315 					badcnt(p1);
316 				}
317 				break;
318 			case 's':
319 				pr->flags = F_STR;
320 				switch(sokay) {
321 				case NOTOKAY:
322 					badsfmt();
323 				case USEBCNT:
324 					pr->bcnt = fu->bcnt;
325 					break;
326 				case USEPREC:
327 					pr->bcnt = prec;
328 					break;
329 				}
330 				break;
331 			case '_':
332 				++p2;
333 				switch(p1[1]) {
334 				case 'A':
335 					endfu = fu;
336 					fu->flags |= F_IGNORE;
337 					/* FALLTHROUGH */
338 				case 'a':
339 					pr->flags = F_ADDRESS;
340 					++p2;
341 					switch(p1[2]) {
342 					case 'd': case 'o': case'x':
343 						cs[0] = 'q';
344 						cs[1] = p1[2];
345 						cs[2] = '\0';
346 						break;
347 					default:
348 						p1[3] = '\0';
349 						badconv(p1);
350 					}
351 					break;
352 				case 'c':
353 					pr->flags = F_C;
354 					/* cs[0] = 'c';	set in conv_c */
355 					goto isint2;
356 				case 'p':
357 					pr->flags = F_P;
358 					cs[0] = 'c';
359 					goto isint2;
360 				case 'u':
361 					pr->flags = F_U;
362 					/* cs[0] = 'c';	set in conv_u */
363 isint2:					switch(fu->bcnt) {
364 					case 0: case 1:
365 						pr->bcnt = 1;
366 						break;
367 					default:
368 						p1[2] = '\0';
369 						badcnt(p1);
370 					}
371 					break;
372 				default:
373 					p1[2] = '\0';
374 					badconv(p1);
375 				}
376 				break;
377 			default:
378 				p1[1] = '\0';
379 				badconv(p1);
380 			}
381 
382 			/*
383 			 * Copy to PR format string, set conversion character
384 			 * pointer, update original.
385 			 */
386 			savech = *p2;
387 			p1[0] = '\0';
388 			pr->fmt = emalloc(strlen(fmtp) + 2);
389 			(void)strcpy(pr->fmt, fmtp);
390 			(void)strcat(pr->fmt, cs);
391 			*p2 = savech;
392 			pr->cchar = pr->fmt + (p1 - fmtp);
393 			fmtp = p2;
394 
395 			/* Only one conversion character if byte count. */
396 			if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++)
397 	    errx(1, "byte count with multiple conversion characters");
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->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 		if (!fu->nextfu)
430 			break;
431 	}
432 #ifdef DEBUG
433 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
434 		(void)printf("fmt:");
435 		for (pr = fu->nextpr; pr; pr = pr->nextpr)
436 			(void)printf(" {%s}", pr->fmt);
437 		(void)printf("\n");
438 	}
439 #endif
440 }
441 
442 void
443 escape(p1)
444 	register char *p1;
445 {
446 	register char *p2;
447 
448 	/* alphabetic escape sequences have to be done in place */
449 	for (p2 = p1;; ++p1, ++p2) {
450 		if (!*p1) {
451 			*p2 = *p1;
452 			break;
453 		}
454 		if (*p1 == '\\')
455 			switch(*++p1) {
456 			case 'a':
457 			     /* *p2 = '\a'; */
458 				*p2 = '\007';
459 				break;
460 			case 'b':
461 				*p2 = '\b';
462 				break;
463 			case 'f':
464 				*p2 = '\f';
465 				break;
466 			case 'n':
467 				*p2 = '\n';
468 				break;
469 			case 'r':
470 				*p2 = '\r';
471 				break;
472 			case 't':
473 				*p2 = '\t';
474 				break;
475 			case 'v':
476 				*p2 = '\v';
477 				break;
478 			default:
479 				*p2 = *p1;
480 				break;
481 			}
482 	}
483 }
484 
485 void
486 badcnt(s)
487 	char *s;
488 {
489 	errx(1, "%s: bad byte count", s);
490 }
491 
492 void
493 badsfmt()
494 {
495 	errx(1, "%%s: requires a precision or a byte count");
496 }
497 
498 void
499 badfmt(fmt)
500 	char *fmt;
501 {
502 	errx(1, "\"%s\": bad format", fmt);
503 }
504 
505 void
506 badconv(ch)
507 	char *ch;
508 {
509 	errx(1, "%%%s: bad conversion character", ch);
510 }
511