xref: /freebsd/usr.bin/hexdump/parse.c (revision 0fddbf874719b9bd50cf66ac26d1140bb3f2be69)
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   "$FreeBSD$";
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 	if ((tfs = calloc(1, sizeof(FS))) == NULL)
91 		err(1, NULL);
92 	if (!fshead)
93 		fshead = tfs;
94 	else
95 		*nextfs = tfs;
96 	nextfs = &tfs->nextfs;
97 	nextfu = &tfs->nextfu;
98 
99 	/* take the format string and break it up into format units */
100 	for (p = fmt;;) {
101 		/* skip leading white space */
102 		for (; isspace(*p); ++p);
103 		if (!*p)
104 			break;
105 
106 		/* allocate a new format unit and link it in */
107 		if ((tfu = calloc(1, sizeof(FU))) == NULL)
108 			err(1, NULL);
109 		*nextfu = tfu;
110 		nextfu = &tfu->nextfu;
111 		tfu->reps = 1;
112 
113 		/* if leading digit, repetition count */
114 		if (isdigit(*p)) {
115 			for (savep = p; isdigit(*p); ++p);
116 			if (!isspace(*p) && *p != '/')
117 				badfmt(fmt);
118 			/* may overwrite either white space or slash */
119 			tfu->reps = atoi(savep);
120 			tfu->flags = F_SETREP;
121 			/* skip trailing white space */
122 			for (++p; isspace(*p); ++p);
123 		}
124 
125 		/* skip slash and trailing white space */
126 		if (*p == '/')
127 			while (isspace(*++p));
128 
129 		/* byte count */
130 		if (isdigit(*p)) {
131 			for (savep = p; isdigit(*p); ++p);
132 			if (!isspace(*p))
133 				badfmt(fmt);
134 			tfu->bcnt = atoi(savep);
135 			/* skip trailing white space */
136 			for (++p; isspace(*p); ++p);
137 		}
138 
139 		/* format */
140 		if (*p != '"')
141 			badfmt(fmt);
142 		for (savep = ++p; *p != '"';)
143 			if (*p++ == 0)
144 				badfmt(fmt);
145 		if (!(tfu->fmt = malloc(p - savep + 1)))
146 			err(1, NULL);
147 		(void) strncpy(tfu->fmt, savep, p - savep);
148 		tfu->fmt[p - savep] = '\0';
149 		escape(tfu->fmt);
150 		p++;
151 	}
152 }
153 
154 static char *spec = ".#-+ 0123456789";
155 
156 int
157 size(fs)
158 	FS *fs;
159 {
160 	register FU *fu;
161 	register int bcnt, cursize;
162 	register unsigned char *fmt;
163 	int prec;
164 
165 	/* figure out the data block size needed for each format unit */
166 	for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) {
167 		if (fu->bcnt) {
168 			cursize += fu->bcnt * fu->reps;
169 			continue;
170 		}
171 		for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) {
172 			if (*fmt != '%')
173 				continue;
174 			/*
175 			 * skip any special chars -- save precision in
176 			 * case it's a %s format.
177 			 */
178 			while (index(spec + 1, *++fmt));
179 			if (*fmt == '.' && isdigit(*++fmt)) {
180 				prec = atoi(fmt);
181 				while (isdigit(*++fmt));
182 			}
183 			switch(*fmt) {
184 			case 'c':
185 				bcnt += 1;
186 				break;
187 			case 'd': case 'i': case 'o': case 'u':
188 			case 'x': case 'X':
189 				bcnt += 4;
190 				break;
191 			case 'e': case 'E': case 'f': case 'g': case 'G':
192 				bcnt += 8;
193 				break;
194 			case 's':
195 				bcnt += prec;
196 				break;
197 			case '_':
198 				switch(*++fmt) {
199 				case 'c': case 'p': case 'u':
200 					bcnt += 1;
201 					break;
202 				}
203 			}
204 		}
205 		cursize += bcnt * fu->reps;
206 	}
207 	return (cursize);
208 }
209 
210 void
211 rewrite(fs)
212 	FS *fs;
213 {
214 	enum { NOTOKAY, USEBCNT, USEPREC } sokay;
215 	register PR *pr, **nextpr;
216 	register FU *fu;
217 	unsigned char *p1, *p2, *fmtp;
218 	char savech, cs[3];
219 	int nconv, prec;
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 					p1[1] = '\0';
318 					badcnt(p1);
319 				}
320 				break;
321 			case 's':
322 				pr->flags = F_STR;
323 				switch(sokay) {
324 				case NOTOKAY:
325 					badsfmt();
326 				case USEBCNT:
327 					pr->bcnt = fu->bcnt;
328 					break;
329 				case USEPREC:
330 					pr->bcnt = prec;
331 					break;
332 				}
333 				break;
334 			case '_':
335 				++p2;
336 				switch(p1[1]) {
337 				case 'A':
338 					endfu = fu;
339 					fu->flags |= F_IGNORE;
340 					/* FALLTHROUGH */
341 				case 'a':
342 					pr->flags = F_ADDRESS;
343 					++p2;
344 					switch(p1[2]) {
345 					case 'd': case 'o': case'x':
346 						cs[0] = 'q';
347 						cs[1] = p1[2];
348 						cs[2] = '\0';
349 						break;
350 					default:
351 						p1[3] = '\0';
352 						badconv(p1);
353 					}
354 					break;
355 				case 'c':
356 					pr->flags = F_C;
357 					/* cs[0] = 'c';	set in conv_c */
358 					goto isint2;
359 				case 'p':
360 					pr->flags = F_P;
361 					cs[0] = 'c';
362 					goto isint2;
363 				case 'u':
364 					pr->flags = F_U;
365 					/* cs[0] = 'c';	set in conv_u */
366 isint2:					switch(fu->bcnt) {
367 					case 0: case 1:
368 						pr->bcnt = 1;
369 						break;
370 					default:
371 						p1[2] = '\0';
372 						badcnt(p1);
373 					}
374 					break;
375 				default:
376 					p1[2] = '\0';
377 					badconv(p1);
378 				}
379 				break;
380 			default:
381 				p1[1] = '\0';
382 				badconv(p1);
383 			}
384 
385 			/*
386 			 * Copy to PR format string, set conversion character
387 			 * pointer, update original.
388 			 */
389 			savech = *p2;
390 			p1[0] = '\0';
391 			if ((pr->fmt = calloc(1, strlen(fmtp) + 2)) == NULL)
392 				err(1, NULL);
393 			(void)strcpy(pr->fmt, fmtp);
394 			(void)strcat(pr->fmt, cs);
395 			*p2 = savech;
396 			pr->cchar = pr->fmt + (p1 - fmtp);
397 			fmtp = p2;
398 
399 			/* Only one conversion character if byte count. */
400 			if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++)
401 	    errx(1, "byte count with multiple conversion characters");
402 		}
403 		/*
404 		 * If format unit byte count not specified, figure it out
405 		 * so can adjust rep count later.
406 		 */
407 		if (!fu->bcnt)
408 			for (pr = fu->nextpr; pr; pr = pr->nextpr)
409 				fu->bcnt += pr->bcnt;
410 	}
411 	/*
412 	 * If the format string interprets any data at all, and it's
413 	 * not the same as the blocksize, and its last format unit
414 	 * interprets any data at all, and has no iteration count,
415 	 * repeat it as necessary.
416 	 *
417 	 * If, rep count is greater than 1, no trailing whitespace
418 	 * gets output from the last iteration of the format unit.
419 	 */
420 	for (fu = fs->nextfu;; fu = fu->nextfu) {
421 		if (!fu->nextfu && fs->bcnt < blocksize &&
422 		    !(fu->flags&F_SETREP) && fu->bcnt)
423 			fu->reps += (blocksize - fs->bcnt) / fu->bcnt;
424 		if (fu->reps > 1) {
425 			for (pr = fu->nextpr;; pr = pr->nextpr)
426 				if (!pr->nextpr)
427 					break;
428 			for (p1 = pr->fmt, p2 = NULL; *p1; ++p1)
429 				p2 = isspace(*p1) ? p1 : NULL;
430 			if (p2)
431 				pr->nospace = p2;
432 		}
433 		if (!fu->nextfu)
434 			break;
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(p1)
448 	register char *p1;
449 {
450 	register 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(s)
491 	char *s;
492 {
493 	errx(1, "%s: bad byte count", s);
494 }
495 
496 void
497 badsfmt()
498 {
499 	errx(1, "%%s: requires a precision or a byte count");
500 }
501 
502 void
503 badfmt(fmt)
504 	char *fmt;
505 {
506 	errx(1, "\"%s\": bad format", fmt);
507 }
508 
509 void
510 badconv(ch)
511 	char *ch;
512 {
513 	errx(1, "%%%s: bad conversion character", ch);
514 }
515