xref: /freebsd/usr.bin/hexdump/parse.c (revision ce834215a70ff69e7e222827437116eee2f9ac6f)
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 	"$Id$";
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 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 	register char *p;
85 	static FS **nextfs;
86 	FS *tfs;
87 	FU *tfu, **nextfu;
88 	char *savep;
89 
90 	/* start new linked list of format units */
91 	tfs = emalloc(sizeof(FS));
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 		tfu = emalloc(sizeof(FU));
108 		*nextfu = tfu;
109 		nextfu = &tfu->nextfu;
110 		tfu->reps = 1;
111 
112 		/* if leading digit, repetition count */
113 		if (isdigit(*p)) {
114 			for (savep = p; isdigit(*p); ++p);
115 			if (!isspace(*p) && *p != '/')
116 				badfmt(fmt);
117 			/* may overwrite either white space or slash */
118 			tfu->reps = atoi(savep);
119 			tfu->flags = F_SETREP;
120 			/* skip trailing white space */
121 			for (++p; isspace(*p); ++p);
122 		}
123 
124 		/* skip slash and trailing white space */
125 		if (*p == '/')
126 			while (isspace(*++p));
127 
128 		/* byte count */
129 		if (isdigit(*p)) {
130 			for (savep = p; isdigit(*p); ++p);
131 			if (!isspace(*p))
132 				badfmt(fmt);
133 			tfu->bcnt = atoi(savep);
134 			/* skip trailing white space */
135 			for (++p; isspace(*p); ++p);
136 		}
137 
138 		/* format */
139 		if (*p != '"')
140 			badfmt(fmt);
141 		for (savep = ++p; *p != '"';)
142 			if (*p++ == 0)
143 				badfmt(fmt);
144 		if (!(tfu->fmt = malloc(p - savep + 1)))
145 			nomem();
146 		(void) strncpy(tfu->fmt, savep, p - savep);
147 		tfu->fmt[p - savep] = '\0';
148 		escape(tfu->fmt);
149 		p++;
150 	}
151 }
152 
153 static char *spec = ".#-+ 0123456789";
154 
155 int
156 size(fs)
157 	FS *fs;
158 {
159 	register FU *fu;
160 	register int bcnt, cursize;
161 	register char *fmt;
162 	int prec;
163 
164 	/* figure out the data block size needed for each format unit */
165 	for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) {
166 		if (fu->bcnt) {
167 			cursize += fu->bcnt * fu->reps;
168 			continue;
169 		}
170 		for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) {
171 			if (*fmt != '%')
172 				continue;
173 			/*
174 			 * skip any special chars -- save precision in
175 			 * case it's a %s format.
176 			 */
177 			while (index(spec + 1, *++fmt));
178 			if (*fmt == '.' && isdigit(*++fmt)) {
179 				prec = atoi(fmt);
180 				while (isdigit(*++fmt));
181 			}
182 			switch(*fmt) {
183 			case 'c':
184 				bcnt += 1;
185 				break;
186 			case 'd': case 'i': case 'o': case 'u':
187 			case 'x': case 'X':
188 				bcnt += 4;
189 				break;
190 			case 'e': case 'E': case 'f': case 'g': case 'G':
191 				bcnt += 8;
192 				break;
193 			case 's':
194 				bcnt += prec;
195 				break;
196 			case '_':
197 				switch(*++fmt) {
198 				case 'c': case 'p': case 'u':
199 					bcnt += 1;
200 					break;
201 				}
202 			}
203 		}
204 		cursize += bcnt * fu->reps;
205 	}
206 	return (cursize);
207 }
208 
209 void
210 rewrite(fs)
211 	FS *fs;
212 {
213 	enum { NOTOKAY, USEBCNT, USEPREC } sokay;
214 	register PR *pr, **nextpr;
215 	register FU *fu;
216 	register char *p1, *p2;
217 	char savech, *fmtp, cs[3];
218 	int nconv, prec;
219 
220 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
221 		/*
222 		 * Break each format unit into print units; each conversion
223 		 * character gets its own.
224 		 */
225 		for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) {
226 			pr = emalloc(sizeof(PR));
227 			if (!fu->nextpr)
228 				fu->nextpr = pr;
229 			else
230 				*nextpr = pr;
231 
232 			/* Skip preceding text and up to the next % sign. */
233 			for (p1 = fmtp; *p1 && *p1 != '%'; ++p1);
234 
235 			/* Only text in the string. */
236 			if (!*p1) {
237 				pr->fmt = fmtp;
238 				pr->flags = F_TEXT;
239 				break;
240 			}
241 
242 			/*
243 			 * Get precision for %s -- if have a byte count, don't
244 			 * need it.
245 			 */
246 			if (fu->bcnt) {
247 				sokay = USEBCNT;
248 				/* Skip to conversion character. */
249 				for (++p1; index(spec, *p1); ++p1);
250 			} else {
251 				/* Skip any special chars, field width. */
252 				while (index(spec + 1, *++p1));
253 				if (*p1 == '.' && isdigit(*++p1)) {
254 					sokay = USEPREC;
255 					prec = atoi(p1);
256 					while (isdigit(*++p1));
257 				} else
258 					sokay = NOTOKAY;
259 			}
260 
261 			p2 = p1 + 1;		/* Set end pointer. */
262 			cs[0] = *p1;		/* Set conversion string. */
263 			cs[1] = '\0';
264 
265 			/*
266 			 * Figure out the byte count for each conversion;
267 			 * rewrite the format as necessary, set up blank-
268 			 * padding for end of data.
269 			 */
270 			switch(cs[0]) {
271 			case 'c':
272 				pr->flags = F_CHAR;
273 				switch(fu->bcnt) {
274 				case 0: case 1:
275 					pr->bcnt = 1;
276 					break;
277 				default:
278 					p1[1] = '\0';
279 					badcnt(p1);
280 				}
281 				break;
282 			case 'd': case 'i':
283 				pr->flags = F_INT;
284 				goto isint;
285 			case 'o': case 'u': case 'x': case 'X':
286 				pr->flags = F_UINT;
287 isint:				cs[2] = '\0';
288 				cs[1] = cs[0];
289 				cs[0] = 'q';
290 				switch(fu->bcnt) {
291 				case 0: case 4:
292 					pr->bcnt = 4;
293 					break;
294 				case 1:
295 					pr->bcnt = 1;
296 					break;
297 				case 2:
298 					pr->bcnt = 2;
299 					break;
300 				default:
301 					p1[1] = '\0';
302 					badcnt(p1);
303 				}
304 				break;
305 			case 'e': case 'E': case 'f': case 'g': case 'G':
306 				pr->flags = F_DBL;
307 				switch(fu->bcnt) {
308 				case 0: case 8:
309 					pr->bcnt = 8;
310 					break;
311 				case 4:
312 					pr->bcnt = 4;
313 					break;
314 				default:
315 					p1[1] = '\0';
316 					badcnt(p1);
317 				}
318 				break;
319 			case 's':
320 				pr->flags = F_STR;
321 				switch(sokay) {
322 				case NOTOKAY:
323 					badsfmt();
324 				case USEBCNT:
325 					pr->bcnt = fu->bcnt;
326 					break;
327 				case USEPREC:
328 					pr->bcnt = prec;
329 					break;
330 				}
331 				break;
332 			case '_':
333 				++p2;
334 				switch(p1[1]) {
335 				case 'A':
336 					endfu = fu;
337 					fu->flags |= F_IGNORE;
338 					/* FALLTHROUGH */
339 				case 'a':
340 					pr->flags = F_ADDRESS;
341 					++p2;
342 					switch(p1[2]) {
343 					case 'd': case 'o': case'x':
344 						cs[0] = 'q';
345 						cs[1] = p1[2];
346 						cs[2] = '\0';
347 						break;
348 					default:
349 						p1[3] = '\0';
350 						badconv(p1);
351 					}
352 					break;
353 				case 'c':
354 					pr->flags = F_C;
355 					/* cs[0] = 'c';	set in conv_c */
356 					goto isint2;
357 				case 'p':
358 					pr->flags = F_P;
359 					cs[0] = 'c';
360 					goto isint2;
361 				case 'u':
362 					pr->flags = F_U;
363 					/* cs[0] = 'c';	set in conv_u */
364 isint2:					switch(fu->bcnt) {
365 					case 0: case 1:
366 						pr->bcnt = 1;
367 						break;
368 					default:
369 						p1[2] = '\0';
370 						badcnt(p1);
371 					}
372 					break;
373 				default:
374 					p1[2] = '\0';
375 					badconv(p1);
376 				}
377 				break;
378 			default:
379 				p1[1] = '\0';
380 				badconv(p1);
381 			}
382 
383 			/*
384 			 * Copy to PR format string, set conversion character
385 			 * pointer, update original.
386 			 */
387 			savech = *p2;
388 			p1[0] = '\0';
389 			pr->fmt = emalloc(strlen(fmtp) + 2);
390 			(void)strcpy(pr->fmt, fmtp);
391 			(void)strcat(pr->fmt, cs);
392 			*p2 = savech;
393 			pr->cchar = pr->fmt + (p1 - fmtp);
394 			fmtp = p2;
395 
396 			/* Only one conversion character if byte count. */
397 			if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++)
398 	    errx(1, "byte count with multiple conversion characters");
399 		}
400 		/*
401 		 * If format unit byte count not specified, figure it out
402 		 * so can adjust rep count later.
403 		 */
404 		if (!fu->bcnt)
405 			for (pr = fu->nextpr; pr; pr = pr->nextpr)
406 				fu->bcnt += pr->bcnt;
407 	}
408 	/*
409 	 * If the format string interprets any data at all, and it's
410 	 * not the same as the blocksize, and its last format unit
411 	 * interprets any data at all, and has no iteration count,
412 	 * repeat it as necessary.
413 	 *
414 	 * If, rep count is greater than 1, no trailing whitespace
415 	 * gets output from the last iteration of the format unit.
416 	 */
417 	for (fu = fs->nextfu;; fu = fu->nextfu) {
418 		if (!fu->nextfu && fs->bcnt < blocksize &&
419 		    !(fu->flags&F_SETREP) && fu->bcnt)
420 			fu->reps += (blocksize - fs->bcnt) / fu->bcnt;
421 		if (fu->reps > 1) {
422 			for (pr = fu->nextpr;; pr = pr->nextpr)
423 				if (!pr->nextpr)
424 					break;
425 			for (p1 = pr->fmt, p2 = NULL; *p1; ++p1)
426 				p2 = isspace(*p1) ? p1 : NULL;
427 			if (p2)
428 				pr->nospace = p2;
429 		}
430 		if (!fu->nextfu)
431 			break;
432 	}
433 #ifdef DEBUG
434 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
435 		(void)printf("fmt:");
436 		for (pr = fu->nextpr; pr; pr = pr->nextpr)
437 			(void)printf(" {%s}", pr->fmt);
438 		(void)printf("\n");
439 	}
440 #endif
441 }
442 
443 void
444 escape(p1)
445 	register char *p1;
446 {
447 	register char *p2;
448 
449 	/* alphabetic escape sequences have to be done in place */
450 	for (p2 = p1;; ++p1, ++p2) {
451 		if (!*p1) {
452 			*p2 = *p1;
453 			break;
454 		}
455 		if (*p1 == '\\')
456 			switch(*++p1) {
457 			case 'a':
458 			     /* *p2 = '\a'; */
459 				*p2 = '\007';
460 				break;
461 			case 'b':
462 				*p2 = '\b';
463 				break;
464 			case 'f':
465 				*p2 = '\f';
466 				break;
467 			case 'n':
468 				*p2 = '\n';
469 				break;
470 			case 'r':
471 				*p2 = '\r';
472 				break;
473 			case 't':
474 				*p2 = '\t';
475 				break;
476 			case 'v':
477 				*p2 = '\v';
478 				break;
479 			default:
480 				*p2 = *p1;
481 				break;
482 			}
483 	}
484 }
485 
486 void
487 badcnt(s)
488 	char *s;
489 {
490 	errx(1, "%s: bad byte count", s);
491 }
492 
493 void
494 badsfmt()
495 {
496 	errx(1, "%%s: requires a precision or a byte count");
497 }
498 
499 void
500 badfmt(fmt)
501 	char *fmt;
502 {
503 	errx(1, "\"%s\": bad format", fmt);
504 }
505 
506 void
507 badconv(ch)
508 	char *ch;
509 {
510 	errx(1, "%%%s: bad conversion character", ch);
511 }
512