xref: /freebsd/usr.bin/hexdump/parse.c (revision afe61c15161c324a7af299a9b8457aba5afc92db)
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 static char sccsid[] = "@(#)parse.c	8.1 (Berkeley) 6/6/93";
36 #endif /* not lint */
37 
38 #include <sys/types.h>
39 
40 #include <errno.h>
41 #include <fcntl.h>
42 #include <stdio.h>
43 #include <stdlib.h>
44 #include <ctype.h>
45 #include <string.h>
46 #include "hexdump.h"
47 
48 FU *endfu;					/* format at end-of-data */
49 
50 void
51 addfile(name)
52 	char *name;
53 {
54 	register char *p;
55 	FILE *fp;
56 	int ch;
57 	char buf[2048 + 1];
58 
59 	if ((fp = fopen(name, "r")) == NULL)
60 		err("%s: %s\n", name, strerror(errno));
61 	while (fgets(buf, sizeof(buf), fp)) {
62 		if (!(p = index(buf, '\n'))) {
63 			(void)fprintf(stderr, "hexdump: line too long.\n");
64 			while ((ch = getchar()) != '\n' && ch != EOF);
65 			continue;
66 		}
67 		*p = '\0';
68 		for (p = buf; *p && isspace(*p); ++p);
69 		if (!*p || *p == '#')
70 			continue;
71 		add(p);
72 	}
73 	(void)fclose(fp);
74 }
75 
76 void
77 add(fmt)
78 	char *fmt;
79 {
80 	register char *p;
81 	static FS **nextfs;
82 	FS *tfs;
83 	FU *tfu, **nextfu;
84 	char *savep;
85 
86 	/* start new linked list of format units */
87 	tfs = emalloc(sizeof(FS));
88 	if (!fshead)
89 		fshead = tfs;
90 	else
91 		*nextfs = tfs;
92 	nextfs = &tfs->nextfs;
93 	nextfu = &tfs->nextfu;
94 
95 	/* take the format string and break it up into format units */
96 	for (p = fmt;;) {
97 		/* skip leading white space */
98 		for (; isspace(*p); ++p);
99 		if (!*p)
100 			break;
101 
102 		/* allocate a new format unit and link it in */
103 		tfu = emalloc(sizeof(FU));
104 		*nextfu = tfu;
105 		nextfu = &tfu->nextfu;
106 		tfu->reps = 1;
107 
108 		/* if leading digit, repetition count */
109 		if (isdigit(*p)) {
110 			for (savep = p; isdigit(*p); ++p);
111 			if (!isspace(*p) && *p != '/')
112 				badfmt(fmt);
113 			/* may overwrite either white space or slash */
114 			tfu->reps = atoi(savep);
115 			tfu->flags = F_SETREP;
116 			/* skip trailing white space */
117 			for (++p; isspace(*p); ++p);
118 		}
119 
120 		/* skip slash and trailing white space */
121 		if (*p == '/')
122 			while (isspace(*++p));
123 
124 		/* byte count */
125 		if (isdigit(*p)) {
126 			for (savep = p; isdigit(*p); ++p);
127 			if (!isspace(*p))
128 				badfmt(fmt);
129 			tfu->bcnt = atoi(savep);
130 			/* skip trailing white space */
131 			for (++p; isspace(*p); ++p);
132 		}
133 
134 		/* format */
135 		if (*p != '"')
136 			badfmt(fmt);
137 		for (savep = ++p; *p != '"';)
138 			if (*p++ == 0)
139 				badfmt(fmt);
140 		if (!(tfu->fmt = malloc(p - savep + 1)))
141 			nomem();
142 		(void) strncpy(tfu->fmt, savep, p - savep);
143 		tfu->fmt[p - savep] = '\0';
144 		escape(tfu->fmt);
145 		p++;
146 	}
147 }
148 
149 static char *spec = ".#-+ 0123456789";
150 
151 int
152 size(fs)
153 	FS *fs;
154 {
155 	register FU *fu;
156 	register int bcnt, cursize;
157 	register char *fmt;
158 	int prec;
159 
160 	/* figure out the data block size needed for each format unit */
161 	for (cursize = 0, fu = fs->nextfu; fu; fu = fu->nextfu) {
162 		if (fu->bcnt) {
163 			cursize += fu->bcnt * fu->reps;
164 			continue;
165 		}
166 		for (bcnt = prec = 0, fmt = fu->fmt; *fmt; ++fmt) {
167 			if (*fmt != '%')
168 				continue;
169 			/*
170 			 * skip any special chars -- save precision in
171 			 * case it's a %s format.
172 			 */
173 			while (index(spec + 1, *++fmt));
174 			if (*fmt == '.' && isdigit(*++fmt)) {
175 				prec = atoi(fmt);
176 				while (isdigit(*++fmt));
177 			}
178 			switch(*fmt) {
179 			case 'c':
180 				bcnt += 1;
181 				break;
182 			case 'd': case 'i': case 'o': case 'u':
183 			case 'x': case 'X':
184 				bcnt += 4;
185 				break;
186 			case 'e': case 'E': case 'f': case 'g': case 'G':
187 				bcnt += 8;
188 				break;
189 			case 's':
190 				bcnt += prec;
191 				break;
192 			case '_':
193 				switch(*++fmt) {
194 				case 'c': case 'p': case 'u':
195 					bcnt += 1;
196 					break;
197 				}
198 			}
199 		}
200 		cursize += bcnt * fu->reps;
201 	}
202 	return (cursize);
203 }
204 
205 void
206 rewrite(fs)
207 	FS *fs;
208 {
209 	enum { NOTOKAY, USEBCNT, USEPREC } sokay;
210 	register PR *pr, **nextpr;
211 	register FU *fu;
212 	register char *p1, *p2;
213 	char savech, *fmtp, cs[3];
214 	int nconv, prec;
215 
216 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
217 		/*
218 		 * Break each format unit into print units; each conversion
219 		 * character gets its own.
220 		 */
221 		for (nconv = 0, fmtp = fu->fmt; *fmtp; nextpr = &pr->nextpr) {
222 			pr = emalloc(sizeof(PR));
223 			if (!fu->nextpr)
224 				fu->nextpr = pr;
225 			else
226 				*nextpr = pr;
227 
228 			/* Skip preceding text and up to the next % sign. */
229 			for (p1 = fmtp; *p1 && *p1 != '%'; ++p1);
230 
231 			/* Only text in the string. */
232 			if (!*p1) {
233 				pr->fmt = fmtp;
234 				pr->flags = F_TEXT;
235 				break;
236 			}
237 
238 			/*
239 			 * Get precision for %s -- if have a byte count, don't
240 			 * need it.
241 			 */
242 			if (fu->bcnt) {
243 				sokay = USEBCNT;
244 				/* Skip to conversion character. */
245 				for (++p1; index(spec, *p1); ++p1);
246 			} else {
247 				/* Skip any special chars, field width. */
248 				while (index(spec + 1, *++p1));
249 				if (*p1 == '.' && isdigit(*++p1)) {
250 					sokay = USEPREC;
251 					prec = atoi(p1);
252 					while (isdigit(*++p1));
253 				} else
254 					sokay = NOTOKAY;
255 			}
256 
257 			p2 = p1 + 1;		/* Set end pointer. */
258 			cs[0] = *p1;		/* Set conversion string. */
259 			cs[1] = '\0';
260 
261 			/*
262 			 * Figure out the byte count for each conversion;
263 			 * rewrite the format as necessary, set up blank-
264 			 * padding for end of data.
265 			 */
266 			switch(cs[0]) {
267 			case 'c':
268 				pr->flags = F_CHAR;
269 				switch(fu->bcnt) {
270 				case 0: case 1:
271 					pr->bcnt = 1;
272 					break;
273 				default:
274 					p1[1] = '\0';
275 					badcnt(p1);
276 				}
277 				break;
278 			case 'd': case 'i':
279 				pr->flags = F_INT;
280 				goto isint;
281 			case 'o': case 'u': case 'x': case 'X':
282 				pr->flags = F_UINT;
283 isint:				cs[2] = '\0';
284 				cs[1] = cs[0];
285 				cs[0] = 'q';
286 				switch(fu->bcnt) {
287 				case 0: case 4:
288 					pr->bcnt = 4;
289 					break;
290 				case 1:
291 					pr->bcnt = 1;
292 					break;
293 				case 2:
294 					pr->bcnt = 2;
295 					break;
296 				default:
297 					p1[1] = '\0';
298 					badcnt(p1);
299 				}
300 				break;
301 			case 'e': case 'E': case 'f': case 'g': case 'G':
302 				pr->flags = F_DBL;
303 				switch(fu->bcnt) {
304 				case 0: case 8:
305 					pr->bcnt = 8;
306 					break;
307 				case 4:
308 					pr->bcnt = 4;
309 					break;
310 				default:
311 					p1[1] = '\0';
312 					badcnt(p1);
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 			pr->fmt = emalloc(strlen(fmtp) + 2);
386 			(void)strcpy(pr->fmt, fmtp);
387 			(void)strcat(pr->fmt, cs);
388 			*p2 = savech;
389 			pr->cchar = pr->fmt + (p1 - fmtp);
390 			fmtp = p2;
391 
392 			/* Only one conversion character if byte count. */
393 			if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++)
394 	    err("byte count with multiple conversion characters");
395 		}
396 		/*
397 		 * If format unit byte count not specified, figure it out
398 		 * so can adjust rep count later.
399 		 */
400 		if (!fu->bcnt)
401 			for (pr = fu->nextpr; pr; pr = pr->nextpr)
402 				fu->bcnt += pr->bcnt;
403 	}
404 	/*
405 	 * If the format string interprets any data at all, and it's
406 	 * not the same as the blocksize, and its last format unit
407 	 * interprets any data at all, and has no iteration count,
408 	 * repeat it as necessary.
409 	 *
410 	 * If, rep count is greater than 1, no trailing whitespace
411 	 * gets output from the last iteration of the format unit.
412 	 */
413 	for (fu = fs->nextfu;; fu = fu->nextfu) {
414 		if (!fu->nextfu && fs->bcnt < blocksize &&
415 		    !(fu->flags&F_SETREP) && fu->bcnt)
416 			fu->reps += (blocksize - fs->bcnt) / fu->bcnt;
417 		if (fu->reps > 1) {
418 			for (pr = fu->nextpr;; pr = pr->nextpr)
419 				if (!pr->nextpr)
420 					break;
421 			for (p1 = pr->fmt, p2 = NULL; *p1; ++p1)
422 				p2 = isspace(*p1) ? p1 : NULL;
423 			if (p2)
424 				pr->nospace = p2;
425 		}
426 		if (!fu->nextfu)
427 			break;
428 	}
429 #ifdef DEBUG
430 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
431 		(void)printf("fmt:");
432 		for (pr = fu->nextpr; pr; pr = pr->nextpr)
433 			(void)printf(" {%s}", pr->fmt);
434 		(void)printf("\n");
435 	}
436 #endif
437 }
438 
439 void
440 escape(p1)
441 	register char *p1;
442 {
443 	register 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
483 badcnt(s)
484 	char *s;
485 {
486 	err("%s: bad byte count", s);
487 }
488 
489 void
490 badsfmt()
491 {
492 	err("%%s: requires a precision or a byte count\n");
493 }
494 
495 void
496 badfmt(fmt)
497 	char *fmt;
498 {
499 	err("\"%s\": bad format\n", fmt);
500 }
501 
502 void
503 badconv(ch)
504 	char *ch;
505 {
506 	err("%%%s: bad conversion character\n", ch);
507 }
508