fatbase

portable OpenBSD tools
git clone git://git.2f30.org/fatbase
Log | Files | Refs

parse.c (11606B)


      1 /*	$OpenBSD: parse.c,v 1.17 2009/10/27 23:59:39 deraadt Exp $	*/
      2 /*	$NetBSD: parse.c,v 1.12 2001/12/07 13:37:39 bjh21 Exp $	*/
      3 
      4 /*
      5  * Copyright (c) 1989, 1993
      6  *	The Regents of the University of California.  All rights reserved.
      7  *
      8  * Redistribution and use in source and binary forms, with or without
      9  * modification, are permitted provided that the following conditions
     10  * are met:
     11  * 1. Redistributions of source code must retain the above copyright
     12  *    notice, this list of conditions and the following disclaimer.
     13  * 2. Redistributions in binary form must reproduce the above copyright
     14  *    notice, this list of conditions and the following disclaimer in the
     15  *    documentation and/or other materials provided with the distribution.
     16  * 3. Neither the name of the University nor the names of its contributors
     17  *    may be used to endorse or promote products derived from this software
     18  *    without specific prior written permission.
     19  *
     20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
     21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
     22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
     23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
     24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
     25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
     26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
     27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
     28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
     29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
     30  * SUCH DAMAGE.
     31  */
     32 
     33 #include <sys/types.h>
     34 #include <sys/file.h>
     35 
     36 #include <ctype.h>
     37 #include <err.h>
     38 #include <errno.h>
     39 #include <fcntl.h>
     40 #include <stdio.h>
     41 #include <stdlib.h>
     42 #include <string.h>
     43 
     44 #include "hexdump.h"
     45 #include "util.h"
     46 
     47 FU *endfu;					/* format at end-of-data */
     48 
     49 void
     50 addfile(char *name)
     51 {
     52 	FILE *fp;
     53 	size_t len;
     54 	char *buf, *lbuf, *p;
     55 
     56 	if ((fp = fopen(name, "r")) == NULL)
     57 		err(1, "fopen %s", name);
     58 
     59 	lbuf = NULL;
     60 	while ((buf = fgetln(fp, &len))) {
     61 		if (buf[len - 1] == '\n')
     62 			buf[len - 1] = '\0';
     63 		else {
     64 			/* EOF without EOL, copy and add the NUL */
     65 			if ((lbuf = malloc(len + 1)) == NULL)
     66 				err(1, NULL);
     67 			memcpy(lbuf, buf, len);
     68 			lbuf[len] = '\0';
     69 			buf = lbuf;
     70 		}
     71 		for (p = buf; isspace((unsigned char)*p); ++p);
     72 		if (!*p || *p == '#')
     73 			continue;
     74 		add(p);
     75 	}
     76 	free(lbuf);
     77 	(void)fclose(fp);
     78 }
     79 
     80 void
     81 add(const char *fmt)
     82 {
     83 	const char *p;
     84 	static FS **nextfs;
     85 	FS *tfs;
     86 	FU *tfu, **nextfu;
     87 	const char *savep;
     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((unsigned char)*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((unsigned char)*p)) {
    113 			for (savep = p; isdigit((unsigned char)*p); ++p);
    114 			if (!isspace((unsigned char)*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((unsigned char)*p); ++p);
    121 		}
    122 
    123 		/* skip slash and trailing white space */
    124 		if (*p == '/')
    125 			while (isspace((unsigned char)*++p));
    126 
    127 		/* byte count */
    128 		if (isdigit((unsigned char)*p)) {
    129 			for (savep = p; isdigit((unsigned char)*p); ++p);
    130 			if (!isspace((unsigned char)*p))
    131 				badfmt(fmt);
    132 			tfu->bcnt = atoi(savep);
    133 			/* skip trailing white space */
    134 			for (++p; isspace((unsigned char)*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 const char *spec = ".#-+ 0123456789";
    153 
    154 int
    155 size(FS *fs)
    156 {
    157 	FU *fu;
    158 	int bcnt, cursize;
    159 	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 (*++fmt && strchr(spec + 1, *fmt));
    176 			if (*fmt == '.' && isdigit((unsigned char)*++fmt)) {
    177 				prec = atoi(fmt);
    178 				while (isdigit((unsigned char)*++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 	char *p1, *p2;
    214 	char savech, *fmtp, cs[3];
    215 	int nconv, prec;
    216 	size_t len;
    217 
    218 	nextpr = NULL;
    219 	prec = 0;
    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; *p1 && strchr(spec, *p1); ++p1);
    250 			} else {
    251 				/* Skip any special chars, field width. */
    252 				while (*++p1 && strchr(spec + 1, *p1));
    253 				if (*p1 == '.' &&
    254 				    isdigit((unsigned char)*++p1)) {
    255 					sokay = USEPREC;
    256 					prec = atoi(p1);
    257 					while (isdigit((unsigned char)*++p1))
    258 						continue;
    259 				} else
    260 					sokay = NOTOKAY;
    261 			}
    262 
    263 			p2 = *p1 ? p1 + 1 : p1;	/* 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 			case 'o': case 'u': case 'x': case 'X':
    286 				if (cs[0] == 'd' || cs[0] == 'i')
    287 					pr->flags = F_INT;
    288 				else
    289 					pr->flags = F_UINT;
    290 
    291 				cs[2] = '\0';
    292 				cs[1] = cs[0];
    293 				cs[0] = 'q';
    294 				switch(fu->bcnt) {
    295 				case 0: case 4:
    296 					pr->bcnt = 4;
    297 					break;
    298 				case 1:
    299 					pr->bcnt = 1;
    300 					break;
    301 				case 2:
    302 					pr->bcnt = 2;
    303 					break;
    304 				case 8:
    305 					pr->bcnt = 8;
    306 					break;
    307 				default:
    308 					p1[1] = '\0';
    309 					badcnt(p1);
    310 				}
    311 				break;
    312 			case 'e': case 'E': case 'f': case 'g': case 'G':
    313 				pr->flags = F_DBL;
    314 				switch(fu->bcnt) {
    315 				case 0: case 8:
    316 					pr->bcnt = 8;
    317 					break;
    318 				case 4:
    319 					pr->bcnt = 4;
    320 					break;
    321 				default:
    322 					p1[1] = '\0';
    323 					badcnt(p1);
    324 				}
    325 				break;
    326 			case 's':
    327 				pr->flags = F_STR;
    328 				switch(sokay) {
    329 				case NOTOKAY:
    330 					badsfmt();
    331 				case USEBCNT:
    332 					pr->bcnt = fu->bcnt;
    333 					break;
    334 				case USEPREC:
    335 					pr->bcnt = prec;
    336 					break;
    337 				}
    338 				break;
    339 			case '_':
    340 				++p2;
    341 				switch(p1[1]) {
    342 				case 'A':
    343 					endfu = fu;
    344 					fu->flags |= F_IGNORE;
    345 					/* FALLTHROUGH */
    346 				case 'a':
    347 					pr->flags = F_ADDRESS;
    348 					++p2;
    349 					switch(p1[2]) {
    350 					case 'd': case 'o': case'x':
    351 						cs[0] = 'q';
    352 						cs[1] = p1[2];
    353 						cs[2] = '\0';
    354 						break;
    355 					default:
    356 						if (p1[2])
    357 							p1[3] = '\0';
    358 						badconv(p1);
    359 					}
    360 					break;
    361 				case 'c':
    362 				case 'p':
    363 				case 'u':
    364 					if (p1[1] == 'c') {
    365 						pr->flags = F_C;
    366 						/* cs[0] = 'c';	set in conv_c */
    367 					} else if (p1[1] == 'p') {
    368 						pr->flags = F_P;
    369 						cs[0] = 'c';
    370 					} else {
    371 						pr->flags = F_U;
    372 						/* cs[0] = 'c';	set in conv_u */
    373 					}
    374 
    375 					switch(fu->bcnt) {
    376 					case 0: case 1:
    377 						pr->bcnt = 1;
    378 						break;
    379 					default:
    380 						p1[2] = '\0';
    381 						badcnt(p1);
    382 					}
    383 					break;
    384 				default:
    385 					if (p1[1])
    386 						p1[2] = '\0';
    387 					badconv(p1);
    388 				}
    389 				break;
    390 			default:
    391 				if (cs[0])
    392 					p1[1] = '\0';
    393 				badconv(p1);
    394 			}
    395 
    396 			/*
    397 			 * Copy to PR format string, set conversion character
    398 			 * pointer, update original.
    399 			 */
    400 			savech = *p2;
    401 			p1[0] = '\0';
    402 			len = strlen(fmtp) + strlen(cs) + 1;
    403 			pr->fmt = emalloc(len);
    404 			snprintf(pr->fmt, len, "%s%s", fmtp, cs);
    405 			*p2 = savech;
    406 			pr->cchar = pr->fmt + (p1 - fmtp);
    407 			fmtp = p2;
    408 
    409 			/* Only one conversion character if byte count. */
    410 			if (!(pr->flags&F_ADDRESS) && fu->bcnt && nconv++)
    411 				errx(1,
    412 			    "byte count with multiple conversion characters");
    413 		}
    414 		/*
    415 		 * If format unit byte count not specified, figure it out
    416 		 * so can adjust rep count later.
    417 		 */
    418 		if (!fu->bcnt)
    419 			for (pr = fu->nextpr; pr; pr = pr->nextpr)
    420 				fu->bcnt += pr->bcnt;
    421 	}
    422 	/*
    423 	 * If the format string interprets any data at all, and it's
    424 	 * not the same as the blocksize, and its last format unit
    425 	 * interprets any data at all, and has no iteration count,
    426 	 * repeat it as necessary.
    427 	 *
    428 	 * If, rep count is greater than 1, no trailing whitespace
    429 	 * gets output from the last iteration of the format unit.
    430 	 */
    431 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
    432 		if (!fu->nextfu && fs->bcnt < blocksize &&
    433 		    !(fu->flags&F_SETREP) && fu->bcnt)
    434 			fu->reps += (blocksize - fs->bcnt) / fu->bcnt;
    435 		if (fu->reps > 1) {
    436 			if (!fu->nextpr)
    437 				break;
    438 			for (pr = fu->nextpr;; pr = pr->nextpr)
    439 				if (!pr->nextpr)
    440 					break;
    441 			for (p1 = pr->fmt, p2 = NULL; *p1; ++p1)
    442 				p2 = isspace((unsigned char)*p1) ? p1 : NULL;
    443 			if (p2)
    444 				pr->nospace = p2;
    445 		}
    446 	}
    447 #ifdef DEBUG
    448 	for (fu = fs->nextfu; fu; fu = fu->nextfu) {
    449 		(void)printf("fmt:");
    450 		for (pr = fu->nextpr; pr; pr = pr->nextpr)
    451 			(void)printf(" {%s}", pr->fmt);
    452 		(void)printf("\n");
    453 	}
    454 #endif
    455 }
    456 
    457 void
    458 escape(char *p1)
    459 {
    460 	char *p2;
    461 
    462 	/* alphabetic escape sequences have to be done in place */
    463 	for (p2 = p1;; ++p1, ++p2) {
    464 		if (!*p1) {
    465 			*p2 = *p1;
    466 			break;
    467 		}
    468 		if (*p1 == '\\') {
    469 			switch(*++p1) {
    470 			case '\0':
    471 				*p2++ = '\\';
    472 				*p2 = '\0';
    473 				return;	/* incomplete escape sequence */
    474 			case 'a':
    475 			     /* *p2 = '\a'; */
    476 				*p2 = '\007';
    477 				break;
    478 			case 'b':
    479 				*p2 = '\b';
    480 				break;
    481 			case 'f':
    482 				*p2 = '\f';
    483 				break;
    484 			case 'n':
    485 				*p2 = '\n';
    486 				break;
    487 			case 'r':
    488 				*p2 = '\r';
    489 				break;
    490 			case 't':
    491 				*p2 = '\t';
    492 				break;
    493 			case 'v':
    494 				*p2 = '\v';
    495 				break;
    496 			default:
    497 				*p2 = *p1;
    498 				break;
    499 			}
    500 		} else
    501 			*p2 = *p1;
    502 	}
    503 }
    504 
    505 void
    506 badcnt(char *s)
    507 {
    508 	errx(1, "%s: bad byte count", s);
    509 }
    510 
    511 void
    512 badsfmt(void)
    513 {
    514 	errx(1, "%%s: requires a precision or a byte count");
    515 }
    516 
    517 void
    518 badfmt(const char *fmt)
    519 {
    520 	errx(1, "\"%s\": bad format", fmt);
    521 }
    522 
    523 void
    524 badconv(char *ch)
    525 {
    526 	errx(1, "%%%s: bad conversion character", ch);
    527 }