xref: /freebsd/contrib/libc-vis/unvis.c (revision 3416500aef140042c64bc149cb1ec6620483bc44)
1 /*	$NetBSD: unvis.c,v 1.44 2014/09/26 15:43:36 roy Exp $	*/
2 
3 /*-
4  * Copyright (c) 1989, 1993
5  *	The Regents of the University of California.  All rights reserved.
6  *
7  * Redistribution and use in source and binary forms, with or without
8  * modification, are permitted provided that the following conditions
9  * are met:
10  * 1. Redistributions of source code must retain the above copyright
11  *    notice, this list of conditions and the following disclaimer.
12  * 2. Redistributions in binary form must reproduce the above copyright
13  *    notice, this list of conditions and the following disclaimer in the
14  *    documentation and/or other materials provided with the distribution.
15  * 3. Neither the name of the University nor the names of its contributors
16  *    may be used to endorse or promote products derived from this software
17  *    without specific prior written permission.
18  *
19  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
20  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
23  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29  * SUCH DAMAGE.
30  */
31 
32 #include <sys/cdefs.h>
33 #if defined(LIBC_SCCS) && !defined(lint)
34 #if 0
35 static char sccsid[] = "@(#)unvis.c	8.1 (Berkeley) 6/4/93";
36 #else
37 __RCSID("$NetBSD: unvis.c,v 1.44 2014/09/26 15:43:36 roy Exp $");
38 #endif
39 #endif /* LIBC_SCCS and not lint */
40 __FBSDID("$FreeBSD$");
41 
42 #include "namespace.h"
43 #include <sys/types.h>
44 
45 #include <assert.h>
46 #include <ctype.h>
47 #include <stdint.h>
48 #include <stdio.h>
49 #include <errno.h>
50 #include <vis.h>
51 
52 #define	_DIAGASSERT(x)	assert(x)
53 
54 /*
55  * Return the number of elements in a statically-allocated array,
56  * __x.
57  */
58 #define	__arraycount(__x)	(sizeof(__x) / sizeof(__x[0]))
59 
60 #ifdef __weak_alias
61 __weak_alias(strnunvisx,_strnunvisx)
62 #endif
63 
64 #if !HAVE_VIS
65 /*
66  * decode driven by state machine
67  */
68 #define	S_GROUND	0	/* haven't seen escape char */
69 #define	S_START		1	/* start decoding special sequence */
70 #define	S_META		2	/* metachar started (M) */
71 #define	S_META1		3	/* metachar more, regular char (-) */
72 #define	S_CTRL		4	/* control char started (^) */
73 #define	S_OCTAL2	5	/* octal digit 2 */
74 #define	S_OCTAL3	6	/* octal digit 3 */
75 #define	S_HEX		7	/* mandatory hex digit */
76 #define	S_HEX1		8	/* http hex digit */
77 #define	S_HEX2		9	/* http hex digit 2 */
78 #define	S_MIME1		10	/* mime hex digit 1 */
79 #define	S_MIME2		11	/* mime hex digit 2 */
80 #define	S_EATCRNL	12	/* mime eating CRNL */
81 #define	S_AMP		13	/* seen & */
82 #define	S_NUMBER	14	/* collecting number */
83 #define	S_STRING	15	/* collecting string */
84 
85 #define	isoctal(c)	(((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
86 #define	xtod(c)		(isdigit(c) ? (c - '0') : ((tolower(c) - 'a') + 10))
87 #define	XTOD(c)		(isdigit(c) ? (c - '0') : ((c - 'A') + 10))
88 
89 /*
90  * RFC 1866
91  */
92 static const struct nv {
93 	char name[7];
94 	uint8_t value;
95 } nv[] = {
96 	{ "AElig",	198 }, /* capital AE diphthong (ligature)  */
97 	{ "Aacute",	193 }, /* capital A, acute accent  */
98 	{ "Acirc",	194 }, /* capital A, circumflex accent  */
99 	{ "Agrave",	192 }, /* capital A, grave accent  */
100 	{ "Aring",	197 }, /* capital A, ring  */
101 	{ "Atilde",	195 }, /* capital A, tilde  */
102 	{ "Auml",	196 }, /* capital A, dieresis or umlaut mark  */
103 	{ "Ccedil",	199 }, /* capital C, cedilla  */
104 	{ "ETH",	208 }, /* capital Eth, Icelandic  */
105 	{ "Eacute",	201 }, /* capital E, acute accent  */
106 	{ "Ecirc",	202 }, /* capital E, circumflex accent  */
107 	{ "Egrave",	200 }, /* capital E, grave accent  */
108 	{ "Euml",	203 }, /* capital E, dieresis or umlaut mark  */
109 	{ "Iacute",	205 }, /* capital I, acute accent  */
110 	{ "Icirc",	206 }, /* capital I, circumflex accent  */
111 	{ "Igrave",	204 }, /* capital I, grave accent  */
112 	{ "Iuml",	207 }, /* capital I, dieresis or umlaut mark  */
113 	{ "Ntilde",	209 }, /* capital N, tilde  */
114 	{ "Oacute",	211 }, /* capital O, acute accent  */
115 	{ "Ocirc",	212 }, /* capital O, circumflex accent  */
116 	{ "Ograve",	210 }, /* capital O, grave accent  */
117 	{ "Oslash",	216 }, /* capital O, slash  */
118 	{ "Otilde",	213 }, /* capital O, tilde  */
119 	{ "Ouml",	214 }, /* capital O, dieresis or umlaut mark  */
120 	{ "THORN",	222 }, /* capital THORN, Icelandic  */
121 	{ "Uacute",	218 }, /* capital U, acute accent  */
122 	{ "Ucirc",	219 }, /* capital U, circumflex accent  */
123 	{ "Ugrave",	217 }, /* capital U, grave accent  */
124 	{ "Uuml",	220 }, /* capital U, dieresis or umlaut mark  */
125 	{ "Yacute",	221 }, /* capital Y, acute accent  */
126 	{ "aacute",	225 }, /* small a, acute accent  */
127 	{ "acirc",	226 }, /* small a, circumflex accent  */
128 	{ "acute",	180 }, /* acute accent  */
129 	{ "aelig",	230 }, /* small ae diphthong (ligature)  */
130 	{ "agrave",	224 }, /* small a, grave accent  */
131 	{ "amp",	 38 }, /* ampersand  */
132 	{ "aring",	229 }, /* small a, ring  */
133 	{ "atilde",	227 }, /* small a, tilde  */
134 	{ "auml",	228 }, /* small a, dieresis or umlaut mark  */
135 	{ "brvbar",	166 }, /* broken (vertical) bar  */
136 	{ "ccedil",	231 }, /* small c, cedilla  */
137 	{ "cedil",	184 }, /* cedilla  */
138 	{ "cent",	162 }, /* cent sign  */
139 	{ "copy",	169 }, /* copyright sign  */
140 	{ "curren",	164 }, /* general currency sign  */
141 	{ "deg",	176 }, /* degree sign  */
142 	{ "divide",	247 }, /* divide sign  */
143 	{ "eacute",	233 }, /* small e, acute accent  */
144 	{ "ecirc",	234 }, /* small e, circumflex accent  */
145 	{ "egrave",	232 }, /* small e, grave accent  */
146 	{ "eth",	240 }, /* small eth, Icelandic  */
147 	{ "euml",	235 }, /* small e, dieresis or umlaut mark  */
148 	{ "frac12",	189 }, /* fraction one-half  */
149 	{ "frac14",	188 }, /* fraction one-quarter  */
150 	{ "frac34",	190 }, /* fraction three-quarters  */
151 	{ "gt",		 62 }, /* greater than  */
152 	{ "iacute",	237 }, /* small i, acute accent  */
153 	{ "icirc",	238 }, /* small i, circumflex accent  */
154 	{ "iexcl",	161 }, /* inverted exclamation mark  */
155 	{ "igrave",	236 }, /* small i, grave accent  */
156 	{ "iquest",	191 }, /* inverted question mark  */
157 	{ "iuml",	239 }, /* small i, dieresis or umlaut mark  */
158 	{ "laquo",	171 }, /* angle quotation mark, left  */
159 	{ "lt",		 60 }, /* less than  */
160 	{ "macr",	175 }, /* macron  */
161 	{ "micro",	181 }, /* micro sign  */
162 	{ "middot",	183 }, /* middle dot  */
163 	{ "nbsp",	160 }, /* no-break space  */
164 	{ "not",	172 }, /* not sign  */
165 	{ "ntilde",	241 }, /* small n, tilde  */
166 	{ "oacute",	243 }, /* small o, acute accent  */
167 	{ "ocirc",	244 }, /* small o, circumflex accent  */
168 	{ "ograve",	242 }, /* small o, grave accent  */
169 	{ "ordf",	170 }, /* ordinal indicator, feminine  */
170 	{ "ordm",	186 }, /* ordinal indicator, masculine  */
171 	{ "oslash",	248 }, /* small o, slash  */
172 	{ "otilde",	245 }, /* small o, tilde  */
173 	{ "ouml",	246 }, /* small o, dieresis or umlaut mark  */
174 	{ "para",	182 }, /* pilcrow (paragraph sign)  */
175 	{ "plusmn",	177 }, /* plus-or-minus sign  */
176 	{ "pound",	163 }, /* pound sterling sign  */
177 	{ "quot",	 34 }, /* double quote  */
178 	{ "raquo",	187 }, /* angle quotation mark, right  */
179 	{ "reg",	174 }, /* registered sign  */
180 	{ "sect",	167 }, /* section sign  */
181 	{ "shy",	173 }, /* soft hyphen  */
182 	{ "sup1",	185 }, /* superscript one  */
183 	{ "sup2",	178 }, /* superscript two  */
184 	{ "sup3",	179 }, /* superscript three  */
185 	{ "szlig",	223 }, /* small sharp s, German (sz ligature)  */
186 	{ "thorn",	254 }, /* small thorn, Icelandic  */
187 	{ "times",	215 }, /* multiply sign  */
188 	{ "uacute",	250 }, /* small u, acute accent  */
189 	{ "ucirc",	251 }, /* small u, circumflex accent  */
190 	{ "ugrave",	249 }, /* small u, grave accent  */
191 	{ "uml",	168 }, /* umlaut (dieresis)  */
192 	{ "uuml",	252 }, /* small u, dieresis or umlaut mark  */
193 	{ "yacute",	253 }, /* small y, acute accent  */
194 	{ "yen",	165 }, /* yen sign  */
195 	{ "yuml",	255 }, /* small y, dieresis or umlaut mark  */
196 };
197 
198 /*
199  * unvis - decode characters previously encoded by vis
200  */
201 int
202 unvis(char *cp, int c, int *astate, int flag)
203 {
204 	unsigned char uc = (unsigned char)c;
205 	unsigned char st, ia, is, lc;
206 
207 /*
208  * Bottom 8 bits of astate hold the state machine state.
209  * Top 8 bits hold the current character in the http 1866 nv string decoding
210  */
211 #define GS(a)		((a) & 0xff)
212 #define SS(a, b)	(((uint32_t)(a) << 24) | (b))
213 #define GI(a)		((uint32_t)(a) >> 24)
214 
215 	_DIAGASSERT(cp != NULL);
216 	_DIAGASSERT(astate != NULL);
217 	st = GS(*astate);
218 
219 	if (flag & UNVIS_END) {
220 		switch (st) {
221 		case S_OCTAL2:
222 		case S_OCTAL3:
223 		case S_HEX2:
224 			*astate = SS(0, S_GROUND);
225 			return UNVIS_VALID;
226 		case S_GROUND:
227 			return UNVIS_NOCHAR;
228 		default:
229 			return UNVIS_SYNBAD;
230 		}
231 	}
232 
233 	switch (st) {
234 
235 	case S_GROUND:
236 		*cp = 0;
237 		if ((flag & VIS_NOESCAPE) == 0 && c == '\\') {
238 			*astate = SS(0, S_START);
239 			return UNVIS_NOCHAR;
240 		}
241 		if ((flag & VIS_HTTP1808) && c == '%') {
242 			*astate = SS(0, S_HEX1);
243 			return UNVIS_NOCHAR;
244 		}
245 		if ((flag & VIS_HTTP1866) && c == '&') {
246 			*astate = SS(0, S_AMP);
247 			return UNVIS_NOCHAR;
248 		}
249 		if ((flag & VIS_MIMESTYLE) && c == '=') {
250 			*astate = SS(0, S_MIME1);
251 			return UNVIS_NOCHAR;
252 		}
253 		*cp = c;
254 		return UNVIS_VALID;
255 
256 	case S_START:
257 		switch(c) {
258 		case '\\':
259 			*cp = c;
260 			*astate = SS(0, S_GROUND);
261 			return UNVIS_VALID;
262 		case '0': case '1': case '2': case '3':
263 		case '4': case '5': case '6': case '7':
264 			*cp = (c - '0');
265 			*astate = SS(0, S_OCTAL2);
266 			return UNVIS_NOCHAR;
267 		case 'M':
268 			*cp = (char)0200;
269 			*astate = SS(0, S_META);
270 			return UNVIS_NOCHAR;
271 		case '^':
272 			*astate = SS(0, S_CTRL);
273 			return UNVIS_NOCHAR;
274 		case 'n':
275 			*cp = '\n';
276 			*astate = SS(0, S_GROUND);
277 			return UNVIS_VALID;
278 		case 'r':
279 			*cp = '\r';
280 			*astate = SS(0, S_GROUND);
281 			return UNVIS_VALID;
282 		case 'b':
283 			*cp = '\b';
284 			*astate = SS(0, S_GROUND);
285 			return UNVIS_VALID;
286 		case 'a':
287 			*cp = '\007';
288 			*astate = SS(0, S_GROUND);
289 			return UNVIS_VALID;
290 		case 'v':
291 			*cp = '\v';
292 			*astate = SS(0, S_GROUND);
293 			return UNVIS_VALID;
294 		case 't':
295 			*cp = '\t';
296 			*astate = SS(0, S_GROUND);
297 			return UNVIS_VALID;
298 		case 'f':
299 			*cp = '\f';
300 			*astate = SS(0, S_GROUND);
301 			return UNVIS_VALID;
302 		case 's':
303 			*cp = ' ';
304 			*astate = SS(0, S_GROUND);
305 			return UNVIS_VALID;
306 		case 'E':
307 			*cp = '\033';
308 			*astate = SS(0, S_GROUND);
309 			return UNVIS_VALID;
310 		case 'x':
311 			*astate = SS(0, S_HEX);
312 			return UNVIS_NOCHAR;
313 		case '\n':
314 			/*
315 			 * hidden newline
316 			 */
317 			*astate = SS(0, S_GROUND);
318 			return UNVIS_NOCHAR;
319 		case '$':
320 			/*
321 			 * hidden marker
322 			 */
323 			*astate = SS(0, S_GROUND);
324 			return UNVIS_NOCHAR;
325 		default:
326 			if (isgraph(c)) {
327 				*cp = c;
328 				*astate = SS(0, S_GROUND);
329 				return UNVIS_VALID;
330 			}
331 		}
332 		goto bad;
333 
334 	case S_META:
335 		if (c == '-')
336 			*astate = SS(0, S_META1);
337 		else if (c == '^')
338 			*astate = SS(0, S_CTRL);
339 		else
340 			goto bad;
341 		return UNVIS_NOCHAR;
342 
343 	case S_META1:
344 		*astate = SS(0, S_GROUND);
345 		*cp |= c;
346 		return UNVIS_VALID;
347 
348 	case S_CTRL:
349 		if (c == '?')
350 			*cp |= 0177;
351 		else
352 			*cp |= c & 037;
353 		*astate = SS(0, S_GROUND);
354 		return UNVIS_VALID;
355 
356 	case S_OCTAL2:	/* second possible octal digit */
357 		if (isoctal(uc)) {
358 			/*
359 			 * yes - and maybe a third
360 			 */
361 			*cp = (*cp << 3) + (c - '0');
362 			*astate = SS(0, S_OCTAL3);
363 			return UNVIS_NOCHAR;
364 		}
365 		/*
366 		 * no - done with current sequence, push back passed char
367 		 */
368 		*astate = SS(0, S_GROUND);
369 		return UNVIS_VALIDPUSH;
370 
371 	case S_OCTAL3:	/* third possible octal digit */
372 		*astate = SS(0, S_GROUND);
373 		if (isoctal(uc)) {
374 			*cp = (*cp << 3) + (c - '0');
375 			return UNVIS_VALID;
376 		}
377 		/*
378 		 * we were done, push back passed char
379 		 */
380 		return UNVIS_VALIDPUSH;
381 
382 	case S_HEX:
383 		if (!isxdigit(uc))
384 			goto bad;
385 		/*FALLTHROUGH*/
386 	case S_HEX1:
387 		if (isxdigit(uc)) {
388 			*cp = xtod(uc);
389 			*astate = SS(0, S_HEX2);
390 			return UNVIS_NOCHAR;
391 		}
392 		/*
393 		 * no - done with current sequence, push back passed char
394 		 */
395 		*astate = SS(0, S_GROUND);
396 		return UNVIS_VALIDPUSH;
397 
398 	case S_HEX2:
399 		*astate = S_GROUND;
400 		if (isxdigit(uc)) {
401 			*cp = xtod(uc) | (*cp << 4);
402 			return UNVIS_VALID;
403 		}
404 		return UNVIS_VALIDPUSH;
405 
406 	case S_MIME1:
407 		if (uc == '\n' || uc == '\r') {
408 			*astate = SS(0, S_EATCRNL);
409 			return UNVIS_NOCHAR;
410 		}
411 		if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
412 			*cp = XTOD(uc);
413 			*astate = SS(0, S_MIME2);
414 			return UNVIS_NOCHAR;
415 		}
416 		goto bad;
417 
418 	case S_MIME2:
419 		if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
420 			*astate = SS(0, S_GROUND);
421 			*cp = XTOD(uc) | (*cp << 4);
422 			return UNVIS_VALID;
423 		}
424 		goto bad;
425 
426 	case S_EATCRNL:
427 		switch (uc) {
428 		case '\r':
429 		case '\n':
430 			return UNVIS_NOCHAR;
431 		case '=':
432 			*astate = SS(0, S_MIME1);
433 			return UNVIS_NOCHAR;
434 		default:
435 			*cp = uc;
436 			*astate = SS(0, S_GROUND);
437 			return UNVIS_VALID;
438 		}
439 
440 	case S_AMP:
441 		*cp = 0;
442 		if (uc == '#') {
443 			*astate = SS(0, S_NUMBER);
444 			return UNVIS_NOCHAR;
445 		}
446 		*astate = SS(0, S_STRING);
447 		/*FALLTHROUGH*/
448 
449 	case S_STRING:
450 		ia = *cp;		/* index in the array */
451 		is = GI(*astate);	/* index in the string */
452 		lc = is == 0 ? 0 : nv[ia].name[is - 1];	/* last character */
453 
454 		if (uc == ';')
455 			uc = '\0';
456 
457 		for (; ia < __arraycount(nv); ia++) {
458 			if (is != 0 && nv[ia].name[is - 1] != lc)
459 				goto bad;
460 			if (nv[ia].name[is] == uc)
461 				break;
462 		}
463 
464 		if (ia == __arraycount(nv))
465 			goto bad;
466 
467 		if (uc != 0) {
468 			*cp = ia;
469 			*astate = SS(is + 1, S_STRING);
470 			return UNVIS_NOCHAR;
471 		}
472 
473 		*cp = nv[ia].value;
474 		*astate = SS(0, S_GROUND);
475 		return UNVIS_VALID;
476 
477 	case S_NUMBER:
478 		if (uc == ';')
479 			return UNVIS_VALID;
480 		if (!isdigit(uc))
481 			goto bad;
482 		*cp += (*cp * 10) + uc - '0';
483 		return UNVIS_NOCHAR;
484 
485 	default:
486 	bad:
487 		/*
488 		 * decoder in unknown state - (probably uninitialized)
489 		 */
490 		*astate = SS(0, S_GROUND);
491 		return UNVIS_SYNBAD;
492 	}
493 }
494 
495 /*
496  * strnunvisx - decode src into dst
497  *
498  *	Number of chars decoded into dst is returned, -1 on error.
499  *	Dst is null terminated.
500  */
501 
502 int
503 strnunvisx(char *dst, size_t dlen, const char *src, int flag)
504 {
505 	char c;
506 	char t = '\0', *start = dst;
507 	int state = 0;
508 
509 	_DIAGASSERT(src != NULL);
510 	_DIAGASSERT(dst != NULL);
511 #define CHECKSPACE() \
512 	do { \
513 		if (dlen-- == 0) { \
514 			errno = ENOSPC; \
515 			return -1; \
516 		} \
517 	} while (/*CONSTCOND*/0)
518 
519 	while ((c = *src++) != '\0') {
520  again:
521 		switch (unvis(&t, c, &state, flag)) {
522 		case UNVIS_VALID:
523 			CHECKSPACE();
524 			*dst++ = t;
525 			break;
526 		case UNVIS_VALIDPUSH:
527 			CHECKSPACE();
528 			*dst++ = t;
529 			goto again;
530 		case 0:
531 		case UNVIS_NOCHAR:
532 			break;
533 		case UNVIS_SYNBAD:
534 			errno = EINVAL;
535 			return -1;
536 		default:
537 			_DIAGASSERT(/*CONSTCOND*/0);
538 			errno = EINVAL;
539 			return -1;
540 		}
541 	}
542 	if (unvis(&t, c, &state, UNVIS_END) == UNVIS_VALID) {
543 		CHECKSPACE();
544 		*dst++ = t;
545 	}
546 	CHECKSPACE();
547 	*dst = '\0';
548 	return (int)(dst - start);
549 }
550 
551 int
552 strunvisx(char *dst, const char *src, int flag)
553 {
554 	return strnunvisx(dst, (size_t)~0, src, flag);
555 }
556 
557 int
558 strunvis(char *dst, const char *src)
559 {
560 	return strnunvisx(dst, (size_t)~0, src, 0);
561 }
562 
563 int
564 strnunvis(char *dst, size_t dlen, const char *src)
565 {
566 	return strnunvisx(dst, dlen, src, 0);
567 }
568 #endif
569