xref: /netbsd-src/lib/libc/gen/unvis.c (revision 56bb44cae5b13a6b74792381ba1e6d930b26aa67)
1 /*	$NetBSD: unvis.c,v 1.33 2011/02/27 01:53:22 christos 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 1.33 2011/02/27 01:53:22 christos Exp $");
38 #endif
39 #endif /* LIBC_SCCS and not lint */
40 
41 #include "namespace.h"
42 #include <sys/types.h>
43 
44 #include <assert.h>
45 #include <ctype.h>
46 #include <stdint.h>
47 #include <stdio.h>
48 #include <vis.h>
49 
50 #ifdef __weak_alias
51 __weak_alias(strunvis,_strunvis)
52 #endif
53 
54 #if !HAVE_VIS
55 /*
56  * decode driven by state machine
57  */
58 #define	S_GROUND	0	/* haven't seen escape char */
59 #define	S_START		1	/* start decoding special sequence */
60 #define	S_META		2	/* metachar started (M) */
61 #define	S_META1		3	/* metachar more, regular char (-) */
62 #define	S_CTRL		4	/* control char started (^) */
63 #define	S_OCTAL2	5	/* octal digit 2 */
64 #define	S_OCTAL3	6	/* octal digit 3 */
65 #define	S_HEX1		7	/* http hex digit */
66 #define	S_HEX2		8	/* http hex digit 2 */
67 #define S_MIME1		9	/* mime hex digit 1 */
68 #define S_MIME2		10	/* mime hex digit 2 */
69 #define S_EATCRNL	11	/* mime eating CRNL */
70 #define S_AMP		12	/* seen & */
71 #define S_NUMBER	13	/* collecting number */
72 #define S_STRING	14	/* collecting string */
73 
74 #define	isoctal(c)	(((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
75 #define xtod(c)		(isdigit(c) ? (c - '0') : ((tolower(c) - 'a') + 10))
76 #define XTOD(c)		(isdigit(c) ? (c - '0') : ((c - 'A') + 10))
77 
78 /*
79  * RFC 1866
80  */
81 static const struct nv {
82 	const char *name;
83 	uint8_t value;
84 } nv[] = {
85 	{ "AElig",	198 }, /* capital AE diphthong (ligature)  */
86 	{ "Aacute",	193 }, /* capital A, acute accent  */
87 	{ "Acirc",	194 }, /* capital A, circumflex accent  */
88 	{ "Agrave",	192 }, /* capital A, grave accent  */
89 	{ "Aring",	197 }, /* capital A, ring  */
90 	{ "Atilde",	195 }, /* capital A, tilde  */
91 	{ "Auml",	196 }, /* capital A, dieresis or umlaut mark  */
92 	{ "Ccedil",	199 }, /* capital C, cedilla  */
93 	{ "ETH",	208 }, /* capital Eth, Icelandic  */
94 	{ "Eacute",	201 }, /* capital E, acute accent  */
95 	{ "Ecirc",	202 }, /* capital E, circumflex accent  */
96 	{ "Egrave",	200 }, /* capital E, grave accent  */
97 	{ "Euml",	203 }, /* capital E, dieresis or umlaut mark  */
98 	{ "Iacute",	205 }, /* capital I, acute accent  */
99 	{ "Icirc",	206 }, /* capital I, circumflex accent  */
100 	{ "Igrave",	204 }, /* capital I, grave accent  */
101 	{ "Iuml",	207 }, /* capital I, dieresis or umlaut mark  */
102 	{ "Ntilde",	209 }, /* capital N, tilde  */
103 	{ "Oacute",	211 }, /* capital O, acute accent  */
104 	{ "Ocirc",	212 }, /* capital O, circumflex accent  */
105 	{ "Ograve",	210 }, /* capital O, grave accent  */
106 	{ "Oslash",	216 }, /* capital O, slash  */
107 	{ "Otilde",	213 }, /* capital O, tilde  */
108 	{ "Ouml",	214 }, /* capital O, dieresis or umlaut mark  */
109 	{ "THORN",	222 }, /* capital THORN, Icelandic  */
110 	{ "Uacute",	218 }, /* capital U, acute accent  */
111 	{ "Ucirc",	219 }, /* capital U, circumflex accent  */
112 	{ "Ugrave",	217 }, /* capital U, grave accent  */
113 	{ "Uuml",	220 }, /* capital U, dieresis or umlaut mark  */
114 	{ "Yacute",	221 }, /* capital Y, acute accent  */
115 	{ "aacute",	225 }, /* small a, acute accent  */
116 	{ "acirc",	226 }, /* small a, circumflex accent  */
117 	{ "acute",	180 }, /* acute accent  */
118 	{ "aelig",	230 }, /* small ae diphthong (ligature)  */
119 	{ "agrave",	224 }, /* small a, grave accent  */
120 	{ "amp",	 38 }, /* ampersand  */
121 	{ "aring",	229 }, /* small a, ring  */
122 	{ "atilde",	227 }, /* small a, tilde  */
123 	{ "auml",	228 }, /* small a, dieresis or umlaut mark  */
124 	{ "brvbar",	166 }, /* broken (vertical) bar  */
125 	{ "ccedil",	231 }, /* small c, cedilla  */
126 	{ "cedil",	184 }, /* cedilla  */
127 	{ "cent",	162 }, /* cent sign  */
128 	{ "copy",	169 }, /* copyright sign  */
129 	{ "curren",	164 }, /* general currency sign  */
130 	{ "deg",	176 }, /* degree sign  */
131 	{ "divide",	247 }, /* divide sign  */
132 	{ "eacute",	233 }, /* small e, acute accent  */
133 	{ "ecirc",	234 }, /* small e, circumflex accent  */
134 	{ "egrave",	232 }, /* small e, grave accent  */
135 	{ "eth",	240 }, /* small eth, Icelandic  */
136 	{ "euml",	235 }, /* small e, dieresis or umlaut mark  */
137 	{ "frac12",	189 }, /* fraction one-half  */
138 	{ "frac14",	188 }, /* fraction one-quarter  */
139 	{ "frac34",	190 }, /* fraction three-quarters  */
140 	{ "gt",		 62 }, /* greater than  */
141 	{ "iacute",	237 }, /* small i, acute accent  */
142 	{ "icirc",	238 }, /* small i, circumflex accent  */
143 	{ "iexcl",	161 }, /* inverted exclamation mark  */
144 	{ "igrave",	236 }, /* small i, grave accent  */
145 	{ "iquest",	191 }, /* inverted question mark  */
146 	{ "iuml",	239 }, /* small i, dieresis or umlaut mark  */
147 	{ "laquo",	171 }, /* angle quotation mark, left  */
148 	{ "lt",		 60 }, /* less than  */
149 	{ "macr",	175 }, /* macron  */
150 	{ "micro",	181 }, /* micro sign  */
151 	{ "middot",	183 }, /* middle dot  */
152 	{ "nbsp",	160 }, /* no-break space  */
153 	{ "not",	172 }, /* not sign  */
154 	{ "ntilde",	241 }, /* small n, tilde  */
155 	{ "oacute",	243 }, /* small o, acute accent  */
156 	{ "ocirc",	244 }, /* small o, circumflex accent  */
157 	{ "ograve",	242 }, /* small o, grave accent  */
158 	{ "ordf",	170 }, /* ordinal indicator, feminine  */
159 	{ "ordm",	186 }, /* ordinal indicator, masculine  */
160 	{ "oslash",	248 }, /* small o, slash  */
161 	{ "otilde",	245 }, /* small o, tilde  */
162 	{ "ouml",	246 }, /* small o, dieresis or umlaut mark  */
163 	{ "para",	182 }, /* pilcrow (paragraph sign)  */
164 	{ "plusmn",	177 }, /* plus-or-minus sign  */
165 	{ "pound",	163 }, /* pound sterling sign  */
166 	{ "quot",	 34 }, /* double quote  */
167 	{ "raquo",	187 }, /* angle quotation mark, right  */
168 	{ "reg",	174 }, /* registered sign  */
169 	{ "sect",	167 }, /* section sign  */
170 	{ "shy",	173 }, /* soft hyphen  */
171 	{ "sup1",	185 }, /* superscript one  */
172 	{ "sup2",	178 }, /* superscript two  */
173 	{ "sup3",	179 }, /* superscript three  */
174 	{ "szlig",	223 }, /* small sharp s, German (sz ligature)  */
175 	{ "thorn",	254 }, /* small thorn, Icelandic  */
176 	{ "times",	215 }, /* multiply sign  */
177 	{ "uacute",	250 }, /* small u, acute accent  */
178 	{ "ucirc",	251 }, /* small u, circumflex accent  */
179 	{ "ugrave",	249 }, /* small u, grave accent  */
180 	{ "uml",	168 }, /* umlaut (dieresis)  */
181 	{ "uuml",	252 }, /* small u, dieresis or umlaut mark  */
182 	{ "yacute",	253 }, /* small y, acute accent  */
183 	{ "yen",	165 }, /* yen sign  */
184 	{ "yuml",	255 }, /* small y, dieresis or umlaut mark  */
185 };
186 
187 /*
188  * unvis - decode characters previously encoded by vis
189  */
190 int
191 unvis(char *cp, int c, int *astate, int flag)
192 {
193 	unsigned char uc = (unsigned char)c;
194 	unsigned char st, ia, is, lc;
195 
196 /*
197  * Bottom 8 bits of astate hold the state machine state.
198  * Top 8 bits hold the current character in the http 1866 nv string decoding
199  */
200 #define GS(a)		((a) & 0xff)
201 #define SS(a, b)	(((uint32_t)(a) << 24) | (b))
202 #define GI(a)		((uint32_t)(a) >> 24)
203 
204 	_DIAGASSERT(cp != NULL);
205 	_DIAGASSERT(astate != NULL);
206 	st = GS(*astate);
207 
208 	if (flag & UNVIS_END) {
209 		switch (st) {
210 		case S_OCTAL2:
211 		case S_OCTAL3:
212 		case S_HEX2:
213 			*astate = SS(0, S_GROUND);
214 			return UNVIS_VALID;
215 		case S_GROUND:
216 			return UNVIS_NOCHAR;
217 		default:
218 			return UNVIS_SYNBAD;
219 		}
220 	}
221 
222 	switch (st) {
223 
224 	case S_GROUND:
225 		*cp = 0;
226 		if ((flag & VIS_NOESCAPE) == 0 && c == '\\') {
227 			*astate = SS(0, S_START);
228 			return UNVIS_NOCHAR;
229 		}
230 		if ((flag & VIS_HTTP1808) && c == '%') {
231 			*astate = SS(0, S_HEX1);
232 			return UNVIS_NOCHAR;
233 		}
234 		if ((flag & VIS_HTTP1866) && c == '&') {
235 			*astate = SS(0, S_AMP);
236 			return UNVIS_NOCHAR;
237 		}
238 		if ((flag & VIS_MIMESTYLE) && c == '=') {
239 			*astate = SS(0, S_MIME1);
240 			return UNVIS_NOCHAR;
241 		}
242 		*cp = c;
243 		return UNVIS_VALID;
244 
245 	case S_START:
246 		switch(c) {
247 		case '\\':
248 			*cp = c;
249 			*astate = SS(0, S_GROUND);
250 			return UNVIS_VALID;
251 		case '0': case '1': case '2': case '3':
252 		case '4': case '5': case '6': case '7':
253 			*cp = (c - '0');
254 			*astate = SS(0, S_OCTAL2);
255 			return UNVIS_NOCHAR;
256 		case 'M':
257 			*cp = (char)0200;
258 			*astate = SS(0, S_META);
259 			return UNVIS_NOCHAR;
260 		case '^':
261 			*astate = SS(0, S_CTRL);
262 			return UNVIS_NOCHAR;
263 		case 'n':
264 			*cp = '\n';
265 			*astate = SS(0, S_GROUND);
266 			return UNVIS_VALID;
267 		case 'r':
268 			*cp = '\r';
269 			*astate = SS(0, S_GROUND);
270 			return UNVIS_VALID;
271 		case 'b':
272 			*cp = '\b';
273 			*astate = SS(0, S_GROUND);
274 			return UNVIS_VALID;
275 		case 'a':
276 			*cp = '\007';
277 			*astate = SS(0, S_GROUND);
278 			return UNVIS_VALID;
279 		case 'v':
280 			*cp = '\v';
281 			*astate = SS(0, S_GROUND);
282 			return UNVIS_VALID;
283 		case 't':
284 			*cp = '\t';
285 			*astate = SS(0, S_GROUND);
286 			return UNVIS_VALID;
287 		case 'f':
288 			*cp = '\f';
289 			*astate = SS(0, S_GROUND);
290 			return UNVIS_VALID;
291 		case 's':
292 			*cp = ' ';
293 			*astate = SS(0, S_GROUND);
294 			return UNVIS_VALID;
295 		case 'E':
296 			*cp = '\033';
297 			*astate = SS(0, S_GROUND);
298 			return UNVIS_VALID;
299 		case '\n':
300 			/*
301 			 * hidden newline
302 			 */
303 			*astate = SS(0, S_GROUND);
304 			return UNVIS_NOCHAR;
305 		case '$':
306 			/*
307 			 * hidden marker
308 			 */
309 			*astate = SS(0, S_GROUND);
310 			return UNVIS_NOCHAR;
311 		}
312 		goto bad;
313 
314 	case S_META:
315 		if (c == '-')
316 			*astate = SS(0, S_META1);
317 		else if (c == '^')
318 			*astate = SS(0, S_CTRL);
319 		else
320 			goto bad;
321 		return UNVIS_NOCHAR;
322 
323 	case S_META1:
324 		*astate = SS(0, S_GROUND);
325 		*cp |= c;
326 		return UNVIS_VALID;
327 
328 	case S_CTRL:
329 		if (c == '?')
330 			*cp |= 0177;
331 		else
332 			*cp |= c & 037;
333 		*astate = SS(0, S_GROUND);
334 		return UNVIS_VALID;
335 
336 	case S_OCTAL2:	/* second possible octal digit */
337 		if (isoctal(uc)) {
338 			/*
339 			 * yes - and maybe a third
340 			 */
341 			*cp = (*cp << 3) + (c - '0');
342 			*astate = SS(0, S_OCTAL3);
343 			return UNVIS_NOCHAR;
344 		}
345 		/*
346 		 * no - done with current sequence, push back passed char
347 		 */
348 		*astate = SS(0, S_GROUND);
349 		return UNVIS_VALIDPUSH;
350 
351 	case S_OCTAL3:	/* third possible octal digit */
352 		*astate = SS(0, S_GROUND);
353 		if (isoctal(uc)) {
354 			*cp = (*cp << 3) + (c - '0');
355 			return UNVIS_VALID;
356 		}
357 		/*
358 		 * we were done, push back passed char
359 		 */
360 		return UNVIS_VALIDPUSH;
361 
362 	case S_HEX1:
363 		if (isxdigit(uc)) {
364 			*cp = xtod(uc);
365 			*astate = SS(0, S_HEX2);
366 			return UNVIS_NOCHAR;
367 		}
368 		/*
369 		 * no - done with current sequence, push back passed char
370 		 */
371 		*astate = SS(0, S_GROUND);
372 		return UNVIS_VALIDPUSH;
373 
374 	case S_HEX2:
375 		*astate = S_GROUND;
376 		if (isxdigit(uc)) {
377 			*cp = xtod(uc) | (*cp << 4);
378 			return UNVIS_VALID;
379 		}
380 		return UNVIS_VALIDPUSH;
381 
382 	case S_MIME1:
383 		if (uc == '\n' || uc == '\r') {
384 			*astate = SS(0, S_EATCRNL);
385 			return UNVIS_NOCHAR;
386 		}
387 		if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
388 			*cp = XTOD(uc);
389 			*astate = SS(0, S_MIME2);
390 			return UNVIS_NOCHAR;
391 		}
392 		goto bad;
393 
394 	case S_MIME2:
395 		if (isxdigit(uc) && (isdigit(uc) || isupper(uc))) {
396 			*astate = SS(0, S_GROUND);
397 			*cp = XTOD(uc) | (*cp << 4);
398 			return UNVIS_VALID;
399 		}
400 		goto bad;
401 
402 	case S_EATCRNL:
403 		switch (uc) {
404 		case '\r':
405 		case '\n':
406 			return UNVIS_NOCHAR;
407 		case '=':
408 			*astate = SS(0, S_MIME1);
409 			return UNVIS_NOCHAR;
410 		default:
411 			*cp = uc;
412 			*astate = SS(0, S_GROUND);
413 			return UNVIS_VALID;
414 		}
415 
416 	case S_AMP:
417 		*cp = 0;
418 		if (uc == '#') {
419 			*astate = SS(0, S_NUMBER);
420 			return UNVIS_NOCHAR;
421 		}
422 		*astate = SS(0, S_STRING);
423 		/*FALLTHROUGH*/
424 
425 	case S_STRING:
426 		ia = *cp;		/* index in the array */
427 		is = GI(*astate);	/* index in the string */
428 		lc = is == 0 ? 0 : nv[ia].name[is - 1];	/* last character */
429 
430 		if (uc == ';')
431 			uc = '\0';
432 
433 		for (; ia < __arraycount(nv); ia++) {
434 			if (is != 0 && nv[ia].name[is - 1] != lc)
435 				goto bad;
436 			if (nv[ia].name[is] == uc)
437 				break;
438 		}
439 
440 		if (*cp == __arraycount(nv))
441 			goto bad;
442 
443 		if (uc != 0) {
444 			*cp = ia;
445 			*astate = SS(is + 1, S_STRING);
446 			return UNVIS_NOCHAR;
447 		}
448 
449 		*cp = nv[ia].value;
450 		*astate = SS(0, S_GROUND);
451 		return UNVIS_VALID;
452 
453 	case S_NUMBER:
454 		if (uc == ';')
455 			return UNVIS_VALID;
456 		if (!isdigit(uc))
457 			goto bad;
458 		*cp += (*cp * 10) + uc - '0';
459 		return UNVIS_NOCHAR;
460 
461 	default:
462 	bad:
463 		/*
464 		 * decoder in unknown state - (probably uninitialized)
465 		 */
466 		*astate = SS(0, S_GROUND);
467 		return UNVIS_SYNBAD;
468 	}
469 }
470 
471 /*
472  * strunvis - decode src into dst
473  *
474  *	Number of chars decoded into dst is returned, -1 on error.
475  *	Dst is null terminated.
476  */
477 
478 int
479 strunvisx(char *dst, const char *src, int flag)
480 {
481 	char c;
482 	char *start = dst;
483 	int state = 0;
484 
485 	_DIAGASSERT(src != NULL);
486 	_DIAGASSERT(dst != NULL);
487 
488 	while ((c = *src++) != '\0') {
489  again:
490 		switch (unvis(dst, c, &state, flag)) {
491 		case UNVIS_VALID:
492 			dst++;
493 			break;
494 		case UNVIS_VALIDPUSH:
495 			dst++;
496 			goto again;
497 		case 0:
498 		case UNVIS_NOCHAR:
499 			break;
500 		default:
501 			return (-1);
502 		}
503 	}
504 	if (unvis(dst, c, &state, UNVIS_END) == UNVIS_VALID)
505 		dst++;
506 	*dst = '\0';
507 	return (int)(dst - start);
508 }
509 
510 int
511 strunvis(char *dst, const char *src)
512 {
513 	return strunvisx(dst, src, 0);
514 }
515 #endif
516