unvis.c revision 1.26 1 /* $NetBSD: unvis.c,v 1.26 2005/05/16 11:40:16 lukem 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 1.26 2005/05/16 11:40:16 lukem Exp $");
38 #endif
39 #endif /* LIBC_SCCS and not lint */
40
41 #define __LIBC12_SOURCE__
42
43 #include "namespace.h"
44 #include <sys/types.h>
45
46 #include <assert.h>
47 #include <ctype.h>
48 #include <stdio.h>
49 #include <vis.h>
50
51 #ifdef __weak_alias
52 __weak_alias(strunvis,_strunvis)
53 __weak_alias(unvis,_unvis)
54 #endif
55
56 #ifdef __warn_references
57 __warn_references(unvis,
58 "warning: reference to compatibility unvis(); include <vis.h> for correct reference")
59 #endif
60
61 #if !HAVE_VIS
62 /*
63 * decode driven by state machine
64 */
65 #define S_GROUND 0 /* haven't seen escape char */
66 #define S_START 1 /* start decoding special sequence */
67 #define S_META 2 /* metachar started (M) */
68 #define S_META1 3 /* metachar more, regular char (-) */
69 #define S_CTRL 4 /* control char started (^) */
70 #define S_OCTAL2 5 /* octal digit 2 */
71 #define S_OCTAL3 6 /* octal digit 3 */
72 #define S_HEX1 7 /* hex digit */
73 #define S_HEX2 8 /* hex digit 2 */
74
75 #define isoctal(c) (((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
76 #define xtod(c) (isdigit(c) ? (c - '0') : ((tolower(c) - 'a') + 10))
77
78 int
79 unvis(cp, c, astate, flag)
80 char *cp;
81 int c;
82 int *astate, flag;
83 {
84 return __unvis13(cp, c, astate, flag);
85 }
86
87 /*
88 * unvis - decode characters previously encoded by vis
89 */
90 int
91 __unvis13(cp, c, astate, flag)
92 char *cp;
93 int c;
94 int *astate, flag;
95 {
96 unsigned char uc = (unsigned char)c;
97
98 _DIAGASSERT(cp != NULL);
99 _DIAGASSERT(astate != NULL);
100
101 if (flag & UNVIS_END) {
102 if (*astate == S_OCTAL2 || *astate == S_OCTAL3
103 || *astate == S_HEX2) {
104 *astate = S_GROUND;
105 return (UNVIS_VALID);
106 }
107 return (*astate == S_GROUND ? UNVIS_NOCHAR : UNVIS_SYNBAD);
108 }
109
110 switch (*astate) {
111
112 case S_GROUND:
113 *cp = 0;
114 if (c == '\\') {
115 *astate = S_START;
116 return (0);
117 }
118 if ((flag & VIS_HTTPSTYLE) && c == '%') {
119 *astate = S_HEX1;
120 return (0);
121 }
122 *cp = c;
123 return (UNVIS_VALID);
124
125 case S_START:
126 switch(c) {
127 case '\\':
128 *cp = c;
129 *astate = S_GROUND;
130 return (UNVIS_VALID);
131 case '0': case '1': case '2': case '3':
132 case '4': case '5': case '6': case '7':
133 *cp = (c - '0');
134 *astate = S_OCTAL2;
135 return (0);
136 case 'M':
137 *cp = (char)0200;
138 *astate = S_META;
139 return (0);
140 case '^':
141 *astate = S_CTRL;
142 return (0);
143 case 'n':
144 *cp = '\n';
145 *astate = S_GROUND;
146 return (UNVIS_VALID);
147 case 'r':
148 *cp = '\r';
149 *astate = S_GROUND;
150 return (UNVIS_VALID);
151 case 'b':
152 *cp = '\b';
153 *astate = S_GROUND;
154 return (UNVIS_VALID);
155 case 'a':
156 *cp = '\007';
157 *astate = S_GROUND;
158 return (UNVIS_VALID);
159 case 'v':
160 *cp = '\v';
161 *astate = S_GROUND;
162 return (UNVIS_VALID);
163 case 't':
164 *cp = '\t';
165 *astate = S_GROUND;
166 return (UNVIS_VALID);
167 case 'f':
168 *cp = '\f';
169 *astate = S_GROUND;
170 return (UNVIS_VALID);
171 case 's':
172 *cp = ' ';
173 *astate = S_GROUND;
174 return (UNVIS_VALID);
175 case 'E':
176 *cp = '\033';
177 *astate = S_GROUND;
178 return (UNVIS_VALID);
179 case '\n':
180 /*
181 * hidden newline
182 */
183 *astate = S_GROUND;
184 return (UNVIS_NOCHAR);
185 case '$':
186 /*
187 * hidden marker
188 */
189 *astate = S_GROUND;
190 return (UNVIS_NOCHAR);
191 }
192 *astate = S_GROUND;
193 return (UNVIS_SYNBAD);
194
195 case S_META:
196 if (c == '-')
197 *astate = S_META1;
198 else if (c == '^')
199 *astate = S_CTRL;
200 else {
201 *astate = S_GROUND;
202 return (UNVIS_SYNBAD);
203 }
204 return (0);
205
206 case S_META1:
207 *astate = S_GROUND;
208 *cp |= c;
209 return (UNVIS_VALID);
210
211 case S_CTRL:
212 if (c == '?')
213 *cp |= 0177;
214 else
215 *cp |= c & 037;
216 *astate = S_GROUND;
217 return (UNVIS_VALID);
218
219 case S_OCTAL2: /* second possible octal digit */
220 if (isoctal(uc)) {
221 /*
222 * yes - and maybe a third
223 */
224 *cp = (*cp << 3) + (c - '0');
225 *astate = S_OCTAL3;
226 return (0);
227 }
228 /*
229 * no - done with current sequence, push back passed char
230 */
231 *astate = S_GROUND;
232 return (UNVIS_VALIDPUSH);
233
234 case S_OCTAL3: /* third possible octal digit */
235 *astate = S_GROUND;
236 if (isoctal(uc)) {
237 *cp = (*cp << 3) + (c - '0');
238 return (UNVIS_VALID);
239 }
240 /*
241 * we were done, push back passed char
242 */
243 return (UNVIS_VALIDPUSH);
244
245 case S_HEX1:
246 if (isxdigit(uc)) {
247 *cp = xtod(uc);
248 *astate = S_HEX2;
249 return (0);
250 }
251 /*
252 * no - done with current sequence, push back passed char
253 */
254 *astate = S_GROUND;
255 return (UNVIS_VALIDPUSH);
256
257 case S_HEX2:
258 *astate = S_GROUND;
259 if (isxdigit(uc)) {
260 *cp = xtod(uc) | (*cp << 4);
261 return (UNVIS_VALID);
262 }
263 return (UNVIS_VALIDPUSH);
264
265 default:
266 /*
267 * decoder in unknown state - (probably uninitialized)
268 */
269 *astate = S_GROUND;
270 return (UNVIS_SYNBAD);
271 }
272 }
273
274 /*
275 * strunvis - decode src into dst
276 *
277 * Number of chars decoded into dst is returned, -1 on error.
278 * Dst is null terminated.
279 */
280
281 int
282 strunvisx(dst, src, flag)
283 char *dst;
284 const char *src;
285 int flag;
286 {
287 char c;
288 char *start = dst;
289 int state = 0;
290
291 _DIAGASSERT(src != NULL);
292 _DIAGASSERT(dst != NULL);
293
294 while ((c = *src++) != '\0') {
295 again:
296 switch (__unvis13(dst, c, &state, flag)) {
297 case UNVIS_VALID:
298 dst++;
299 break;
300 case UNVIS_VALIDPUSH:
301 dst++;
302 goto again;
303 case 0:
304 case UNVIS_NOCHAR:
305 break;
306 default:
307 return (-1);
308 }
309 }
310 if (__unvis13(dst, c, &state, UNVIS_END) == UNVIS_VALID)
311 dst++;
312 *dst = '\0';
313 return (dst - start);
314 }
315
316 int
317 strunvis(dst, src)
318 char *dst;
319 const char *src;
320 {
321 return strunvisx(dst, src, 0);
322 }
323 #endif
324