unvis.c revision 1.24 1 /* $NetBSD: unvis.c,v 1.24 2003/08/07 16:42:59 agc 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 1.24 2003/08/07 16:42:59 agc 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, (int)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
97 _DIAGASSERT(cp != NULL);
98 _DIAGASSERT(astate != NULL);
99
100 if (flag & UNVIS_END) {
101 if (*astate == S_OCTAL2 || *astate == S_OCTAL3
102 || *astate == S_HEX2) {
103 *astate = S_GROUND;
104 return (UNVIS_VALID);
105 }
106 return (*astate == S_GROUND ? UNVIS_NOCHAR : UNVIS_SYNBAD);
107 }
108
109 switch (*astate) {
110
111 case S_GROUND:
112 *cp = 0;
113 if (c == '\\') {
114 *astate = S_START;
115 return (0);
116 }
117 if ((flag & VIS_HTTPSTYLE) && c == '%') {
118 *astate = S_HEX1;
119 return (0);
120 }
121 *cp = c;
122 return (UNVIS_VALID);
123
124 case S_START:
125 switch(c) {
126 case '\\':
127 *cp = c;
128 *astate = S_GROUND;
129 return (UNVIS_VALID);
130 case '0': case '1': case '2': case '3':
131 case '4': case '5': case '6': case '7':
132 *cp = (c - '0');
133 *astate = S_OCTAL2;
134 return (0);
135 case 'M':
136 *cp = (char)0200;
137 *astate = S_META;
138 return (0);
139 case '^':
140 *astate = S_CTRL;
141 return (0);
142 case 'n':
143 *cp = '\n';
144 *astate = S_GROUND;
145 return (UNVIS_VALID);
146 case 'r':
147 *cp = '\r';
148 *astate = S_GROUND;
149 return (UNVIS_VALID);
150 case 'b':
151 *cp = '\b';
152 *astate = S_GROUND;
153 return (UNVIS_VALID);
154 case 'a':
155 *cp = '\007';
156 *astate = S_GROUND;
157 return (UNVIS_VALID);
158 case 'v':
159 *cp = '\v';
160 *astate = S_GROUND;
161 return (UNVIS_VALID);
162 case 't':
163 *cp = '\t';
164 *astate = S_GROUND;
165 return (UNVIS_VALID);
166 case 'f':
167 *cp = '\f';
168 *astate = S_GROUND;
169 return (UNVIS_VALID);
170 case 's':
171 *cp = ' ';
172 *astate = S_GROUND;
173 return (UNVIS_VALID);
174 case 'E':
175 *cp = '\033';
176 *astate = S_GROUND;
177 return (UNVIS_VALID);
178 case '\n':
179 /*
180 * hidden newline
181 */
182 *astate = S_GROUND;
183 return (UNVIS_NOCHAR);
184 case '$':
185 /*
186 * hidden marker
187 */
188 *astate = S_GROUND;
189 return (UNVIS_NOCHAR);
190 }
191 *astate = S_GROUND;
192 return (UNVIS_SYNBAD);
193
194 case S_META:
195 if (c == '-')
196 *astate = S_META1;
197 else if (c == '^')
198 *astate = S_CTRL;
199 else {
200 *astate = S_GROUND;
201 return (UNVIS_SYNBAD);
202 }
203 return (0);
204
205 case S_META1:
206 *astate = S_GROUND;
207 *cp |= c;
208 return (UNVIS_VALID);
209
210 case S_CTRL:
211 if (c == '?')
212 *cp |= 0177;
213 else
214 *cp |= c & 037;
215 *astate = S_GROUND;
216 return (UNVIS_VALID);
217
218 case S_OCTAL2: /* second possible octal digit */
219 if (isoctal(c)) {
220 /*
221 * yes - and maybe a third
222 */
223 *cp = (*cp << 3) + (c - '0');
224 *astate = S_OCTAL3;
225 return (0);
226 }
227 /*
228 * no - done with current sequence, push back passed char
229 */
230 *astate = S_GROUND;
231 return (UNVIS_VALIDPUSH);
232
233 case S_OCTAL3: /* third possible octal digit */
234 *astate = S_GROUND;
235 if (isoctal(c)) {
236 *cp = (*cp << 3) + (c - '0');
237 return (UNVIS_VALID);
238 }
239 /*
240 * we were done, push back passed char
241 */
242 return (UNVIS_VALIDPUSH);
243 case S_HEX1:
244 if (isxdigit(c)) {
245 *cp = xtod(c);
246 *astate = S_HEX2;
247 return (0);
248 }
249 /*
250 * no - done with current sequence, push back passed char
251 */
252 *astate = S_GROUND;
253 return (UNVIS_VALIDPUSH);
254 case S_HEX2:
255 *astate = S_GROUND;
256 if (isxdigit(c)) {
257 *cp = xtod(c) | (*cp << 4);
258 return (UNVIS_VALID);
259 }
260 return (UNVIS_VALIDPUSH);
261 default:
262 /*
263 * decoder in unknown state - (probably uninitialized)
264 */
265 *astate = S_GROUND;
266 return (UNVIS_SYNBAD);
267 }
268 }
269
270 /*
271 * strunvis - decode src into dst
272 *
273 * Number of chars decoded into dst is returned, -1 on error.
274 * Dst is null terminated.
275 */
276
277 int
278 strunvisx(dst, src, flag)
279 char *dst;
280 const char *src;
281 int flag;
282 {
283 char c;
284 char *start = dst;
285 int state = 0;
286
287 _DIAGASSERT(src != NULL);
288 _DIAGASSERT(dst != NULL);
289
290 while ((c = *src++) != '\0') {
291 again:
292 switch (__unvis13(dst, c, &state, flag)) {
293 case UNVIS_VALID:
294 dst++;
295 break;
296 case UNVIS_VALIDPUSH:
297 dst++;
298 goto again;
299 case 0:
300 case UNVIS_NOCHAR:
301 break;
302 default:
303 return (-1);
304 }
305 }
306 if (__unvis13(dst, c, &state, UNVIS_END) == UNVIS_VALID)
307 dst++;
308 *dst = '\0';
309 return (dst - start);
310 }
311
312 int
313 strunvis(dst, src)
314 char *dst;
315 const char *src;
316 {
317 return strunvisx(dst, src, 0);
318 }
319 #endif
320