unvis.c revision 178825 
1/*	$NetBSD: unvis.c,v 1.19 2000/01/22 22:19:13 mycroft 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#if 1
33#ifdef HAVE_CONFIG_H
34#include <config.h>
35RCSID("$Id: unvis.c 21005 2007-06-08 01:54:35Z lha $");
36#endif
37#include "roken.h"
38#ifndef _DIAGASSERT
39#define _DIAGASSERT(X)
40#endif
41#else
42#include <sys/cdefs.h>
43#if defined(LIBC_SCCS) && !defined(lint)
44#if 0
45static char sccsid[] = "@(#)unvis.c	8.1 (Berkeley) 6/4/93";
46#else
47__RCSID("$NetBSD: unvis.c,v 1.19 2000/01/22 22:19:13 mycroft Exp $");
48#endif
49#endif /* LIBC_SCCS and not lint */
50
51#define __LIBC12_SOURCE__
52
53#include "namespace.h"
54#endif
55#include <sys/types.h>
56
57#include <assert.h>
58#include <ctype.h>
59#include <stdio.h>
60#include <vis.h>
61
62#if 0
63#ifdef __weak_alias
64__weak_alias(strunvis,_strunvis)
65__weak_alias(unvis,_unvis)
66#endif
67
68__warn_references(unvis,
69    "warning: reference to compatibility unvis(); include <vis.h> for correct reference")
70#endif
71
72/*
73 * decode driven by state machine
74 */
75#define	S_GROUND	0	/* haven't seen escape char */
76#define	S_START		1	/* start decoding special sequence */
77#define	S_META		2	/* metachar started (M) */
78#define	S_META1		3	/* metachar more, regular char (-) */
79#define	S_CTRL		4	/* control char started (^) */
80#define	S_OCTAL2	5	/* octal digit 2 */
81#define	S_OCTAL3	6	/* octal digit 3 */
82
83#define	isoctal(c)	(((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
84
85int ROKEN_LIB_FUNCTION
86	rk_strunvis (char *, const char *);
87int ROKEN_LIB_FUNCTION
88	rk_unvis (char *, int, int *, int);
89
90/*
91 * unvis - decode characters previously encoded by vis
92 */
93
94int ROKEN_LIB_FUNCTION
95rk_unvis(char *cp, int c, int *astate, int flag)
96{
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_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		*cp = c;
118		return (UNVIS_VALID);
119
120	case S_START:
121		switch(c) {
122		case '\\':
123			*cp = c;
124			*astate = S_GROUND;
125			return (UNVIS_VALID);
126		case '0': case '1': case '2': case '3':
127		case '4': case '5': case '6': case '7':
128			*cp = (c - '0');
129			*astate = S_OCTAL2;
130			return (0);
131		case 'M':
132			*cp = (char)0200;
133			*astate = S_META;
134			return (0);
135		case '^':
136			*astate = S_CTRL;
137			return (0);
138		case 'n':
139			*cp = '\n';
140			*astate = S_GROUND;
141			return (UNVIS_VALID);
142		case 'r':
143			*cp = '\r';
144			*astate = S_GROUND;
145			return (UNVIS_VALID);
146		case 'b':
147			*cp = '\b';
148			*astate = S_GROUND;
149			return (UNVIS_VALID);
150		case 'a':
151			*cp = '\007';
152			*astate = S_GROUND;
153			return (UNVIS_VALID);
154		case 'v':
155			*cp = '\v';
156			*astate = S_GROUND;
157			return (UNVIS_VALID);
158		case 't':
159			*cp = '\t';
160			*astate = S_GROUND;
161			return (UNVIS_VALID);
162		case 'f':
163			*cp = '\f';
164			*astate = S_GROUND;
165			return (UNVIS_VALID);
166		case 's':
167			*cp = ' ';
168			*astate = S_GROUND;
169			return (UNVIS_VALID);
170		case 'E':
171			*cp = '\033';
172			*astate = S_GROUND;
173			return (UNVIS_VALID);
174		case '\n':
175			/*
176			 * hidden newline
177			 */
178			*astate = S_GROUND;
179			return (UNVIS_NOCHAR);
180		case '$':
181			/*
182			 * hidden marker
183			 */
184			*astate = S_GROUND;
185			return (UNVIS_NOCHAR);
186		}
187		*astate = S_GROUND;
188		return (UNVIS_SYNBAD);
189
190	case S_META:
191		if (c == '-')
192			*astate = S_META1;
193		else if (c == '^')
194			*astate = S_CTRL;
195		else {
196			*astate = S_GROUND;
197			return (UNVIS_SYNBAD);
198		}
199		return (0);
200
201	case S_META1:
202		*astate = S_GROUND;
203		*cp |= c;
204		return (UNVIS_VALID);
205
206	case S_CTRL:
207		if (c == '?')
208			*cp |= 0177;
209		else
210			*cp |= c & 037;
211		*astate = S_GROUND;
212		return (UNVIS_VALID);
213
214	case S_OCTAL2:	/* second possible octal digit */
215		if (isoctal(c)) {
216			/*
217			 * yes - and maybe a third
218			 */
219			*cp = (*cp << 3) + (c - '0');
220			*astate = S_OCTAL3;
221			return (0);
222		}
223		/*
224		 * no - done with current sequence, push back passed char
225		 */
226		*astate = S_GROUND;
227		return (UNVIS_VALIDPUSH);
228
229	case S_OCTAL3:	/* third possible octal digit */
230		*astate = S_GROUND;
231		if (isoctal(c)) {
232			*cp = (*cp << 3) + (c - '0');
233			return (UNVIS_VALID);
234		}
235		/*
236		 * we were done, push back passed char
237		 */
238		return (UNVIS_VALIDPUSH);
239
240	default:
241		/*
242		 * decoder in unknown state - (probably uninitialized)
243		 */
244		*astate = S_GROUND;
245		return (UNVIS_SYNBAD);
246	}
247}
248
249/*
250 * strunvis - decode src into dst
251 *
252 *	Number of chars decoded into dst is returned, -1 on error.
253 *	Dst is null terminated.
254 */
255
256int ROKEN_LIB_FUNCTION
257rk_strunvis(char *dst, const char *src)
258{
259	char c;
260	char *start = dst;
261	int state = 0;
262
263	_DIAGASSERT(src != NULL);
264	_DIAGASSERT(dst != NULL);
265
266	while ((c = *src++) != '\0') {
267	again:
268		switch (rk_unvis(dst, (unsigned char)c, &state, 0)) {
269		case UNVIS_VALID:
270			dst++;
271			break;
272		case UNVIS_VALIDPUSH:
273			dst++;
274			goto again;
275		case 0:
276		case UNVIS_NOCHAR:
277			break;
278		default:
279			return (-1);
280		}
281	}
282	if (unvis(dst, (unsigned char)c, &state, UNVIS_END) == UNVIS_VALID)
283		dst++;
284	*dst = '\0';
285	return (dst - start);
286}
287