1/*	$NetBSD: fparseln.c,v 1.10 2009/10/21 01:07:45 snj Exp $	*/
2
3/*-
4 * SPDX-License-Identifier: BSD-2-Clause
5 *
6 * Copyright (c) 1997 Christos Zoulas.  All rights reserved.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 *    notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 *    notice, this list of conditions and the following disclaimer in the
15 *    documentation and/or other materials provided with the distribution.
16 *
17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
18 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
19 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
20 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
21 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
22 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
23 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
24 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
26 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27 */
28
29#include <sys/types.h>
30#include <assert.h>
31#include <errno.h>
32#include <stdio.h>
33#include <string.h>
34#include <stdlib.h>
35#include <libutil.h>
36
37static int isescaped(const char *, const char *, int);
38
39/* isescaped():
40 *	Return true if the character in *p that belongs to a string
41 *	that starts in *sp, is escaped by the escape character esc.
42 */
43static int
44isescaped(const char *sp, const char *p, int esc)
45{
46	const char     *cp;
47	size_t		ne;
48
49#if 0
50	_DIAGASSERT(sp != NULL);
51	_DIAGASSERT(p != NULL);
52#endif
53
54	/* No escape character */
55	if (esc == '\0')
56		return 0;
57
58	/* Count the number of escape characters that precede ours */
59	for (ne = 0, cp = p; --cp >= sp && *cp == esc; ne++)
60		continue;
61
62	/* Return true if odd number of escape characters */
63	return (ne & 1) != 0;
64}
65
66
67/* fparseln():
68 *	Read a line from a file parsing continuations ending in \
69 *	and eliminating trailing newlines, or comments starting with
70 *	the comment char.
71 */
72char *
73fparseln(FILE *fp, size_t *size, size_t *lineno, const char str[3], int flags)
74{
75	static const char dstr[3] = { '\\', '\\', '#' };
76
77	size_t	s, len;
78	char   *buf;
79	char   *ptr, *cp;
80	int	cnt;
81	char	esc, con, nl, com;
82
83#if 0
84	_DIAGASSERT(fp != NULL);
85#endif
86
87	len = 0;
88	buf = NULL;
89	cnt = 1;
90
91	if (str == NULL)
92		str = dstr;
93
94	esc = str[0];
95	con = str[1];
96	com = str[2];
97	/*
98	 * XXX: it would be cool to be able to specify the newline character,
99	 * but unfortunately, fgetln does not let us
100	 */
101	nl  = '\n';
102
103	while (cnt) {
104		cnt = 0;
105
106		if (lineno)
107			(*lineno)++;
108
109		if ((ptr = fgetln(fp, &s)) == NULL)
110			break;
111
112		if (s && com) {		/* Check and eliminate comments */
113			for (cp = ptr; cp < ptr + s; cp++)
114				if (*cp == com && !isescaped(ptr, cp, esc)) {
115					s = cp - ptr;
116					cnt = s == 0 && buf == NULL;
117					break;
118				}
119		}
120
121		if (s && nl) { 		/* Check and eliminate newlines */
122			cp = &ptr[s - 1];
123
124			if (*cp == nl)
125				s--;	/* forget newline */
126		}
127
128		if (s && con) {		/* Check and eliminate continuations */
129			cp = &ptr[s - 1];
130
131			if (*cp == con && !isescaped(ptr, cp, esc)) {
132				s--;	/* forget continuation char */
133				cnt = 1;
134			}
135		}
136
137		if (s == 0) {
138			/*
139			 * nothing to add, skip realloc except in case
140			 * we need a minimal buf to return an empty line
141			 */
142			if (cnt || buf != NULL)
143				continue;
144		}
145
146		if ((cp = realloc(buf, len + s + 1)) == NULL) {
147			free(buf);
148			return NULL;
149		}
150		buf = cp;
151
152		(void) memcpy(buf + len, ptr, s);
153		len += s;
154		buf[len] = '\0';
155	}
156
157	if ((flags & FPARSELN_UNESCALL) != 0 && esc && buf != NULL &&
158	    strchr(buf, esc) != NULL) {
159		ptr = cp = buf;
160		while (cp[0] != '\0') {
161			int skipesc;
162
163			while (cp[0] != '\0' && cp[0] != esc)
164				*ptr++ = *cp++;
165			if (cp[0] == '\0' || cp[1] == '\0')
166				break;
167
168			skipesc = 0;
169			if (cp[1] == com)
170				skipesc += (flags & FPARSELN_UNESCCOMM);
171			if (cp[1] == con)
172				skipesc += (flags & FPARSELN_UNESCCONT);
173			if (cp[1] == esc)
174				skipesc += (flags & FPARSELN_UNESCESC);
175			if (cp[1] != com && cp[1] != con && cp[1] != esc)
176				skipesc = (flags & FPARSELN_UNESCREST);
177
178			if (skipesc)
179				cp++;
180			else
181				*ptr++ = *cp++;
182			*ptr++ = *cp++;
183		}
184		*ptr = '\0';
185		len = strlen(buf);
186	}
187
188	if (size)
189		*size = len;
190	return buf;
191}
192
193#ifdef TEST
194
195int
196main(int argc, char *argv[])
197{
198	char   *ptr;
199	size_t	size, line;
200
201	line = 0;
202	while ((ptr = fparseln(stdin, &size, &line, NULL,
203	    FPARSELN_UNESCALL)) != NULL)
204		printf("line %d (%d) |%s|\n", line, size, ptr);
205	return 0;
206}
207
208/*
209
210# This is a test
211line 1
212line 2 \
213line 3 # Comment
214line 4 \# Not comment \\\\
215
216# And a comment \
217line 5 \\\
218line 6
219
220*/
221
222#endif /* TEST */
223