1/*	$NetBSD: fparseln.c,v 1.1 2015/01/22 03:48:07 christos Exp $	*/
2
3/*
4 * Copyright (c) 1997 Christos Zoulas.  All rights reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions
8 * are met:
9 * 1. Redistributions of source code must retain the above copyright
10 *    notice, this list of conditions and the following disclaimer.
11 * 2. Redistributions in binary form must reproduce the above copyright
12 *    notice, this list of conditions and the following disclaimer in the
13 *    documentation and/or other materials provided with the distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
16 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
17 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
18 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
19 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
20 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
21 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
22 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
24 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25 */
26#ifdef HAVE_CONFIG_H
27#include "config.h"
28#endif
29
30#include <sys/cdefs.h>
31#if defined(LIBC_SCCS) && !defined(lint)
32__RCSID("$NetBSD: fparseln.c,v 1.1 2015/01/22 03:48:07 christos Exp $");
33#endif /* LIBC_SCCS and not lint */
34
35#include <assert.h>
36#include <errno.h>
37#include <stdio.h>
38#include <string.h>
39#include <stdlib.h>
40
41#if ! HAVE_FPARSELN || BROKEN_FPARSELN
42
43#define FLOCKFILE(fp)
44#define FUNLOCKFILE(fp)
45
46#if defined(_REENTRANT) && !HAVE_NBTOOL_CONFIG_H
47#define __fgetln(f, l) __fgetstr(f, l, '\n')
48#else
49#define __fgetln(f, l) fgetln(f, l)
50#endif
51
52static int isescaped(const char *, const char *, int);
53
54/* isescaped():
55 *	Return true if the character in *p that belongs to a string
56 *	that starts in *sp, is escaped by the escape character esc.
57 */
58static int
59isescaped(const char *sp, const char *p, int esc)
60{
61	const char     *cp;
62	size_t		ne;
63
64	/* No escape character */
65	if (esc == '\0')
66		return 0;
67
68	/* Count the number of escape characters that precede ours */
69	for (ne = 0, cp = p; --cp >= sp && *cp == esc; ne++)
70		continue;
71
72	/* Return true if odd number of escape characters */
73	return (ne & 1) != 0;
74}
75
76
77/* fparseln():
78 *	Read a line from a file parsing continuations ending in \
79 *	and eliminating trailing newlines, or comments starting with
80 *	the comment char.
81 */
82char *
83fparseln(FILE *fp, size_t *size, size_t *lineno, const char str[3], int flags)
84{
85	static const char dstr[3] = { '\\', '\\', '#' };
86
87	size_t	s, len;
88	char   *buf;
89	char   *ptr, *cp;
90	int	cnt;
91	char	esc, con, nl, com;
92
93	len = 0;
94	buf = NULL;
95	cnt = 1;
96
97	if (str == NULL)
98		str = dstr;
99
100	esc = str[0];
101	con = str[1];
102	com = str[2];
103	/*
104	 * XXX: it would be cool to be able to specify the newline character,
105	 * but unfortunately, fgetln does not let us
106	 */
107	nl  = '\n';
108
109	FLOCKFILE(fp);
110
111	while (cnt) {
112		cnt = 0;
113
114		if (lineno)
115			(*lineno)++;
116
117		if ((ptr = __fgetln(fp, &s)) == NULL)
118			break;
119
120		if (s && com) {		/* Check and eliminate comments */
121			for (cp = ptr; cp < ptr + s; cp++)
122				if (*cp == com && !isescaped(ptr, cp, esc)) {
123					s = cp - ptr;
124					cnt = s == 0 && buf == NULL;
125					break;
126				}
127		}
128
129		if (s && nl) { 		/* Check and eliminate newlines */
130			cp = &ptr[s - 1];
131
132			if (*cp == nl)
133				s--;	/* forget newline */
134		}
135
136		if (s && con) {		/* Check and eliminate continuations */
137			cp = &ptr[s - 1];
138
139			if (*cp == con && !isescaped(ptr, cp, esc)) {
140				s--;	/* forget continuation char */
141				cnt = 1;
142			}
143		}
144
145		if (s == 0) {
146			/*
147			 * nothing to add, skip realloc except in case
148			 * we need a minimal buf to return an empty line
149			 */
150			if (cnt || buf != NULL)
151				continue;
152		}
153
154		if ((cp = realloc(buf, len + s + 1)) == NULL) {
155			FUNLOCKFILE(fp);
156			free(buf);
157			return NULL;
158		}
159		buf = cp;
160
161		(void) memcpy(buf + len, ptr, s);
162		len += s;
163		buf[len] = '\0';
164	}
165
166	FUNLOCKFILE(fp);
167
168	if ((flags & FPARSELN_UNESCALL) != 0 && esc && buf != NULL &&
169	    strchr(buf, esc) != NULL) {
170		ptr = cp = buf;
171		while (cp[0] != '\0') {
172			int skipesc;
173
174			while (cp[0] != '\0' && cp[0] != esc)
175				*ptr++ = *cp++;
176			if (cp[0] == '\0' || cp[1] == '\0')
177				break;
178
179			skipesc = 0;
180			if (cp[1] == com)
181				skipesc += (flags & FPARSELN_UNESCCOMM);
182			if (cp[1] == con)
183				skipesc += (flags & FPARSELN_UNESCCONT);
184			if (cp[1] == esc)
185				skipesc += (flags & FPARSELN_UNESCESC);
186			if (cp[1] != com && cp[1] != con && cp[1] != esc)
187				skipesc = (flags & FPARSELN_UNESCREST);
188
189			if (skipesc)
190				cp++;
191			else
192				*ptr++ = *cp++;
193			*ptr++ = *cp++;
194		}
195		*ptr = '\0';
196		len = strlen(buf);
197	}
198
199	if (size)
200		*size = len;
201	return buf;
202}
203
204#ifdef TEST
205
206int main(int, char **);
207
208int
209main(int argc, char **argv)
210{
211	char   *ptr;
212	size_t	size, line;
213
214	line = 0;
215	while ((ptr = fparseln(stdin, &size, &line, NULL,
216	    FPARSELN_UNESCALL)) != NULL)
217		printf("line %d (%d) |%s|\n", line, size, ptr);
218	return 0;
219}
220
221/*
222
223# This is a test
224line 1
225line 2 \
226line 3 # Comment
227line 4 \# Not comment \\\\
228
229# And a comment \
230line 5 \\\
231line 6
232
233*/
234
235#endif /* TEST */
236#endif	/* ! HAVE_FPARSELN || BROKEN_FPARSELN */
237