xref: /freebsd/lib/libutil/fparseln.c (revision 23f282aa31e9b6fceacd449020e936e98d6f2298)
1 /*	$NetBSD: fparseln.c,v 1.9 1999/09/20 04:48:06 lukem Exp $	*/
2 /* $FreeBSD$ */
3 
4 /*
5  * Copyright (c) 1997 Christos Zoulas.  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. All advertising materials mentioning features or use of this software
16  *    must display the following acknowledgement:
17  *	This product includes software developed by Christos Zoulas.
18  * 4. The name of the author may not be used to endorse or promote products
19  *    derived from this software without specific prior written permission.
20  *
21  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
22  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
23  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
24  * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
25  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
26  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
27  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
28  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
29  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
30  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31  */
32 
33 #include <sys/cdefs.h>
34 #if defined(LIBC_SCCS) && !defined(lint)
35 __RCSID("$FreeBSD$");
36 #endif
37 
38 #include <sys/types.h>
39 #include <assert.h>
40 #include <errno.h>
41 #include <stdio.h>
42 #include <string.h>
43 #include <stdlib.h>
44 #include <libutil.h>
45 
46 static int isescaped __P((const char *, const char *, int));
47 
48 /* isescaped():
49  *	Return true if the character in *p that belongs to a string
50  *	that starts in *sp, is escaped by the escape character esc.
51  */
52 static int
53 isescaped(sp, p, esc)
54 	const char *sp, *p;
55 	int esc;
56 {
57 	const char     *cp;
58 	size_t		ne;
59 
60 #if 0
61 	_DIAGASSERT(sp != NULL);
62 	_DIAGASSERT(p != NULL);
63 #endif
64 
65 	/* No escape character */
66 	if (esc == '\0')
67 		return 1;
68 
69 	/* Count the number of escape characters that precede ours */
70 	for (ne = 0, cp = p; --cp >= sp && *cp == esc; ne++)
71 		continue;
72 
73 	/* Return true if odd number of escape characters */
74 	return (ne & 1) != 0;
75 }
76 
77 
78 /* fparseln():
79  *	Read a line from a file parsing continuations ending in \
80  *	and eliminating trailing newlines, or comments starting with
81  *	the comment char.
82  */
83 char *
84 fparseln(fp, size, lineno, str, flags)
85 	FILE		*fp;
86 	size_t		*size;
87 	size_t		*lineno;
88 	const char	 str[3];
89 	int		 flags;
90 {
91 	static const char dstr[3] = { '\\', '\\', '#' };
92 
93 	size_t	s, len;
94 	char   *buf;
95 	char   *ptr, *cp;
96 	int	cnt;
97 	char	esc, con, nl, com;
98 
99 #if 0
100 	_DIAGASSERT(fp != NULL);
101 #endif
102 
103 	len = 0;
104 	buf = NULL;
105 	cnt = 1;
106 
107 	if (str == NULL)
108 		str = dstr;
109 
110 	esc = str[0];
111 	con = str[1];
112 	com = str[2];
113 	/*
114 	 * XXX: it would be cool to be able to specify the newline character,
115 	 * but unfortunately, fgetln does not let us
116 	 */
117 	nl  = '\n';
118 
119 	while (cnt) {
120 		cnt = 0;
121 
122 		if (lineno)
123 			(*lineno)++;
124 
125 		if ((ptr = fgetln(fp, &s)) == NULL)
126 			break;
127 
128 		if (s && com) {		/* Check and eliminate comments */
129 			for (cp = ptr; cp < ptr + s; cp++)
130 				if (*cp == com && !isescaped(ptr, cp, esc)) {
131 					s = cp - ptr;
132 					cnt = s == 0 && buf == NULL;
133 					break;
134 				}
135 		}
136 
137 		if (s && nl) { 		/* Check and eliminate newlines */
138 			cp = &ptr[s - 1];
139 
140 			if (*cp == nl)
141 				s--;	/* forget newline */
142 		}
143 
144 		if (s && con) {		/* Check and eliminate continuations */
145 			cp = &ptr[s - 1];
146 
147 			if (*cp == con && !isescaped(ptr, cp, esc)) {
148 				s--;	/* forget escape */
149 				cnt = 1;
150 			}
151 		}
152 
153 		if (s == 0 && buf != NULL)
154 			continue;
155 
156 		if ((cp = realloc(buf, len + s + 1)) == NULL) {
157 			free(buf);
158 			return NULL;
159 		}
160 		buf = cp;
161 
162 		(void) memcpy(buf + len, ptr, s);
163 		len += s;
164 		buf[len] = '\0';
165 	}
166 
167 	if ((flags & FPARSELN_UNESCALL) != 0 && esc && buf != NULL &&
168 	    strchr(buf, esc) != NULL) {
169 		ptr = cp = buf;
170 		while (cp[0] != '\0') {
171 			int skipesc;
172 
173 			while (cp[0] != '\0' && cp[0] != esc)
174 				*ptr++ = *cp++;
175 			if (cp[0] == '\0' || cp[1] == '\0')
176 				break;
177 
178 			skipesc = 0;
179 			if (cp[1] == com)
180 				skipesc += (flags & FPARSELN_UNESCCOMM);
181 			if (cp[1] == con)
182 				skipesc += (flags & FPARSELN_UNESCCONT);
183 			if (cp[1] == esc)
184 				skipesc += (flags & FPARSELN_UNESCESC);
185 			if (cp[1] != com && cp[1] != con && cp[1] != esc)
186 				skipesc = (flags & FPARSELN_UNESCREST);
187 
188 			if (skipesc)
189 				cp++;
190 			else
191 				*ptr++ = *cp++;
192 			*ptr++ = *cp++;
193 		}
194 		*ptr = '\0';
195 		len = strlen(buf);
196 	}
197 
198 	if (size)
199 		*size = len;
200 	return buf;
201 }
202 
203 #ifdef TEST
204 
205 int main __P((int, char **));
206 
207 int
208 main(argc, argv)
209 	int argc;
210 	char **argv;
211 {
212 	char   *ptr;
213 	size_t	size, line;
214 
215 	line = 0;
216 	while ((ptr = fparseln(stdin, &size, &line, NULL,
217 	    FPARSELN_UNESCALL)) != NULL)
218 		printf("line %d (%d) |%s|\n", line, size, ptr);
219 	return 0;
220 }
221 
222 /*
223 
224 # This is a test
225 line 1
226 line 2 \
227 line 3 # Comment
228 line 4 \# Not comment \\\\
229 
230 # And a comment \
231 line 5 \\\
232 line 6
233 
234 */
235 
236 #endif /* TEST */
237