tools/nfotiser/parser.c

/* $FreeWRT: src/share/misc/licence.template,v 1.20 2006/12/11 21:04:56 tg Rel $ */

/*-
 * Copyright (c) 2007
 *      Thorsten Glaser <tg@mirbsd.de>
 *
 * Provided that these terms and disclaimer and all copyright notices
 * are retained or reproduced in an accompanying document, permission
 * is granted to deal in this work without restriction, including un-
 * limited rights to use, publicly perform, distribute, sell, modify,
 * merge, give away, or sublicence.
 *
 * Advertising materials mentioning features or use of this work must
 * display the following acknowledgement:
 *      This product includes material provided by Thorsten Glaser.
 * This acknowledgement does not need to be reprinted if this work is
 * linked into a bigger work whose licence does not allow such clause
 * and the author of this work is given due credit in the bigger work
 * or its accompanying documents, where such information is generally
 * kept, provided that said credits are retained.
 *
 * This work is provided "AS IS" and WITHOUT WARRANTY of any kind, to
 * the utmost extent permitted by applicable law, neither express nor
 * implied; without malicious intent or gross negligence. In no event
 * may a licensor, author or contributor be held liable for indirect,
 * direct, other damage, loss, or other issues arising in any way out
 * of dealing in the work, even if advised of the possibility of such
 * damage or existence of a defect, except proven that it results out
 * of said person's immediate fault when using the work as intended.
 */

#include <sys/param.h>
#include <sys/mman.h>
#include <sys/stat.h>

#include <err.h>
#include <errno.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

#include "nfotiser.h"

/*
 * Parsing works as follows:
 *
 * - strip completely empty line
 * - strip line beginning with hash mark
 * - if line ends with backslash, get next line that is not
 *   + empty
 *   + beginning with a hash mark
 *   and look if it begins with a tab (if not: syntax error)
 *   if so, strip backslash + newline + tab and repeat
 * - if line begins with a tab, strip it and append line to
 *   the last line, including the newline separator
 * - match line with '([A-Za-z_][A-Za-z0-9_]*)\t(.*)$' and
 *   call \1 the key and \2 the value (else: syntax error)
 * - uppercase the key
 * - enter the key/value pair in the system
 *
 * The following is also part of parsing, but left to the caller:
 * - replace ${foo} with the value of key "foo"
 * - undouble all backslashes
 *
 * We enter the key into the system up to three-fold:
 * - KWT_NORMAL => ([A-Za-z_][A-Za-z0-9_]*)
 *   + \1 = keyword (toupper'd)
 * - KWT_MULTI => ([A-Za-z_][A-Za-z0-9_]*)_([A-Za-z0-9_]*)
 *   + \1 = keyword (toupper'd)
 *   + \2 = kw_multi (case preserving)
 * - KWT_ITERATED => ([A-Za-z_][A-Za-z0-9_]*)_([0-9]*)
 *   + \1 = keyword (toupper'd)
 *   + \2 = kw_iter (unsigned integer value)
 * - KWT_MULTITER => ([A-Za-z_][A-Za-z0-9_]*)_([0-9]*)_([A-Za-z0-9_]*)
 *   + \1 = keyword (toupper'd)
 *   + \2 = kw_iter (unsigned integer value)
 *   + \3 = kw_multi (case preserving)
 *
 * Cf. https://www.freewrt.org/trac/wiki/Documentation/Specs/Freewrt_info_files
 * for more examples and a more human-readable version of this specification.
 */

struct parser_result *
nfo_parse(int fd, const struct parser_keywords *kws)
{
        struct parser_result *res;
        struct parser_res *entry;
        char ch, *cp, *t, *buf;
        size_t len, n;
        struct stat sb;

        res = xmalloc(sizeof (struct parser_result));
        CIRCLEQ_INIT(res);

        if (fstat(fd, &sb))
                err(255, "cannot stat");

        buf = mmap(NULL, (len = sb.st_size), PROT_READ, MAP_FILE, fd, 0);
        if (buf == MAP_FAILED)
                err(255, "cannot mmap %zu bytes", len);

// Log In
// CVSweb
Project: FreeWRT
// Summary // Activity // Search // Tracker // Lists // News // SCM // Wiki
1	/* $FreeWRT: src/share/misc/licence.template,v 1.20 2006/12/11 21:04:56 tg Rel $ */
2
3	/*-
4	* Copyright (c) 2007
5	* Thorsten Glaser <tg@mirbsd.de>
6	*
7	* Provided that these terms and disclaimer and all copyright notices
8	* are retained or reproduced in an accompanying document, permission
9	* is granted to deal in this work without restriction, including un-
10	* limited rights to use, publicly perform, distribute, sell, modify,
11	* merge, give away, or sublicence.
12	*
13	* Advertising materials mentioning features or use of this work must
14	* display the following acknowledgement:
15	* This product includes material provided by Thorsten Glaser.
16	* This acknowledgement does not need to be reprinted if this work is
17	* linked into a bigger work whose licence does not allow such clause
18	* and the author of this work is given due credit in the bigger work
19	* or its accompanying documents, where such information is generally
20	* kept, provided that said credits are retained.
21	*
22	* This work is provided "AS IS" and WITHOUT WARRANTY of any kind, to
23	* the utmost extent permitted by applicable law, neither express nor
24	* implied; without malicious intent or gross negligence. In no event
25	* may a licensor, author or contributor be held liable for indirect,
26	* direct, other damage, loss, or other issues arising in any way out
27	* of dealing in the work, even if advised of the possibility of such
28	* damage or existence of a defect, except proven that it results out
29	* of said person's immediate fault when using the work as intended.
30	*/
31
32	#include <sys/param.h>
33	#include <sys/mman.h>
34	#include <sys/stat.h>
35
36	#include <err.h>
37	#include <errno.h>
38	#include <stdint.h>
39	#include <stdio.h>
40	#include <stdlib.h>
41	#include <string.h>
42	#include <unistd.h>
43
44	#include "nfotiser.h"
45
46	/*
47	* Parsing works as follows:
48	*
49	* - strip completely empty line
50	* - strip line beginning with hash mark
51	* - if line ends with backslash, get next line that is not
52	* + empty
53	* + beginning with a hash mark
54	* and look if it begins with a tab (if not: syntax error)
55	* if so, strip backslash + newline + tab and repeat
56	* - if line begins with a tab, strip it and append line to
57	* the last line, including the newline separator
58	* - match line with '([A-Za-z_][A-Za-z0-9_])\t(.)$' and
59	* call \1 the key and \2 the value (else: syntax error)
60	* - uppercase the key
61	* - enter the key/value pair in the system
62	*
63	* The following is also part of parsing, but left to the caller:
64	* - replace ${foo} with the value of key "foo"
65	* - undouble all backslashes
66	*
67	* We enter the key into the system up to three-fold:
68	* - KWT_NORMAL => ([A-Za-z_][A-Za-z0-9_]*)
69	* + \1 = keyword (toupper'd)
70	* - KWT_MULTI => ([A-Za-z_][A-Za-z0-9_])_([A-Za-z0-9_])
71	* + \1 = keyword (toupper'd)
72	* + \2 = kw_multi (case preserving)
73	* - KWT_ITERATED => ([A-Za-z_][A-Za-z0-9_])_([0-9])
74	* + \1 = keyword (toupper'd)
75	* + \2 = kw_iter (unsigned integer value)
76	* - KWT_MULTITER => ([A-Za-z_][A-Za-z0-9_])_([0-9])_([A-Za-z0-9_]*)
77	* + \1 = keyword (toupper'd)
78	* + \2 = kw_iter (unsigned integer value)
79	* + \3 = kw_multi (case preserving)
80	*
81	* Cf. https://www.freewrt.org/trac/wiki/Documentation/Specs/Freewrt_info_files
82	* for more examples and a more human-readable version of this specification.
83	*/
84
85	struct parser_result *
86	nfo_parse(int fd, const struct parser_keywords *kws)
87	{
88	struct parser_result *res;
89	struct parser_res *entry;
90	char ch, cp, t, *buf;
91	size_t len, n;
92	struct stat sb;
93
94	res = xmalloc(sizeof (struct parser_result));
95	CIRCLEQ_INIT(res);
96
97	if (fstat(fd, &sb))
98	err(255, "cannot stat");
99
100	buf = mmap(NULL, (len = sb.st_size), PROT_READ, MAP_FILE, fd, 0);
101	if (buf == MAP_FAILED)
102	err(255, "cannot mmap %zu bytes", len);
103