1 /*
2  * "streamable kanji code filter and converter"
3  * Copyright (c) 1998-2002 HappySize, Inc. All rights reserved.
4  *
5  * LICENSE NOTICES
6  *
7  * This file is part of "streamable kanji code filter and converter",
8  * which is distributed under the terms of GNU Lesser General Public
9  * License (version 2) as published by the Free Software Foundation.
10  *
11  * This software is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14  * GNU Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with "streamable kanji code filter and converter";
18  * if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19  * Suite 330, Boston, MA  02111-1307  USA
20  *
21  * The author of this file:
22  *
23  */
24 /*
25  * The source code included in this files was separated from mbfilter.c
26  * by Moriyoshi Koizumi <moriyoshi@php.net> on 20 Dec 2002. The file
27  * mbfilter.c is included in this package .
28  *
29  */
30 
31 #include "libmbfl/config.h"
32 
33 #ifdef HAVE_STRINGS_H
34 	/* For strcasecmp */
35 	#include <strings.h>
36 #endif
37 
38 #include "mbfl_encoding.h"
39 #include "mbfilter_pass.h"
40 #include "mbfilter_8bit.h"
41 
42 #include "filters/mbfilter_base64.h"
43 #include "filters/mbfilter_cjk.h"
44 #include "filters/mbfilter_qprint.h"
45 #include "filters/mbfilter_uuencode.h"
46 #include "filters/mbfilter_7bit.h"
47 #include "filters/mbfilter_utf7.h"
48 #include "filters/mbfilter_utf7imap.h"
49 #include "filters/mbfilter_utf8.h"
50 #include "filters/mbfilter_utf8_mobile.h"
51 #include "filters/mbfilter_utf16.h"
52 #include "filters/mbfilter_utf32.h"
53 #include "filters/mbfilter_ucs4.h"
54 #include "filters/mbfilter_ucs2.h"
55 #include "filters/mbfilter_htmlent.h"
56 #include "filters/mbfilter_singlebyte.h"
57 
58 #ifndef HAVE_STRCASECMP
59 #ifdef HAVE_STRICMP
60 #define strcasecmp stricmp
61 #endif
62 #endif
63 
64 
65 static const mbfl_encoding *mbfl_encoding_ptr_list[] = {
66 	&mbfl_encoding_base64,
67 	&mbfl_encoding_uuencode,
68 	&mbfl_encoding_html_ent,
69 	&mbfl_encoding_qprint,
70 	&mbfl_encoding_7bit,
71 	&mbfl_encoding_8bit,
72 	&mbfl_encoding_ucs4,
73 	&mbfl_encoding_ucs4be,
74 	&mbfl_encoding_ucs4le,
75 	&mbfl_encoding_ucs2,
76 	&mbfl_encoding_ucs2be,
77 	&mbfl_encoding_ucs2le,
78 	&mbfl_encoding_utf32,
79 	&mbfl_encoding_utf32be,
80 	&mbfl_encoding_utf32le,
81 	&mbfl_encoding_utf16,
82 	&mbfl_encoding_utf16be,
83 	&mbfl_encoding_utf16le,
84 	&mbfl_encoding_utf8,
85 	&mbfl_encoding_utf7,
86 	&mbfl_encoding_utf7imap,
87 	&mbfl_encoding_ascii,
88 	&mbfl_encoding_euc_jp,
89 	&mbfl_encoding_sjis,
90 	&mbfl_encoding_eucjp_win,
91 	&mbfl_encoding_eucjp2004,
92  	&mbfl_encoding_sjis_docomo,
93  	&mbfl_encoding_sjis_kddi,
94  	&mbfl_encoding_sjis_sb,
95  	&mbfl_encoding_sjis_mac,
96 	&mbfl_encoding_sjis2004,
97 	&mbfl_encoding_utf8_docomo,
98 	&mbfl_encoding_utf8_kddi_a,
99 	&mbfl_encoding_utf8_kddi_b,
100 	&mbfl_encoding_utf8_sb,
101 	&mbfl_encoding_cp932,
102 	&mbfl_encoding_sjiswin,
103 	&mbfl_encoding_cp51932,
104 	&mbfl_encoding_jis,
105 	&mbfl_encoding_2022jp,
106 	&mbfl_encoding_2022jpms,
107 	&mbfl_encoding_gb18030,
108 	&mbfl_encoding_cp1252,
109 	&mbfl_encoding_cp1254,
110 	&mbfl_encoding_8859_1,
111 	&mbfl_encoding_8859_2,
112 	&mbfl_encoding_8859_3,
113 	&mbfl_encoding_8859_4,
114 	&mbfl_encoding_8859_5,
115 	&mbfl_encoding_8859_6,
116 	&mbfl_encoding_8859_7,
117 	&mbfl_encoding_8859_8,
118 	&mbfl_encoding_8859_9,
119 	&mbfl_encoding_8859_10,
120 	&mbfl_encoding_8859_13,
121 	&mbfl_encoding_8859_14,
122 	&mbfl_encoding_8859_15,
123 	&mbfl_encoding_8859_16,
124 	&mbfl_encoding_euc_cn,
125 	&mbfl_encoding_cp936,
126 	&mbfl_encoding_hz,
127 	&mbfl_encoding_euc_tw,
128 	&mbfl_encoding_big5,
129 	&mbfl_encoding_cp950,
130 	&mbfl_encoding_euc_kr,
131 	&mbfl_encoding_uhc,
132 	&mbfl_encoding_2022kr,
133 	&mbfl_encoding_cp1251,
134 	&mbfl_encoding_cp866,
135 	&mbfl_encoding_koi8r,
136 	&mbfl_encoding_koi8u,
137 	&mbfl_encoding_armscii8,
138 	&mbfl_encoding_cp850,
139 	&mbfl_encoding_2022jp_2004,
140 	&mbfl_encoding_2022jp_kddi,
141 	&mbfl_encoding_cp50220,
142 	&mbfl_encoding_cp50221,
143 	&mbfl_encoding_cp50222,
144 	NULL
145 };
146 
mbfl_name2encoding(const char * name)147 const mbfl_encoding *mbfl_name2encoding(const char *name)
148 {
149 	const mbfl_encoding **encoding;
150 
151 	for (encoding = mbfl_encoding_ptr_list; *encoding; encoding++) {
152 		if (strcasecmp((*encoding)->name, name) == 0) {
153 			return *encoding;
154 		}
155 	}
156 
157 	/* search MIME charset name */
158 	for (encoding = mbfl_encoding_ptr_list; *encoding; encoding++) {
159 		if ((*encoding)->mime_name) {
160 			if (strcasecmp((*encoding)->mime_name, name) == 0) {
161 				return *encoding;
162 			}
163 		}
164 	}
165 
166 	/* search aliases */
167 	for (encoding = mbfl_encoding_ptr_list; *encoding; encoding++) {
168 		if ((*encoding)->aliases) {
169 			for (const char **alias = (*encoding)->aliases; *alias; alias++) {
170 				if (strcasecmp(*alias, name) == 0) {
171 					return *encoding;
172 				}
173 			}
174 		}
175 	}
176 
177 	return NULL;
178 }
179 
mbfl_no2encoding(enum mbfl_no_encoding no_encoding)180 const mbfl_encoding *mbfl_no2encoding(enum mbfl_no_encoding no_encoding)
181 {
182 	const mbfl_encoding **encoding;
183 
184 	for (encoding = mbfl_encoding_ptr_list; *encoding; encoding++) {
185 		if ((*encoding)->no_encoding == no_encoding) {
186 			return *encoding;
187 		}
188 	}
189 
190 	return NULL;
191 }
192 
mbfl_no_encoding2name(enum mbfl_no_encoding no_encoding)193 const char *mbfl_no_encoding2name(enum mbfl_no_encoding no_encoding)
194 {
195 	const mbfl_encoding *encoding = mbfl_no2encoding(no_encoding);
196 	return encoding ? encoding->name : "";
197 }
198 
mbfl_get_supported_encodings(void)199 const mbfl_encoding **mbfl_get_supported_encodings(void)
200 {
201 	return mbfl_encoding_ptr_list;
202 }
203 
mbfl_encoding_preferred_mime_name(const mbfl_encoding * encoding)204 const char *mbfl_encoding_preferred_mime_name(const mbfl_encoding *encoding)
205 {
206 	if (encoding->mime_name && encoding->mime_name[0] != '\0') {
207 		return encoding->mime_name;
208 	}
209 	return NULL;
210 }
211