Commit | Line | Data |
---|---|---|
460b9539 | 1 | /* |
2 | * This file is part of DisOrder. | |
3 | * Copyright (C) 2004, 2005 Richard Kettlewell | |
4 | * | |
5 | * This program is free software; you can redistribute it and/or modify | |
6 | * it under the terms of the GNU General Public License as published by | |
7 | * the Free Software Foundation; either version 2 of the License, or | |
8 | * (at your option) any later version. | |
9 | * | |
10 | * This program is distributed in the hope that it will be useful, but | |
11 | * WITHOUT ANY WARRANTY; without even the implied warranty of | |
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
13 | * General Public License for more details. | |
14 | * | |
15 | * You should have received a copy of the GNU General Public License | |
16 | * along with this program; if not, write to the Free Software | |
17 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 | |
18 | * USA | |
19 | */ | |
14ad73b9 | 20 | /** @file lib/charset.c @brief Character set conversion */ |
460b9539 | 21 | |
22 | #include <config.h> | |
23 | #include "types.h" | |
24 | ||
25 | #include <iconv.h> | |
26 | #include <string.h> | |
27 | #include <errno.h> | |
28 | #include <langinfo.h> | |
29 | ||
30 | #include "mem.h" | |
31 | #include "log.h" | |
32 | #include "charset.h" | |
33 | #include "configuration.h" | |
34 | #include "utf8.h" | |
35 | #include "vector.h" | |
36 | ||
14ad73b9 RK |
37 | /** @brief Low-level converstion routine |
38 | * @param from Source encoding | |
39 | * @param to Destination encoding | |
40 | * @param ptr First byte to convert | |
41 | * @param n Number of bytes to convert | |
42 | * @return Converted text, 0-terminated; or NULL on error. | |
43 | */ | |
460b9539 | 44 | static void *convert(const char *from, const char *to, |
45 | const void *ptr, size_t n) { | |
46 | iconv_t i; | |
47 | size_t len; | |
48 | char *buf = 0, *s, *d; | |
49 | size_t bufsize = 0, sl, dl; | |
50 | ||
51 | if((i = iconv_open(to, from)) == (iconv_t)-1) | |
52 | fatal(errno, "error calling iconv_open"); | |
53 | do { | |
54 | bufsize = bufsize ? 2 * bufsize : 32; | |
55 | buf = xrealloc_noptr(buf, bufsize); | |
56 | iconv(i, 0, 0, 0, 0); | |
57 | s = (char *)ptr; | |
58 | sl = n; | |
59 | d = buf; | |
60 | dl = bufsize; | |
61 | /* (void *) to work around FreeBSD's nonstandard iconv prototype */ | |
62 | len = iconv(i, (void *)&s, &sl, &d, &dl); | |
63 | } while(len == (size_t)-1 && errno == E2BIG); | |
64 | iconv_close(i); | |
65 | if(len == (size_t)-1) { | |
66 | error(errno, "error converting from %s to %s", from, to); | |
67 | return 0; | |
68 | } | |
69 | return buf; | |
70 | } | |
71 | ||
14ad73b9 RK |
72 | /** @brief Convert UTF-8 to UCS-4 |
73 | * @param mb Pointer to 0-terminated UTF-8 string | |
74 | * @return Pointer to 0-terminated UCS-4 string | |
75 | * | |
76 | * Not everybody's iconv supports UCS-4, and it's inconvenient to have to know | |
77 | * our endianness, and it's easy to convert it ourselves, so we do. See also | |
78 | * @ref ucs42utf8(). | |
79 | */ | |
460b9539 | 80 | uint32_t *utf82ucs4(const char *mb) { |
81 | struct dynstr_ucs4 d; | |
82 | uint32_t c; | |
83 | ||
84 | dynstr_ucs4_init(&d); | |
85 | while(*mb) { | |
86 | PARSE_UTF8(mb, c, | |
87 | error(0, "invalid UTF-8 sequence"); return 0;); | |
88 | dynstr_ucs4_append(&d, c); | |
89 | } | |
90 | dynstr_ucs4_terminate(&d); | |
91 | return d.vec; | |
92 | } | |
93 | ||
14ad73b9 RK |
94 | /** @brief Convert UCS-4 to UTF-8 |
95 | * @param u Pointer to 0-terminated UCS-4 string | |
96 | * @return Pointer to 0-terminated UTF-8 string | |
97 | * | |
98 | * See @ref utf82ucs4(). | |
99 | */ | |
460b9539 | 100 | char *ucs42utf8(const uint32_t *u) { |
101 | struct dynstr d; | |
102 | uint32_t c; | |
103 | ||
104 | dynstr_init(&d); | |
105 | while((c = *u++)) { | |
106 | if(c < 0x80) | |
107 | dynstr_append(&d, c); | |
108 | else if(c < 0x800) { | |
109 | dynstr_append(&d, 0xC0 | (c >> 6)); | |
110 | dynstr_append(&d, 0x80 | (c & 0x3F)); | |
111 | } else if(c < 0x10000) { | |
112 | dynstr_append(&d, 0xE0 | (c >> 12)); | |
113 | dynstr_append(&d, 0x80 | ((c >> 6) & 0x3F)); | |
114 | dynstr_append(&d, 0x80 | (c & 0x3F)); | |
115 | } else if(c < 0x110000) { | |
116 | dynstr_append(&d, 0xF0 | (c >> 18)); | |
117 | dynstr_append(&d, 0x80 | ((c >> 12) & 0x3F)); | |
118 | dynstr_append(&d, 0x80 | ((c >> 6) & 0x3F)); | |
119 | dynstr_append(&d, 0x80 | (c & 0x3F)); | |
120 | } else { | |
121 | error(0, "invalid UCS-4 character"); | |
122 | return 0; | |
123 | } | |
124 | } | |
125 | dynstr_terminate(&d); | |
126 | return d.vec; | |
127 | } | |
128 | ||
14ad73b9 | 129 | /** @brief Convert from the local multibyte encoding to UTF-8 */ |
460b9539 | 130 | char *mb2utf8(const char *mb) { |
131 | return convert(nl_langinfo(CODESET), "UTF-8", mb, strlen(mb) + 1); | |
132 | } | |
133 | ||
14ad73b9 | 134 | /** @brief Convert from UTF-8 to the local multibyte encoding */ |
460b9539 | 135 | char *utf82mb(const char *utf8) { |
136 | return convert("UTF-8", nl_langinfo(CODESET), utf8, strlen(utf8) + 1); | |
137 | } | |
138 | ||
14ad73b9 | 139 | /** @brief Convert from encoding @p from to UTF-8 */ |
460b9539 | 140 | char *any2utf8(const char *from, const char *any) { |
141 | return convert(from, "UTF-8", any, strlen(any) + 1); | |
142 | } | |
143 | ||
14ad73b9 | 144 | /** @brief Convert from encoding @p from to the local multibyte encoding */ |
460b9539 | 145 | char *any2mb(const char *from, const char *any) { |
146 | if(from) return convert(from, nl_langinfo(CODESET), any, strlen(any) + 1); | |
147 | else return xstrdup(any); | |
148 | } | |
149 | ||
14ad73b9 | 150 | /** @brief Convert from encoding @p from to encoding @p to */ |
460b9539 | 151 | char *any2any(const char *from, |
152 | const char *to, | |
153 | const char *any) { | |
154 | if(from || to) return convert(from, to, any, strlen(any) + 1); | |
155 | else return xstrdup(any); | |
156 | } | |
157 | ||
14ad73b9 RK |
158 | /** @brief strlen workalike for UCS-4 strings |
159 | * | |
160 | * We don't rely on the local @c wchar_t being UCS-4. | |
161 | */ | |
460b9539 | 162 | int ucs4cmp(const uint32_t *a, const uint32_t *b) { |
163 | while(*a && *b && *a == *b) ++a, ++b; | |
164 | if(*a > *b) return 1; | |
165 | else if(*a < *b) return -1; | |
166 | else return 0; | |
167 | } | |
168 | ||
169 | /* | |
170 | Local Variables: | |
171 | c-basic-offset:2 | |
172 | comment-column:40 | |
173 | End: | |
174 | */ |