1 /* Conversions between Unicode and legacy encodings.
2 Copyright (C) 2002, 2005, 2007, 2009-2023 Free Software Foundation, Inc.
3
4 This file is free software: you can redistribute it and/or modify
5 it under the terms of the GNU Lesser General Public License as
6 published by the Free Software Foundation; either version 2.1 of the
7 License, or (at your option) any later version.
8
9 This file is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU Lesser General Public License for more details.
13
14 You should have received a copy of the GNU Lesser General Public License
15 along with this program. If not, see <https://www.gnu.org/licenses/>. */
16
17 #ifndef _UNICONV_H
18 #define _UNICONV_H
19
20 /* Get size_t. */
21 #include <stddef.h>
22
23 #include "unitypes.h"
24
25 /* Get enum iconv_ilseq_handler. */
26 #include "iconveh.h"
27
28 /* Get uniconv_register_autodetect() declaration. */
29 #include "striconveha.h"
30
31 /* Get locale_charset() declaration. */
32 #include "localcharset.h"
33
34
35 #ifdef __cplusplus
36 extern "C" {
37 #endif
38
39
40 /* Converts an entire string, possibly including NUL bytes, from one encoding
41 to a Unicode encoding.
42 Converts a memory region given in encoding FROMCODE. FROMCODE is as for
43 iconv_open(3).
44 The input is in the memory region between SRC (inclusive) and SRC + SRCLEN
45 (exclusive).
46 If OFFSETS is not NULL, it should point to an array of SRCLEN integers; this
47 array is filled with offsets into the result, i.e. the character starting
48 at SRC[i] corresponds to the character starting at (*RESULTP)[OFFSETS[i]],
49 and other offsets are set to (size_t)(-1).
50 RESULTBUF and *LENGTHP should initially be a scratch buffer and its size,
51 or *RESULTBUF can be NULL.
52 May erase the contents of the memory at RESULTBUF.
53 If successful: The resulting Unicode string (non-NULL) is returned and its
54 length stored in *LENGTHP. The resulting string is RESULTBUF if no dynamic
55 memory allocation was necessary, or a freshly allocated memory block
56 otherwise.
57 In case of error: NULL is returned and errno is set. Particular errno
58 values: EINVAL, EILSEQ, ENOMEM. */
59 extern uint8_t *
60 u8_conv_from_encoding (const char *fromcode,
61 enum iconv_ilseq_handler handler,
62 const char *src, size_t srclen,
63 size_t *offsets,
64 uint8_t *resultbuf, size_t *lengthp);
65 extern uint16_t *
66 u16_conv_from_encoding (const char *fromcode,
67 enum iconv_ilseq_handler handler,
68 const char *src, size_t srclen,
69 size_t *offsets,
70 uint16_t *resultbuf, size_t *lengthp);
71 extern uint32_t *
72 u32_conv_from_encoding (const char *fromcode,
73 enum iconv_ilseq_handler handler,
74 const char *src, size_t srclen,
75 size_t *offsets,
76 uint32_t *resultbuf, size_t *lengthp);
77
78 /* Converts an entire Unicode string, possibly including NUL units, from a
79 Unicode encoding to a given encoding.
80 Converts a memory region to encoding TOCODE. TOCODE is as for
81 iconv_open(3).
82 The input is in the memory region between SRC (inclusive) and SRC + SRCLEN
83 (exclusive).
84 If OFFSETS is not NULL, it should point to an array of SRCLEN integers; this
85 array is filled with offsets into the result, i.e. the character starting
86 at SRC[i] corresponds to the character starting at (*RESULTP)[OFFSETS[i]],
87 and other offsets are set to (size_t)(-1).
88 RESULTBUF and *LENGTHP should initially be a scratch buffer and its size,
89 or RESULTBUF can be NULL.
90 May erase the contents of the memory at RESULTBUF.
91 If successful: The resulting string (non-NULL) is returned and its length
92 stored in *LENGTHP. The resulting string is RESULTBUF if no dynamic memory
93 allocation was necessary, or a freshly allocated memory block otherwise.
94 In case of error: NULL is returned and errno is set. Particular errno
95 values: EINVAL, EILSEQ, ENOMEM. */
96 extern char *
97 u8_conv_to_encoding (const char *tocode,
98 enum iconv_ilseq_handler handler,
99 const uint8_t *src, size_t srclen,
100 size_t *offsets,
101 char *_UC_RESTRICT resultbuf, size_t *lengthp);
102 extern char *
103 u16_conv_to_encoding (const char *tocode,
104 enum iconv_ilseq_handler handler,
105 const uint16_t *src, size_t srclen,
106 size_t *offsets,
107 char *_UC_RESTRICT resultbuf, size_t *lengthp);
108 extern char *
109 u32_conv_to_encoding (const char *tocode,
110 enum iconv_ilseq_handler handler,
111 const uint32_t *src, size_t srclen,
112 size_t *offsets,
113 char *_UC_RESTRICT resultbuf, size_t *lengthp);
114
115 /* Converts a NUL terminated string from a given encoding.
116 The result is malloc allocated, or NULL (with errno set) in case of error.
117 Particular errno values: EILSEQ, ENOMEM. */
118 extern uint8_t *
119 u8_strconv_from_encoding (const char *string,
120 const char *fromcode,
121 enum iconv_ilseq_handler handler);
122 extern uint16_t *
123 u16_strconv_from_encoding (const char *string,
124 const char *fromcode,
125 enum iconv_ilseq_handler handler);
126 extern uint32_t *
127 u32_strconv_from_encoding (const char *string,
128 const char *fromcode,
129 enum iconv_ilseq_handler handler);
130
131 /* Converts a NUL terminated string to a given encoding.
132 The result is malloc allocated, or NULL (with errno set) in case of error.
133 Particular errno values: EILSEQ, ENOMEM. */
134 extern char *
135 u8_strconv_to_encoding (const uint8_t *string,
136 const char *tocode,
137 enum iconv_ilseq_handler handler);
138 extern char *
139 u16_strconv_to_encoding (const uint16_t *string,
140 const char *tocode,
141 enum iconv_ilseq_handler handler);
142 extern char *
143 u32_strconv_to_encoding (const uint32_t *string,
144 const char *tocode,
145 enum iconv_ilseq_handler handler);
146
147 /* Converts a NUL terminated string from the locale encoding.
148 The result is malloc allocated, or NULL (with errno set) in case of error.
149 Particular errno values: ENOMEM. */
150 extern uint8_t *
151 u8_strconv_from_locale (const char *string);
152 extern uint16_t *
153 u16_strconv_from_locale (const char *string);
154 extern uint32_t *
155 u32_strconv_from_locale (const char *string);
156
157 /* Converts a NUL terminated string to the locale encoding.
158 The result is malloc allocated, or NULL (with errno set) in case of error.
159 Particular errno values: ENOMEM. */
160 extern char *
161 u8_strconv_to_locale (const uint8_t *string);
162 extern char *
163 u16_strconv_to_locale (const uint16_t *string);
164 extern char *
165 u32_strconv_to_locale (const uint32_t *string);
166
167
168 #ifdef __cplusplus
169 }
170 #endif
171
172 #endif /* _UNICONV_H */