1 #ifndef TAGLIB_UNICODE_H
2 #define TAGLIB_UNICODE_H
4 /*******************************************************************************
6 * THIS FILE IS INCLUDED IN TAGLIB, BUT IS NOT COPYRIGHTED BY THE TAGLIB *
7 * AUTHORS, NOT PART OF THE TAGLIB API AND COULD GO AWAY AT ANY POINT IN TIME. *
8 * AS SUCH IT SHOULD BE CONSIERED FOR INTERNAL USE ONLY. *
10 *******************************************************************************/
12 #ifndef DO_NOT_DOCUMENT // tell Doxygen not to document this header
15 * Copyright 2001 Unicode, Inc.
19 * This source code is provided as is by Unicode, Inc. No claims are
20 * made as to fitness for any particular purpose. No warranties of any
21 * kind are expressed or implied. The recipient agrees to determine
22 * applicability of information provided. If this file has been
23 * purchased on magnetic or optical media from Unicode, Inc., the
24 * sole remedy for any claim will be exchange of defective media
25 * within 90 days of receipt.
27 * Limitations on Rights to Redistribute This Code
29 * Unicode, Inc. hereby grants the right to freely use the information
30 * supplied in this file in the creation of products supporting the
31 * Unicode Standard, and to make copies of this file in any form
32 * for internal or external distribution as long as this notice
37 * This file has been modified by Scott Wheeler <wheeler@kde.org> to remove
38 * the UTF32 conversion functions and to place the appropriate functions
39 * in their own C++ namespace.
42 /* ---------------------------------------------------------------------
44 Conversions between UTF32, UTF-16, and UTF-8. Header file.
46 Several functions are included here, forming a complete set of
47 conversions between the three formats. UTF-7 is not included
48 here, but is handled in a separate source file.
50 Each of these routines takes pointers to input buffers and output
51 buffers. The input buffers are const.
53 Each routine converts the text between *sourceStart and sourceEnd,
54 putting the result into the buffer between *targetStart and
55 targetEnd. Note: the end pointers are *after* the last item: e.g.
56 *(sourceEnd - 1) is the last item.
58 The return result indicates whether the conversion was successful,
59 and if not, whether the problem was in the source or target buffers.
60 (Only the first encountered problem is indicated.)
62 After the conversion, *sourceStart and *targetStart are both
63 updated to point to the end of last text successfully converted in
64 the respective buffers.
67 sourceStart - pointer to a pointer to the source buffer.
68 The contents of this are modified on return so that
69 it points at the next thing to be converted.
70 targetStart - similarly, pointer to pointer to the target buffer.
71 sourceEnd, targetEnd - respectively pointers to the ends of the
72 two buffers, for overflow checking only.
74 These conversion functions take a ConversionFlags argument. When this
75 flag is set to strict, both irregular sequences and isolated surrogates
76 will cause an error. When the flag is set to lenient, both irregular
77 sequences and isolated surrogates are converted.
79 Whether the flag is strict or lenient, all illegal sequences will cause
80 an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>,
81 or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code
82 must check for illegal sequences.
84 When the flag is set to lenient, characters over 0x10FFFF are converted
85 to the replacement character; otherwise (when the flag is set to strict)
86 they constitute an error.
89 The value "sourceIllegal" is returned from some routines if the input
90 sequence is malformed. When "sourceIllegal" is returned, the source
91 value will point to the illegal value that caused the problem. E.g.,
92 in UTF-8 when a sequence is malformed, it points to the start of the
95 Author: Mark E. Davis, 1994.
96 Rev History: Rick McGowan, fixes & updates May 2001.
97 Fixes & updates, Sept 2001.
99 ------------------------------------------------------------------------ */
101 /* ---------------------------------------------------------------------
102 The following 4 definitions are compiler-specific.
103 The C standard does not guarantee that wchar_t has at least
104 16 bits, so wchar_t is no less portable than unsigned short!
105 All should be unsigned values to avoid sign extension during
106 bit mask & shift operations.
107 ------------------------------------------------------------------------ */
109 /* Some fundamental constants */
110 #define UNI_REPLACEMENT_CHAR (UTF32)0x0000FFFD
111 #define UNI_MAX_BMP (UTF32)0x0000FFFF
112 #define UNI_MAX_UTF16 (UTF32)0x0010FFFF
113 #define UNI_MAX_UTF32 (UTF32)0x7FFFFFFF
117 typedef unsigned long UTF32; /* at least 32 bits */
118 typedef unsigned short UTF16; /* at least 16 bits */
119 typedef unsigned char UTF8; /* typically 8 bits */
120 typedef unsigned char Boolean; /* 0 or 1 */
123 conversionOK = 0, /* conversion successful */
124 sourceExhausted = 1, /* partial character in source, but hit end */
125 targetExhausted = 2, /* insuff. room in target for conversion */
126 sourceIllegal = 3 /* source sequence is illegal/malformed */
130 strictConversion = 0,
134 ConversionResult ConvertUTF8toUTF16 (
135 const UTF8** sourceStart, const UTF8* sourceEnd,
136 UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags);
138 ConversionResult ConvertUTF16toUTF8 (
139 const UTF16** sourceStart, const UTF16* sourceEnd,
140 UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags);
142 Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd);
144 } // namespace Unicode
146 /* --------------------------------------------------------------------- */