1 /* 2 * Copyright 2001-2004 Unicode, Inc. 3 * 4 * Disclaimer 5 * 6 * This source code is provided as is by Unicode, Inc. No claims are 7 * made as to fitness for any particular purpose. No warranties of any 8 * kind are expressed or implied. The recipient agrees to determine 9 * applicability of information provided. If this file has been 10 * purchased on magnetic or optical media from Unicode, Inc., the 11 * sole remedy for any claim will be exchange of defective media 12 * within 90 days of receipt. 13 * 14 * Limitations on Rights to Redistribute This Code 15 * 16 * Unicode, Inc. hereby grants the right to freely use the information 17 * supplied in this file in the creation of products supporting the 18 * Unicode Standard, and to make copies of this file in any form 19 * for internal or external distribution as long as this notice 20 * remains attached. 21 */ 22 23 // 24 // Copyright (c) ZeroC, Inc. All rights reserved. 25 // 26 27 #ifndef ICE_UTIL_CONVERT_UTF_H 28 #define ICE_UTIL_CONVERT_UTF_H 29 30 /* --------------------------------------------------------------------- 31 32 Conversions between UTF32, UTF-16, and UTF-8. Header file. 33 34 Several funtions are included here, forming a complete set of 35 conversions between the three formats. UTF-7 is not included 36 here, but is handled in a separate source file. 37 38 Each of these routines takes pointers to input buffers and output 39 buffers. The input buffers are const. 40 41 Each routine converts the text between *sourceStart and sourceEnd, 42 putting the result into the buffer between *targetStart and 43 targetEnd. Note: the end pointers are *after* the last item: e.g. 44 *(sourceEnd - 1) is the last item. 45 46 The return result indicates whether the conversion was successful, 47 and if not, whether the problem was in the source or target buffers. 48 (Only the first encountered problem is indicated.) 49 50 After the conversion, *sourceStart and *targetStart are both 51 updated to point to the end of last text successfully converted in 52 the respective buffers. 53 54 Input parameters: 55 sourceStart - pointer to a pointer to the source buffer. 56 The contents of this are modified on return so that 57 it points at the next thing to be converted. 58 targetStart - similarly, pointer to pointer to the target buffer. 59 sourceEnd, targetEnd - respectively pointers to the ends of the 60 two buffers, for overflow checking only. 61 62 These conversion functions take a ConversionFlags argument. When this 63 flag is set to strict, both irregular sequences and isolated surrogates 64 will cause an error. When the flag is set to lenient, both irregular 65 sequences and isolated surrogates are converted. 66 67 Whether the flag is strict or lenient, all illegal sequences will cause 68 an error return. This includes sequences such as: <F4 90 80 80>, <C0 80>, 69 or <A0> in UTF-8, and values above 0x10FFFF in UTF-32. Conformant code 70 must check for illegal sequences. 71 72 When the flag is set to lenient, characters over 0x10FFFF are converted 73 to the replacement character; otherwise (when the flag is set to strict) 74 they constitute an error. 75 76 Output parameters: 77 The value "sourceIllegal" is returned from some routines if the input 78 sequence is malformed. When "sourceIllegal" is returned, the source 79 value will point to the illegal value that caused the problem. E.g., 80 in UTF-8 when a sequence is malformed, it points to the start of the 81 malformed sequence. 82 83 Author: Mark E. Davis, 1994. 84 Rev History: Rick McGowan, fixes & updates May 2001. 85 Fixes & updates, Sept 2001. 86 87 ------------------------------------------------------------------------ */ 88 89 namespace IceUtilInternal 90 { 91 92 enum ConversionFlags 93 { 94 strictConversion = 0, 95 lenientConversion 96 }; 97 98 enum ConversionResult 99 { 100 conversionOK, /* conversion successful */ 101 sourceExhausted, /* partial character in source, but hit end */ 102 targetExhausted, /* insuff. room in target for conversion */ 103 sourceIllegal /* source sequence is illegal/malformed */ 104 }; 105 106 /* --------------------------------------------------------------------- 107 The following 4 definitions are compiler-specific. 108 The C standard does not guarantee that wchar_t has at least 109 16 bits, so wchar_t is no less portable than unsigned short! 110 All should be unsigned values to avoid sign extension during 111 bit mask & shift operations. 112 ------------------------------------------------------------------------ */ 113 114 typedef unsigned int UTF32; /* at least 32 bits */ 115 typedef unsigned short UTF16; /* at least 16 bits */ 116 typedef unsigned char UTF8; /* typically 8 bits */ 117 typedef bool Boolean; /* 0 or 1 */ 118 119 /* Some fundamental constants */ 120 #define UNI_REPLACEMENT_CHAR static_cast<UTF32>(0x0000FFFD) 121 #define UNI_MAX_BMP static_cast<UTF32>(0x0000FFFF) 122 #define UNI_MAX_UTF16 static_cast<UTF32>(0x0010FFFF) 123 #define UNI_MAX_UTF32 static_cast<UTF32>(0x7FFFFFFF) 124 #define UNI_MAX_LEGAL_UTF32 static_cast<UTF32>(0x0010FFFF) 125 126 ConversionResult ConvertUTF8toUTF16( 127 const UTF8** sourceStart, const UTF8* sourceEnd, 128 UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags); 129 130 ConversionResult ConvertUTF16toUTF8 ( 131 const UTF16** sourceStart, const UTF16* sourceEnd, 132 UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); 133 134 ConversionResult ConvertUTF8toUTF32( 135 const UTF8** sourceStart, const UTF8* sourceEnd, 136 UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags); 137 138 ConversionResult ConvertUTF32toUTF8( 139 const UTF32** sourceStart, const UTF32* sourceEnd, 140 UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags); 141 142 bool isLegalUTF8Sequence(const UTF8* source, const UTF8* end); 143 144 /* --------------------------------------------------------------------- */ 145 146 } 147 #endif 148