Victor Chang | 7322950 | 2020-09-17 13:39:19 +0100 | [diff] [blame] | 1 | // © 2016 and later: Unicode, Inc. and others. |
| 2 | // License & terms of use: http://www.unicode.org/copyright.html |
| 3 | /* |
| 4 | ********************************************************************** |
| 5 | * Copyright (C) 2005-2012, International Business Machines |
| 6 | * Corporation and others. All Rights Reserved. |
| 7 | ********************************************************************** |
| 8 | */ |
| 9 | |
| 10 | #ifndef __CSRMBCS_H |
| 11 | #define __CSRMBCS_H |
| 12 | |
| 13 | #include "unicode/utypes.h" |
| 14 | |
| 15 | #if !UCONFIG_NO_CONVERSION |
| 16 | |
| 17 | #include "csrecog.h" |
| 18 | |
| 19 | U_NAMESPACE_BEGIN |
| 20 | |
| 21 | // "Character" iterated character class. |
| 22 | // Recognizers for specific mbcs encodings make their "characters" available |
| 23 | // by providing a nextChar() function that fills in an instance of IteratedChar |
| 24 | // with the next char from the input. |
| 25 | // The returned characters are not converted to Unicode, but remain as the raw |
| 26 | // bytes (concatenated into an int) from the codepage data. |
| 27 | // |
| 28 | // For Asian charsets, use the raw input rather than the input that has been |
| 29 | // stripped of markup. Detection only considers multi-byte chars, effectively |
| 30 | // stripping markup anyway, and double byte chars do occur in markup too. |
| 31 | // |
| 32 | class IteratedChar : public UMemory |
| 33 | { |
| 34 | public: |
| 35 | uint32_t charValue; // 1-4 bytes from the raw input data |
| 36 | int32_t index; |
| 37 | int32_t nextIndex; |
| 38 | UBool error; |
| 39 | UBool done; |
| 40 | |
| 41 | public: |
| 42 | IteratedChar(); |
| 43 | //void reset(); |
| 44 | int32_t nextByte(InputText* det); |
| 45 | }; |
| 46 | |
| 47 | |
| 48 | class CharsetRecog_mbcs : public CharsetRecognizer { |
| 49 | |
| 50 | protected: |
| 51 | /** |
| 52 | * Test the match of this charset with the input text data |
| 53 | * which is obtained via the CharsetDetector object. |
| 54 | * |
| 55 | * @param det The CharsetDetector, which contains the input text |
| 56 | * to be checked for being in this charset. |
| 57 | * @return Two values packed into one int (Damn java, anyhow) |
| 58 | * <br/> |
| 59 | * bits 0-7: the match confidence, ranging from 0-100 |
| 60 | * <br/> |
| 61 | * bits 8-15: The match reason, an enum-like value. |
| 62 | */ |
| 63 | int32_t match_mbcs(InputText* det, const uint16_t commonChars[], int32_t commonCharsLen) const; |
| 64 | |
| 65 | public: |
| 66 | |
| 67 | virtual ~CharsetRecog_mbcs(); |
| 68 | |
| 69 | /** |
| 70 | * Get the IANA name of this charset. |
| 71 | * @return the charset name. |
| 72 | */ |
| 73 | |
| 74 | const char *getName() const = 0; |
| 75 | const char *getLanguage() const = 0; |
| 76 | UBool match(InputText* input, CharsetMatch *results) const = 0; |
| 77 | |
| 78 | /** |
| 79 | * Get the next character (however many bytes it is) from the input data |
| 80 | * Subclasses for specific charset encodings must implement this function |
| 81 | * to get characters according to the rules of their encoding scheme. |
| 82 | * |
| 83 | * This function is not a method of class IteratedChar only because |
| 84 | * that would require a lot of extra derived classes, which is awkward. |
| 85 | * @param it The IteratedChar "struct" into which the returned char is placed. |
| 86 | * @param det The charset detector, which is needed to get at the input byte data |
| 87 | * being iterated over. |
| 88 | * @return True if a character was returned, false at end of input. |
| 89 | */ |
| 90 | virtual UBool nextChar(IteratedChar *it, InputText *textIn) const = 0; |
| 91 | |
| 92 | }; |
| 93 | |
| 94 | |
| 95 | /** |
| 96 | * Shift-JIS charset recognizer. |
| 97 | * |
| 98 | */ |
| 99 | class CharsetRecog_sjis : public CharsetRecog_mbcs { |
| 100 | public: |
| 101 | virtual ~CharsetRecog_sjis(); |
| 102 | |
| 103 | UBool nextChar(IteratedChar *it, InputText *det) const; |
| 104 | |
| 105 | UBool match(InputText* input, CharsetMatch *results) const; |
| 106 | |
| 107 | const char *getName() const; |
| 108 | const char *getLanguage() const; |
| 109 | |
| 110 | }; |
| 111 | |
| 112 | |
| 113 | /** |
| 114 | * EUC charset recognizers. One abstract class that provides the common function |
| 115 | * for getting the next character according to the EUC encoding scheme, |
| 116 | * and nested derived classes for EUC_KR, EUC_JP, EUC_CN. |
| 117 | * |
| 118 | */ |
| 119 | class CharsetRecog_euc : public CharsetRecog_mbcs |
| 120 | { |
| 121 | public: |
| 122 | virtual ~CharsetRecog_euc(); |
| 123 | |
| 124 | const char *getName() const = 0; |
| 125 | const char *getLanguage() const = 0; |
| 126 | |
| 127 | UBool match(InputText* input, CharsetMatch *results) const = 0; |
| 128 | /* |
| 129 | * (non-Javadoc) |
| 130 | * Get the next character value for EUC based encodings. |
| 131 | * Character "value" is simply the raw bytes that make up the character |
| 132 | * packed into an int. |
| 133 | */ |
| 134 | UBool nextChar(IteratedChar *it, InputText *det) const; |
| 135 | }; |
| 136 | |
| 137 | /** |
| 138 | * The charset recognize for EUC-JP. A singleton instance of this class |
| 139 | * is created and kept by the public CharsetDetector class |
| 140 | */ |
| 141 | class CharsetRecog_euc_jp : public CharsetRecog_euc |
| 142 | { |
| 143 | public: |
| 144 | virtual ~CharsetRecog_euc_jp(); |
| 145 | |
| 146 | const char *getName() const; |
| 147 | const char *getLanguage() const; |
| 148 | |
| 149 | UBool match(InputText* input, CharsetMatch *results) const; |
| 150 | }; |
| 151 | |
| 152 | /** |
| 153 | * The charset recognize for EUC-KR. A singleton instance of this class |
| 154 | * is created and kept by the public CharsetDetector class |
| 155 | */ |
| 156 | class CharsetRecog_euc_kr : public CharsetRecog_euc |
| 157 | { |
| 158 | public: |
| 159 | virtual ~CharsetRecog_euc_kr(); |
| 160 | |
| 161 | const char *getName() const; |
| 162 | const char *getLanguage() const; |
| 163 | |
| 164 | UBool match(InputText* input, CharsetMatch *results) const; |
| 165 | }; |
| 166 | |
| 167 | /** |
| 168 | * |
| 169 | * Big5 charset recognizer. |
| 170 | * |
| 171 | */ |
| 172 | class CharsetRecog_big5 : public CharsetRecog_mbcs |
| 173 | { |
| 174 | public: |
| 175 | virtual ~CharsetRecog_big5(); |
| 176 | |
| 177 | UBool nextChar(IteratedChar* it, InputText* det) const; |
| 178 | |
| 179 | const char *getName() const; |
| 180 | const char *getLanguage() const; |
| 181 | |
| 182 | UBool match(InputText* input, CharsetMatch *results) const; |
| 183 | }; |
| 184 | |
| 185 | |
| 186 | /** |
| 187 | * |
| 188 | * GB-18030 recognizer. Uses simplified Chinese statistics. |
| 189 | * |
| 190 | */ |
| 191 | class CharsetRecog_gb_18030 : public CharsetRecog_mbcs |
| 192 | { |
| 193 | public: |
| 194 | virtual ~CharsetRecog_gb_18030(); |
| 195 | |
| 196 | UBool nextChar(IteratedChar* it, InputText* det) const; |
| 197 | |
| 198 | const char *getName() const; |
| 199 | const char *getLanguage() const; |
| 200 | |
| 201 | UBool match(InputText* input, CharsetMatch *results) const; |
| 202 | }; |
| 203 | |
| 204 | U_NAMESPACE_END |
| 205 | |
| 206 | #endif |
| 207 | #endif /* __CSRMBCS_H */ |