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) 2006-2012, International Business Machines Corporation and others. * |
| 6 | * All Rights Reserved. * |
| 7 | ************************************************************************************ |
| 8 | */ |
| 9 | |
| 10 | #ifndef BRKENG_H |
| 11 | #define BRKENG_H |
| 12 | |
| 13 | #include "unicode/utypes.h" |
| 14 | #include "unicode/uobject.h" |
| 15 | #include "unicode/utext.h" |
| 16 | #include "unicode/uscript.h" |
| 17 | |
| 18 | U_NAMESPACE_BEGIN |
| 19 | |
| 20 | class UnicodeSet; |
| 21 | class UStack; |
| 22 | class UVector32; |
| 23 | class DictionaryMatcher; |
| 24 | |
| 25 | /******************************************************************* |
| 26 | * LanguageBreakEngine |
| 27 | */ |
| 28 | |
| 29 | /** |
| 30 | * <p>LanguageBreakEngines implement language-specific knowledge for |
| 31 | * finding text boundaries within a run of characters belonging to a |
| 32 | * specific set. The boundaries will be of a specific kind, e.g. word, |
| 33 | * line, etc.</p> |
| 34 | * |
| 35 | * <p>LanguageBreakEngines should normally be implemented so as to |
| 36 | * be shared between threads without locking.</p> |
| 37 | */ |
| 38 | class LanguageBreakEngine : public UMemory { |
| 39 | public: |
| 40 | |
| 41 | /** |
| 42 | * <p>Default constructor.</p> |
| 43 | * |
| 44 | */ |
| 45 | LanguageBreakEngine(); |
| 46 | |
| 47 | /** |
| 48 | * <p>Virtual destructor.</p> |
| 49 | */ |
| 50 | virtual ~LanguageBreakEngine(); |
| 51 | |
| 52 | /** |
| 53 | * <p>Indicate whether this engine handles a particular character for |
| 54 | * a particular kind of break.</p> |
| 55 | * |
| 56 | * @param c A character which begins a run that the engine might handle |
Victor Chang | ce4bf3c | 2021-01-19 16:34:24 +0000 | [diff] [blame] | 57 | * @return true if this engine handles the particular character and break |
Victor Chang | 7322950 | 2020-09-17 13:39:19 +0100 | [diff] [blame] | 58 | * type. |
| 59 | */ |
| 60 | virtual UBool handles(UChar32 c) const = 0; |
| 61 | |
| 62 | /** |
| 63 | * <p>Find any breaks within a run in the supplied text.</p> |
| 64 | * |
| 65 | * @param text A UText representing the text. The |
| 66 | * iterator is left at the end of the run of characters which the engine |
| 67 | * is capable of handling. |
| 68 | * @param startPos The start of the run within the supplied text. |
| 69 | * @param endPos The end of the run within the supplied text. |
| 70 | * @param foundBreaks A Vector of int32_t to receive the breaks. |
| 71 | * @return The number of breaks found. |
| 72 | */ |
| 73 | virtual int32_t findBreaks( UText *text, |
| 74 | int32_t startPos, |
| 75 | int32_t endPos, |
| 76 | UVector32 &foundBreaks ) const = 0; |
| 77 | |
| 78 | }; |
| 79 | |
| 80 | /******************************************************************* |
| 81 | * LanguageBreakFactory |
| 82 | */ |
| 83 | |
| 84 | /** |
| 85 | * <p>LanguageBreakFactorys find and return a LanguageBreakEngine |
| 86 | * that can determine breaks for characters in a specific set, if |
| 87 | * such an object can be found.</p> |
| 88 | * |
| 89 | * <p>If a LanguageBreakFactory is to be shared between threads, |
| 90 | * appropriate synchronization must be used; there is none internal |
| 91 | * to the factory.</p> |
| 92 | * |
| 93 | * <p>A LanguageBreakEngine returned by a LanguageBreakFactory can |
| 94 | * normally be shared between threads without synchronization, unless |
| 95 | * the specific subclass of LanguageBreakFactory indicates otherwise.</p> |
| 96 | * |
| 97 | * <p>A LanguageBreakFactory is responsible for deleting any LanguageBreakEngine |
| 98 | * it returns when it itself is deleted, unless the specific subclass of |
| 99 | * LanguageBreakFactory indicates otherwise. Naturally, the factory should |
| 100 | * not be deleted until the LanguageBreakEngines it has returned are no |
| 101 | * longer needed.</p> |
| 102 | */ |
| 103 | class LanguageBreakFactory : public UMemory { |
| 104 | public: |
| 105 | |
| 106 | /** |
| 107 | * <p>Default constructor.</p> |
| 108 | * |
| 109 | */ |
| 110 | LanguageBreakFactory(); |
| 111 | |
| 112 | /** |
| 113 | * <p>Virtual destructor.</p> |
| 114 | */ |
| 115 | virtual ~LanguageBreakFactory(); |
| 116 | |
| 117 | /** |
| 118 | * <p>Find and return a LanguageBreakEngine that can find the desired |
| 119 | * kind of break for the set of characters to which the supplied |
| 120 | * character belongs. It is up to the set of available engines to |
| 121 | * determine what the sets of characters are.</p> |
| 122 | * |
| 123 | * @param c A character that begins a run for which a LanguageBreakEngine is |
| 124 | * sought. |
| 125 | * @return A LanguageBreakEngine with the desired characteristics, or 0. |
| 126 | */ |
| 127 | virtual const LanguageBreakEngine *getEngineFor(UChar32 c) = 0; |
| 128 | |
| 129 | }; |
| 130 | |
| 131 | /******************************************************************* |
| 132 | * UnhandledEngine |
| 133 | */ |
| 134 | |
| 135 | /** |
| 136 | * <p>UnhandledEngine is a special subclass of LanguageBreakEngine that |
| 137 | * handles characters that no other LanguageBreakEngine is available to |
| 138 | * handle. It is told the character and the type of break; at its |
| 139 | * discretion it may handle more than the specified character (e.g., |
| 140 | * the entire script to which that character belongs.</p> |
| 141 | * |
| 142 | * <p>UnhandledEngines may not be shared between threads without |
| 143 | * external synchronization.</p> |
| 144 | */ |
| 145 | |
| 146 | class UnhandledEngine : public LanguageBreakEngine { |
| 147 | private: |
| 148 | |
| 149 | /** |
| 150 | * The sets of characters handled. |
| 151 | * @internal |
| 152 | */ |
| 153 | |
| 154 | UnicodeSet *fHandled; |
| 155 | |
| 156 | public: |
| 157 | |
| 158 | /** |
| 159 | * <p>Default constructor.</p> |
| 160 | * |
| 161 | */ |
| 162 | UnhandledEngine(UErrorCode &status); |
| 163 | |
| 164 | /** |
| 165 | * <p>Virtual destructor.</p> |
| 166 | */ |
| 167 | virtual ~UnhandledEngine(); |
| 168 | |
| 169 | /** |
| 170 | * <p>Indicate whether this engine handles a particular character for |
| 171 | * a particular kind of break.</p> |
| 172 | * |
| 173 | * @param c A character which begins a run that the engine might handle |
Victor Chang | ce4bf3c | 2021-01-19 16:34:24 +0000 | [diff] [blame] | 174 | * @return true if this engine handles the particular character and break |
Victor Chang | 7322950 | 2020-09-17 13:39:19 +0100 | [diff] [blame] | 175 | * type. |
| 176 | */ |
| 177 | virtual UBool handles(UChar32 c) const; |
| 178 | |
| 179 | /** |
| 180 | * <p>Find any breaks within a run in the supplied text.</p> |
| 181 | * |
| 182 | * @param text A UText representing the text (TODO: UText). The |
| 183 | * iterator is left at the end of the run of characters which the engine |
| 184 | * is capable of handling. |
| 185 | * @param startPos The start of the run within the supplied text. |
| 186 | * @param endPos The end of the run within the supplied text. |
| 187 | * @param foundBreaks An allocated C array of the breaks found, if any |
| 188 | * @return The number of breaks found. |
| 189 | */ |
| 190 | virtual int32_t findBreaks( UText *text, |
| 191 | int32_t startPos, |
| 192 | int32_t endPos, |
| 193 | UVector32 &foundBreaks ) const; |
| 194 | |
| 195 | /** |
| 196 | * <p>Tell the engine to handle a particular character and break type.</p> |
| 197 | * |
| 198 | * @param c A character which the engine should handle |
| 199 | */ |
| 200 | virtual void handleCharacter(UChar32 c); |
| 201 | |
| 202 | }; |
| 203 | |
| 204 | /******************************************************************* |
| 205 | * ICULanguageBreakFactory |
| 206 | */ |
| 207 | |
| 208 | /** |
| 209 | * <p>ICULanguageBreakFactory is the default LanguageBreakFactory for |
| 210 | * ICU. It creates dictionary-based LanguageBreakEngines from dictionary |
| 211 | * data in the ICU data file.</p> |
| 212 | */ |
| 213 | class ICULanguageBreakFactory : public LanguageBreakFactory { |
| 214 | private: |
| 215 | |
| 216 | /** |
| 217 | * The stack of break engines created by this factory |
| 218 | * @internal |
| 219 | */ |
| 220 | |
| 221 | UStack *fEngines; |
| 222 | |
| 223 | public: |
| 224 | |
| 225 | /** |
| 226 | * <p>Standard constructor.</p> |
| 227 | * |
| 228 | */ |
| 229 | ICULanguageBreakFactory(UErrorCode &status); |
| 230 | |
| 231 | /** |
| 232 | * <p>Virtual destructor.</p> |
| 233 | */ |
| 234 | virtual ~ICULanguageBreakFactory(); |
| 235 | |
| 236 | /** |
| 237 | * <p>Find and return a LanguageBreakEngine that can find the desired |
| 238 | * kind of break for the set of characters to which the supplied |
| 239 | * character belongs. It is up to the set of available engines to |
| 240 | * determine what the sets of characters are.</p> |
| 241 | * |
| 242 | * @param c A character that begins a run for which a LanguageBreakEngine is |
| 243 | * sought. |
| 244 | * @return A LanguageBreakEngine with the desired characteristics, or 0. |
| 245 | */ |
| 246 | virtual const LanguageBreakEngine *getEngineFor(UChar32 c); |
| 247 | |
| 248 | protected: |
| 249 | /** |
| 250 | * <p>Create a LanguageBreakEngine for the set of characters to which |
| 251 | * the supplied character belongs, for the specified break type.</p> |
| 252 | * |
| 253 | * @param c A character that begins a run for which a LanguageBreakEngine is |
| 254 | * sought. |
| 255 | * @return A LanguageBreakEngine with the desired characteristics, or 0. |
| 256 | */ |
| 257 | virtual const LanguageBreakEngine *loadEngineFor(UChar32 c); |
| 258 | |
| 259 | /** |
| 260 | * <p>Create a DictionaryMatcher for the specified script and break type.</p> |
| 261 | * @param script An ISO 15924 script code that identifies the dictionary to be |
| 262 | * created. |
| 263 | * @return A DictionaryMatcher with the desired characteristics, or NULL. |
| 264 | */ |
| 265 | virtual DictionaryMatcher *loadDictionaryMatcherFor(UScriptCode script); |
| 266 | }; |
| 267 | |
| 268 | U_NAMESPACE_END |
| 269 | |
| 270 | /* BRKENG_H */ |
| 271 | #endif |