| 1 | // © 2016 and later: Unicode, Inc. and others. |
| 2 | // License & terms of use: http://www.unicode.org/copyright.html |
| 3 | /* |
| 4 | ********************************************************************** |
| 5 | * Copyright (C) 1999-2007, International Business Machines Corporation |
| 6 | * and others. All Rights Reserved. |
| 7 | ********************************************************************** |
| 8 | * Date Name Description |
| 9 | * 11/17/99 aliu Creation. |
| 10 | ********************************************************************** |
| 11 | */ |
| 12 | #ifndef RBT_DATA_H |
| 13 | #define RBT_DATA_H |
| 14 | |
| 15 | #include "unicode/utypes.h" |
| 16 | #include "unicode/uclean.h" |
| 17 | |
| 18 | #if !UCONFIG_NO_TRANSLITERATION |
| 19 | |
| 20 | #include "unicode/uobject.h" |
| 21 | #include "rbt_set.h" |
| 22 | #include "hash.h" |
| 23 | |
| 24 | U_NAMESPACE_BEGIN |
| 25 | |
| 26 | class UnicodeFunctor; |
| 27 | class UnicodeMatcher; |
| 28 | class UnicodeReplacer; |
| 29 | |
| 30 | /** |
| 31 | * The rule data for a RuleBasedTransliterators. RBT objects hold |
| 32 | * a const pointer to a TRD object that they do not own. TRD objects |
| 33 | * are essentially the parsed rules in compact, usable form. The |
| 34 | * TRD objects themselves are held for the life of the process in |
| 35 | * a static cache owned by Transliterator. |
| 36 | * |
| 37 | * This class' API is a little asymmetric. There is a method to |
| 38 | * define a variable, but no way to define a set. This is because the |
| 39 | * sets are defined by the parser in a UVector, and the vector is |
| 40 | * copied into a fixed-size array here. Once this is done, no new |
| 41 | * sets may be defined. In practice, there is no need to do so, since |
| 42 | * generating the data and using it are discrete phases. When there |
| 43 | * is a need to access the set data during the parse phase, another |
| 44 | * data structure handles this. See the parsing code for more |
| 45 | * details. |
| 46 | */ |
| 47 | class TransliterationRuleData : public UMemory { |
| 48 | |
| 49 | public: |
| 50 | |
| 51 | // PUBLIC DATA MEMBERS |
| 52 | |
| 53 | /** |
| 54 | * Rule table. May be empty. |
| 55 | */ |
| 56 | TransliterationRuleSet ruleSet; |
| 57 | |
| 58 | /** |
| 59 | * Map variable name (String) to variable (UnicodeString). A variable name |
| 60 | * corresponds to zero or more characters, stored in a UnicodeString in |
| 61 | * this hash. One or more of these chars may also correspond to a |
| 62 | * UnicodeMatcher, in which case the character in the UnicodeString in this hash is |
| 63 | * a stand-in: it is an index for a secondary lookup in |
| 64 | * data.variables. The stand-in also represents the UnicodeMatcher in |
| 65 | * the stored rules. |
| 66 | */ |
| 67 | Hashtable variableNames; |
| 68 | |
| 69 | /** |
| 70 | * Map category variable (UChar) to set (UnicodeFunctor). |
| 71 | * Variables that correspond to a set of characters are mapped |
| 72 | * from variable name to a stand-in character in data.variableNames. |
| 73 | * The stand-in then serves as a key in this hash to lookup the |
| 74 | * actual UnicodeFunctor object. In addition, the stand-in is |
| 75 | * stored in the rule text to represent the set of characters. |
| 76 | * variables[i] represents character (variablesBase + i). |
| 77 | */ |
| 78 | UnicodeFunctor** variables; |
| 79 | |
| 80 | /** |
| 81 | * Flag that indicates whether the variables are owned (if a single |
| 82 | * call to Transliterator::createFromRules() produces a CompoundTransliterator |
| 83 | * with more than one RuleBasedTransliterator as children, they all share |
| 84 | * the same variables list, so only the first one is considered to own |
| 85 | * the variables) |
| 86 | */ |
| 87 | UBool variablesAreOwned; |
| 88 | |
| 89 | /** |
| 90 | * The character that represents variables[0]. Characters |
| 91 | * variablesBase through variablesBase + |
| 92 | * variablesLength - 1 represent UnicodeFunctor objects. |
| 93 | */ |
| 94 | UChar variablesBase; |
| 95 | |
| 96 | /** |
| 97 | * The length of variables. |
| 98 | */ |
| 99 | int32_t variablesLength; |
| 100 | |
| 101 | public: |
| 102 | |
| 103 | /** |
| 104 | * Constructor |
| 105 | * @param status Output param set to success/failure code on exit. |
| 106 | */ |
| 107 | TransliterationRuleData(UErrorCode& status); |
| 108 | |
| 109 | /** |
| 110 | * Copy Constructor |
| 111 | */ |
| 112 | TransliterationRuleData(const TransliterationRuleData&); |
| 113 | |
| 114 | /** |
| 115 | * destructor |
| 116 | */ |
| 117 | ~TransliterationRuleData(); |
| 118 | |
| 119 | /** |
| 120 | * Given a stand-in character, return the UnicodeFunctor that it |
| 121 | * represents, or NULL if it doesn't represent anything. |
| 122 | * @param standIn the given stand-in character. |
| 123 | * @return the UnicodeFunctor that 'standIn' represents |
| 124 | */ |
| 125 | UnicodeFunctor* lookup(UChar32 standIn) const; |
| 126 | |
| 127 | /** |
| 128 | * Given a stand-in character, return the UnicodeMatcher that it |
| 129 | * represents, or NULL if it doesn't represent anything or if it |
| 130 | * represents something that is not a matcher. |
| 131 | * @param standIn the given stand-in character. |
| 132 | * @return return the UnicodeMatcher that 'standIn' represents |
| 133 | */ |
| 134 | UnicodeMatcher* lookupMatcher(UChar32 standIn) const; |
| 135 | |
| 136 | /** |
| 137 | * Given a stand-in character, return the UnicodeReplacer that it |
| 138 | * represents, or NULL if it doesn't represent anything or if it |
| 139 | * represents something that is not a replacer. |
| 140 | * @param standIn the given stand-in character. |
| 141 | * @return return the UnicodeReplacer that 'standIn' represents |
| 142 | */ |
| 143 | UnicodeReplacer* lookupReplacer(UChar32 standIn) const; |
| 144 | |
| 145 | |
| 146 | private: |
| 147 | TransliterationRuleData &operator=(const TransliterationRuleData &other); // forbid copying of this class |
| 148 | }; |
| 149 | |
| 150 | U_NAMESPACE_END |
| 151 | |
| 152 | #endif /* #if !UCONFIG_NO_TRANSLITERATION */ |
| 153 | |
| 154 | #endif |
| 155 | |