1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/i18n/rbt_data.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,152 @@
1.4 +/*
1.5 +**********************************************************************
1.6 +* Copyright (C) 1999-2007, International Business Machines Corporation
1.7 +* and others. All Rights Reserved.
1.8 +**********************************************************************
1.9 +* Date Name Description
1.10 +* 11/17/99 aliu Creation.
1.11 +**********************************************************************
1.12 +*/
1.13 +#ifndef RBT_DATA_H
1.14 +#define RBT_DATA_H
1.15 +
1.16 +#include "unicode/utypes.h"
1.17 +#include "unicode/uclean.h"
1.18 +
1.19 +#if !UCONFIG_NO_TRANSLITERATION
1.20 +
1.21 +#include "unicode/uobject.h"
1.22 +#include "rbt_set.h"
1.23 +#include "hash.h"
1.24 +
1.25 +U_NAMESPACE_BEGIN
1.26 +
1.27 +class UnicodeFunctor;
1.28 +class UnicodeMatcher;
1.29 +class UnicodeReplacer;
1.30 +
1.31 +/**
1.32 + * The rule data for a RuleBasedTransliterators. RBT objects hold
1.33 + * a const pointer to a TRD object that they do not own. TRD objects
1.34 + * are essentially the parsed rules in compact, usable form. The
1.35 + * TRD objects themselves are held for the life of the process in
1.36 + * a static cache owned by Transliterator.
1.37 + *
1.38 + * This class' API is a little asymmetric. There is a method to
1.39 + * define a variable, but no way to define a set. This is because the
1.40 + * sets are defined by the parser in a UVector, and the vector is
1.41 + * copied into a fixed-size array here. Once this is done, no new
1.42 + * sets may be defined. In practice, there is no need to do so, since
1.43 + * generating the data and using it are discrete phases. When there
1.44 + * is a need to access the set data during the parse phase, another
1.45 + * data structure handles this. See the parsing code for more
1.46 + * details.
1.47 + */
1.48 +class TransliterationRuleData : public UMemory {
1.49 +
1.50 +public:
1.51 +
1.52 + // PUBLIC DATA MEMBERS
1.53 +
1.54 + /**
1.55 + * Rule table. May be empty.
1.56 + */
1.57 + TransliterationRuleSet ruleSet;
1.58 +
1.59 + /**
1.60 + * Map variable name (String) to variable (UnicodeString). A variable name
1.61 + * corresponds to zero or more characters, stored in a UnicodeString in
1.62 + * this hash. One or more of these chars may also correspond to a
1.63 + * UnicodeMatcher, in which case the character in the UnicodeString in this hash is
1.64 + * a stand-in: it is an index for a secondary lookup in
1.65 + * data.variables. The stand-in also represents the UnicodeMatcher in
1.66 + * the stored rules.
1.67 + */
1.68 + Hashtable variableNames;
1.69 +
1.70 + /**
1.71 + * Map category variable (UChar) to set (UnicodeFunctor).
1.72 + * Variables that correspond to a set of characters are mapped
1.73 + * from variable name to a stand-in character in data.variableNames.
1.74 + * The stand-in then serves as a key in this hash to lookup the
1.75 + * actual UnicodeFunctor object. In addition, the stand-in is
1.76 + * stored in the rule text to represent the set of characters.
1.77 + * variables[i] represents character (variablesBase + i).
1.78 + */
1.79 + UnicodeFunctor** variables;
1.80 +
1.81 + /**
1.82 + * Flag that indicates whether the variables are owned (if a single
1.83 + * call to Transliterator::createFromRules() produces a CompoundTransliterator
1.84 + * with more than one RuleBasedTransliterator as children, they all share
1.85 + * the same variables list, so only the first one is considered to own
1.86 + * the variables)
1.87 + */
1.88 + UBool variablesAreOwned;
1.89 +
1.90 + /**
1.91 + * The character that represents variables[0]. Characters
1.92 + * variablesBase through variablesBase +
1.93 + * variablesLength - 1 represent UnicodeFunctor objects.
1.94 + */
1.95 + UChar variablesBase;
1.96 +
1.97 + /**
1.98 + * The length of variables.
1.99 + */
1.100 + int32_t variablesLength;
1.101 +
1.102 +public:
1.103 +
1.104 + /**
1.105 + * Constructor
1.106 + * @param status Output param set to success/failure code on exit.
1.107 + */
1.108 + TransliterationRuleData(UErrorCode& status);
1.109 +
1.110 + /**
1.111 + * Copy Constructor
1.112 + */
1.113 + TransliterationRuleData(const TransliterationRuleData&);
1.114 +
1.115 + /**
1.116 + * destructor
1.117 + */
1.118 + ~TransliterationRuleData();
1.119 +
1.120 + /**
1.121 + * Given a stand-in character, return the UnicodeFunctor that it
1.122 + * represents, or NULL if it doesn't represent anything.
1.123 + * @param standIn the given stand-in character.
1.124 + * @return the UnicodeFunctor that 'standIn' represents
1.125 + */
1.126 + UnicodeFunctor* lookup(UChar32 standIn) const;
1.127 +
1.128 + /**
1.129 + * Given a stand-in character, return the UnicodeMatcher that it
1.130 + * represents, or NULL if it doesn't represent anything or if it
1.131 + * represents something that is not a matcher.
1.132 + * @param standIn the given stand-in character.
1.133 + * @return return the UnicodeMatcher that 'standIn' represents
1.134 + */
1.135 + UnicodeMatcher* lookupMatcher(UChar32 standIn) const;
1.136 +
1.137 + /**
1.138 + * Given a stand-in character, return the UnicodeReplacer that it
1.139 + * represents, or NULL if it doesn't represent anything or if it
1.140 + * represents something that is not a replacer.
1.141 + * @param standIn the given stand-in character.
1.142 + * @return return the UnicodeReplacer that 'standIn' represents
1.143 + */
1.144 + UnicodeReplacer* lookupReplacer(UChar32 standIn) const;
1.145 +
1.146 +
1.147 +private:
1.148 + TransliterationRuleData &operator=(const TransliterationRuleData &other); // forbid copying of this class
1.149 +};
1.150 +
1.151 +U_NAMESPACE_END
1.152 +
1.153 +#endif /* #if !UCONFIG_NO_TRANSLITERATION */
1.154 +
1.155 +#endif
