1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/intl/icu/source/i18n/strmatch.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,250 @@
1.4 +/*
1.5 + * Copyright (C) 2001-2011, International Business Machines Corporation
1.6 + * and others. All Rights Reserved.
1.7 + **********************************************************************
1.8 + * Date Name Description
1.9 + * 07/23/01 aliu Creation.
1.10 + **********************************************************************
1.11 + */
1.12 +#ifndef STRMATCH_H
1.13 +#define STRMATCH_H
1.14 +
1.15 +#include "unicode/utypes.h"
1.16 +
1.17 +#if !UCONFIG_NO_TRANSLITERATION
1.18 +
1.19 +#include "unicode/unistr.h"
1.20 +#include "unicode/unifunct.h"
1.21 +#include "unicode/unimatch.h"
1.22 +#include "unicode/unirepl.h"
1.23 +
1.24 +U_NAMESPACE_BEGIN
1.25 +
1.26 +class TransliterationRuleData;
1.27 +
1.28 +/**
1.29 + * An object that matches a fixed input string, implementing the
1.30 + * UnicodeMatcher API. This object also implements the
1.31 + * UnicodeReplacer API, allowing it to emit the matched text as
1.32 + * output. Since the match text may contain flexible match elements,
1.33 + * such as UnicodeSets, the emitted text is not the match pattern, but
1.34 + * instead a substring of the actual matched text. Following
1.35 + * convention, the output text is the leftmost match seen up to this
1.36 + * point.
1.37 + *
1.38 + * A StringMatcher may represent a segment, in which case it has a
1.39 + * positive segment number. This affects how the matcher converts
1.40 + * itself to a pattern but does not otherwise affect its function.
1.41 + *
1.42 + * A StringMatcher that is not a segment should not be used as a
1.43 + * UnicodeReplacer.
1.44 + */
1.45 +class StringMatcher : public UnicodeFunctor, public UnicodeMatcher, public UnicodeReplacer {
1.46 +
1.47 + public:
1.48 +
1.49 + /**
1.50 + * Construct a matcher that matches the given pattern string.
1.51 + * @param string the pattern to be matched, possibly containing
1.52 + * stand-ins that represent nested UnicodeMatcher objects.
1.53 + * @param start inclusive start index of text to be replaced
1.54 + * @param limit exclusive end index of text to be replaced;
1.55 + * must be greater than or equal to start
1.56 + * @param segmentNum the segment number from 1..n, or 0 if this is
1.57 + * not a segment.
1.58 + * @param data context object mapping stand-ins to
1.59 + * UnicodeMatcher objects.
1.60 + */
1.61 + StringMatcher(const UnicodeString& string,
1.62 + int32_t start,
1.63 + int32_t limit,
1.64 + int32_t segmentNum,
1.65 + const TransliterationRuleData& data);
1.66 +
1.67 + /**
1.68 + * Copy constructor
1.69 + * @param o the object to be copied.
1.70 + */
1.71 + StringMatcher(const StringMatcher& o);
1.72 +
1.73 + /**
1.74 + * Destructor
1.75 + */
1.76 + virtual ~StringMatcher();
1.77 +
1.78 + /**
1.79 + * Implement UnicodeFunctor
1.80 + * @return a copy of the object.
1.81 + */
1.82 + virtual UnicodeFunctor* clone() const;
1.83 +
1.84 + /**
1.85 + * UnicodeFunctor API. Cast 'this' to a UnicodeMatcher* pointer
1.86 + * and return the pointer.
1.87 + * @return the UnicodeMatcher point.
1.88 + */
1.89 + virtual UnicodeMatcher* toMatcher() const;
1.90 +
1.91 + /**
1.92 + * UnicodeFunctor API. Cast 'this' to a UnicodeReplacer* pointer
1.93 + * and return the pointer.
1.94 + * @return the UnicodeReplacer pointer.
1.95 + */
1.96 + virtual UnicodeReplacer* toReplacer() const;
1.97 +
1.98 + /**
1.99 + * Implement UnicodeMatcher
1.100 + * @param text the text to be matched
1.101 + * @param offset on input, the index into text at which to begin
1.102 + * matching. On output, the limit of the matched text. The
1.103 + * number of matched characters is the output value of offset
1.104 + * minus the input value. Offset should always point to the
1.105 + * HIGH SURROGATE (leading code unit) of a pair of surrogates,
1.106 + * both on entry and upon return.
1.107 + * @param limit the limit index of text to be matched. Greater
1.108 + * than offset for a forward direction match, less than offset for
1.109 + * a backward direction match. The last character to be
1.110 + * considered for matching will be text.charAt(limit-1) in the
1.111 + * forward direction or text.charAt(limit+1) in the backward
1.112 + * direction.
1.113 + * @param incremental if TRUE, then assume further characters may
1.114 + * be inserted at limit and check for partial matching. Otherwise
1.115 + * assume the text as given is complete.
1.116 + * @return a match degree value indicating a full match, a partial
1.117 + * match, or a mismatch. If incremental is FALSE then
1.118 + * U_PARTIAL_MATCH should never be returned.
1.119 + */
1.120 + virtual UMatchDegree matches(const Replaceable& text,
1.121 + int32_t& offset,
1.122 + int32_t limit,
1.123 + UBool incremental);
1.124 +
1.125 + /**
1.126 + * Implement UnicodeMatcher
1.127 + * @param result Output param to receive the pattern.
1.128 + * @param escapeUnprintable if True then escape the unprintable characters.
1.129 + * @return A reference to 'result'.
1.130 + */
1.131 + virtual UnicodeString& toPattern(UnicodeString& result,
1.132 + UBool escapeUnprintable = FALSE) const;
1.133 +
1.134 + /**
1.135 + * Implement UnicodeMatcher
1.136 + * Returns TRUE if this matcher will match a character c, where c
1.137 + * & 0xFF == v, at offset, in the forward direction (with limit >
1.138 + * offset). This is used by <tt>RuleBasedTransliterator</tt> for
1.139 + * indexing.
1.140 + * @param v the given value
1.141 + * @return TRUE if this matcher will match a character c,
1.142 + * where c & 0xFF == v
1.143 + */
1.144 + virtual UBool matchesIndexValue(uint8_t v) const;
1.145 +
1.146 + /**
1.147 + * Implement UnicodeMatcher
1.148 + */
1.149 + virtual void addMatchSetTo(UnicodeSet& toUnionTo) const;
1.150 +
1.151 + /**
1.152 + * Implement UnicodeFunctor
1.153 + */
1.154 + virtual void setData(const TransliterationRuleData*);
1.155 +
1.156 + /**
1.157 + * Replace characters in 'text' from 'start' to 'limit' with the
1.158 + * output text of this object. Update the 'cursor' parameter to
1.159 + * give the cursor position and return the length of the
1.160 + * replacement text.
1.161 + *
1.162 + * @param text the text to be matched
1.163 + * @param start inclusive start index of text to be replaced
1.164 + * @param limit exclusive end index of text to be replaced;
1.165 + * must be greater than or equal to start
1.166 + * @param cursor output parameter for the cursor position.
1.167 + * Not all replacer objects will update this, but in a complete
1.168 + * tree of replacer objects, representing the entire output side
1.169 + * of a transliteration rule, at least one must update it.
1.170 + * @return the number of 16-bit code units in the text replacing
1.171 + * the characters at offsets start..(limit-1) in text
1.172 + */
1.173 + virtual int32_t replace(Replaceable& text,
1.174 + int32_t start,
1.175 + int32_t limit,
1.176 + int32_t& cursor);
1.177 +
1.178 + /**
1.179 + * Returns a string representation of this replacer. If the
1.180 + * result of calling this function is passed to the appropriate
1.181 + * parser, typically TransliteratorParser, it will produce another
1.182 + * replacer that is equal to this one.
1.183 + * @param result the string to receive the pattern. Previous
1.184 + * contents will be deleted.
1.185 + * @param escapeUnprintable if TRUE then convert unprintable
1.186 + * character to their hex escape representations, \\uxxxx or
1.187 + * \\Uxxxxxxxx. Unprintable characters are defined by
1.188 + * Utility.isUnprintable().
1.189 + * @return a reference to 'result'.
1.190 + */
1.191 + virtual UnicodeString& toReplacerPattern(UnicodeString& result,
1.192 + UBool escapeUnprintable) const;
1.193 +
1.194 + /**
1.195 + * Remove any match data. This must be called before performing a
1.196 + * set of matches with this segment.
1.197 + */
1.198 + void resetMatch();
1.199 +
1.200 + /**
1.201 + * ICU "poor man's RTTI", returns a UClassID for the actual class.
1.202 + */
1.203 + virtual UClassID getDynamicClassID() const;
1.204 +
1.205 + /**
1.206 + * ICU "poor man's RTTI", returns a UClassID for this class.
1.207 + */
1.208 + static UClassID U_EXPORT2 getStaticClassID();
1.209 +
1.210 + /**
1.211 + * Union the set of all characters that may output by this object
1.212 + * into the given set.
1.213 + * @param toUnionTo the set into which to union the output characters
1.214 + */
1.215 + virtual void addReplacementSetTo(UnicodeSet& toUnionTo) const;
1.216 +
1.217 + private:
1.218 +
1.219 + /**
1.220 + * The text to be matched.
1.221 + */
1.222 + UnicodeString pattern;
1.223 +
1.224 + /**
1.225 + * Context object that maps stand-ins to matcher and replacer
1.226 + * objects.
1.227 + */
1.228 + const TransliterationRuleData* data;
1.229 +
1.230 + /**
1.231 + * The segment number, 1-based, or 0 if not a segment.
1.232 + */
1.233 + int32_t segmentNumber;
1.234 +
1.235 + /**
1.236 + * Start offset, in the match text, of the <em>rightmost</em>
1.237 + * match.
1.238 + */
1.239 + int32_t matchStart;
1.240 +
1.241 + /**
1.242 + * Limit offset, in the match text, of the <em>rightmost</em>
1.243 + * match.
1.244 + */
1.245 + int32_t matchLimit;
1.246 +
1.247 +};
1.248 +
1.249 +U_NAMESPACE_END
1.250 +
1.251 +#endif /* #if !UCONFIG_NO_TRANSLITERATION */
1.252 +
1.253 +#endif
