icuSources/common/unifilt.cpp

   1 /*
   2 **********************************************************************
   3 * Copyright (c) 2001-2004, International Business Machines
   4 * Corporation and others.  All Rights Reserved.
   5 **********************************************************************
   6 *   Date        Name        Description
   7 *   07/18/01    aliu        Creation.
   8 **********************************************************************
   9 */
  10
  11 #include "unicode/unifilt.h"
  12 #include "unicode/rep.h"
  13
  14 U_NAMESPACE_BEGIN
  15 UOBJECT_DEFINE_ABSTRACT_RTTI_IMPLEMENTATION(UnicodeFilter)
  16
  17
  18 /* Define this here due to the lack of another file.
  19    It can't be defined in the header */
  20 UnicodeMatcher::~UnicodeMatcher() {}
  21
  22 UnicodeFilter::~UnicodeFilter() {}
  23
  24 /**
  25  * UnicodeFunctor API.  Cast 'this' to a UnicodeMatcher* pointer
  26  * and return the pointer.
  27  */
  28 UnicodeMatcher* UnicodeFilter::toMatcher() const {
  29     return (UnicodeMatcher*) this;
  30 }
  31
  32 void UnicodeFilter::setData(const TransliterationRuleData*) {}
  33
  34 /**
  35  * Default implementation of UnicodeMatcher::matches() for Unicode
  36  * filters.  Matches a single code point at offset (either one or
  37  * two 16-bit code units).
  38  */
  39 UMatchDegree UnicodeFilter::matches(const Replaceable& text,
  40                                     int32_t& offset,
  41                                     int32_t limit,
  42                                     UBool incremental) {
  43     UChar32 c;
  44     if (offset < limit &&
  45         contains(c = text.char32At(offset))) {
  46         offset += UTF_CHAR_LENGTH(c);
  47         return U_MATCH;
  48     }
  49     if (offset > limit &&
  50         contains(c = text.char32At(offset))) {
  51         // Backup offset by 1, unless the preceding character is a
  52         // surrogate pair -- then backup by 2 (keep offset pointing at
  53         // the lead surrogate).
  54         --offset;
  55         if (offset >= 0) {
  56             offset -= UTF_CHAR_LENGTH(text.char32At(offset)) - 1;
  57         }
  58         return U_MATCH;
  59     }
  60     if (incremental && offset == limit) {
  61         return U_PARTIAL_MATCH;
  62     }
  63     return U_MISMATCH;
  64 }
  65
  66 U_NAMESPACE_END
  67
  68 //eof