michael@0: /*
michael@0: **********************************************************************
michael@0: * Copyright (c) 2001-2012, International Business Machines
michael@0: * Corporation and others.  All Rights Reserved.
michael@0: **********************************************************************
michael@0: *   Date        Name        Description
michael@0: *   07/18/01    aliu        Creation.
michael@0: **********************************************************************
michael@0: */
michael@0: 
michael@0: #include "unicode/unifilt.h"
michael@0: #include "unicode/rep.h"
michael@0: #include "unicode/utf16.h"
michael@0: 
michael@0: U_NAMESPACE_BEGIN
michael@0: UOBJECT_DEFINE_ABSTRACT_RTTI_IMPLEMENTATION(UnicodeFilter)
michael@0: 
michael@0: 
michael@0: /* Define this here due to the lack of another file.
michael@0:    It can't be defined in the header */
michael@0: UnicodeMatcher::~UnicodeMatcher() {}
michael@0: 
michael@0: UnicodeFilter::~UnicodeFilter() {}
michael@0: 
michael@0: /**
michael@0:  * UnicodeFunctor API.
michael@0:  *   Note that UnicodeMatcher is a base class of UnicodeFilter.
michael@0:  */
michael@0: UnicodeMatcher* UnicodeFilter::toMatcher() const {
michael@0:   return const_cast<UnicodeFilter *>(this);
michael@0: }
michael@0: 
michael@0: void UnicodeFilter::setData(const TransliterationRuleData*) {}
michael@0: 
michael@0: /**
michael@0:  * Default implementation of UnicodeMatcher::matches() for Unicode
michael@0:  * filters.  Matches a single code point at offset (either one or
michael@0:  * two 16-bit code units).
michael@0:  */
michael@0: UMatchDegree UnicodeFilter::matches(const Replaceable& text,
michael@0:                                     int32_t& offset,
michael@0:                                     int32_t limit,
michael@0:                                     UBool incremental) {
michael@0:     UChar32 c;
michael@0:     if (offset < limit &&
michael@0:         contains(c = text.char32At(offset))) {
michael@0:         offset += U16_LENGTH(c);
michael@0:         return U_MATCH;
michael@0:     }
michael@0:     if (offset > limit &&
michael@0:         contains(c = text.char32At(offset))) {
michael@0:         // Backup offset by 1, unless the preceding character is a
michael@0:         // surrogate pair -- then backup by 2 (keep offset pointing at
michael@0:         // the lead surrogate).
michael@0:         --offset;
michael@0:         if (offset >= 0) {
michael@0:             offset -= U16_LENGTH(text.char32At(offset)) - 1;
michael@0:         }
michael@0:         return U_MATCH;
michael@0:     }
michael@0:     if (incremental && offset == limit) {
michael@0:         return U_PARTIAL_MATCH;
michael@0:     }
michael@0:     return U_MISMATCH;
michael@0: }
michael@0: 
michael@0: U_NAMESPACE_END
michael@0: 
michael@0: //eof