michael@0: /*
michael@0: **********************************************************************
michael@0: * Copyright (C) 1999-2010, International Business Machines Corporation and others.
michael@0: * All Rights Reserved.
michael@0: **********************************************************************
michael@0: *   Date        Name        Description
michael@0: *   11/17/99    aliu        Creation.
michael@0: **********************************************************************
michael@0: */
michael@0: #ifndef UNIFILT_H
michael@0: #define UNIFILT_H
michael@0: 
michael@0: #include "unicode/unifunct.h"
michael@0: #include "unicode/unimatch.h"
michael@0: 
michael@0: /**
michael@0:  * \file 
michael@0:  * \brief C++ API: Unicode Filter
michael@0:  */
michael@0: 
michael@0: U_NAMESPACE_BEGIN
michael@0: 
michael@0: /**
michael@0:  * U_ETHER is used to represent character values for positions outside
michael@0:  * a range.  For example, transliterator uses this to represent
michael@0:  * characters outside the range contextStart..contextLimit-1.  This
michael@0:  * allows explicit matching by rules and UnicodeSets of text outside a
michael@0:  * defined range.
michael@0:  * @stable ICU 3.0
michael@0:  */
michael@0: #define U_ETHER ((UChar)0xFFFF)
michael@0: 
michael@0: /**
michael@0:  *
michael@0:  * <code>UnicodeFilter</code> defines a protocol for selecting a
michael@0:  * subset of the full range (U+0000 to U+10FFFF) of Unicode characters.
michael@0:  * Currently, filters are used in conjunction with classes like {@link
michael@0:  * Transliterator} to only process selected characters through a
michael@0:  * transformation.
michael@0:  *
michael@0:  * <p>Note: UnicodeFilter currently stubs out two pure virtual methods
michael@0:  * of its base class, UnicodeMatcher.  These methods are toPattern()
michael@0:  * and matchesIndexValue().  This is done so that filter classes that
michael@0:  * are not actually used as matchers -- specifically, those in the
michael@0:  * UnicodeFilterLogic component, and those in tests -- can continue to
michael@0:  * work without defining these methods.  As long as a filter is not
michael@0:  * used in an RBT during real transliteration, these methods will not
michael@0:  * be called.  However, this breaks the UnicodeMatcher base class
michael@0:  * protocol, and it is not a correct solution.
michael@0:  *
michael@0:  * <p>In the future we may revisit the UnicodeMatcher / UnicodeFilter
michael@0:  * hierarchy and either redesign it, or simply remove the stubs in
michael@0:  * UnicodeFilter and force subclasses to implement the full
michael@0:  * UnicodeMatcher protocol.
michael@0:  *
michael@0:  * @see UnicodeFilterLogic
michael@0:  * @stable ICU 2.0
michael@0:  */
michael@0: class U_COMMON_API UnicodeFilter : public UnicodeFunctor, public UnicodeMatcher {
michael@0: 
michael@0: public:
michael@0:     /**
michael@0:      * Destructor
michael@0:      * @stable ICU 2.0
michael@0:      */
michael@0:     virtual ~UnicodeFilter();
michael@0: 
michael@0:     /**
michael@0:      * Returns <tt>true</tt> for characters that are in the selected
michael@0:      * subset.  In other words, if a character is <b>to be
michael@0:      * filtered</b>, then <tt>contains()</tt> returns
michael@0:      * <b><tt>false</tt></b>.
michael@0:      * @stable ICU 2.0
michael@0:      */
michael@0:     virtual UBool contains(UChar32 c) const = 0;
michael@0: 
michael@0:     /**
michael@0:      * UnicodeFunctor API.  Cast 'this' to a UnicodeMatcher* pointer
michael@0:      * and return the pointer.
michael@0:      * @stable ICU 2.4
michael@0:      */
michael@0:     virtual UnicodeMatcher* toMatcher() const;
michael@0: 
michael@0:     /**
michael@0:      * Implement UnicodeMatcher API.
michael@0:      * @stable ICU 2.4
michael@0:      */
michael@0:     virtual UMatchDegree matches(const Replaceable& text,
michael@0:                                  int32_t& offset,
michael@0:                                  int32_t limit,
michael@0:                                  UBool incremental);
michael@0: 
michael@0:     /**
michael@0:      * UnicodeFunctor API.  Nothing to do.
michael@0:      * @stable ICU 2.4
michael@0:      */
michael@0:     virtual void setData(const TransliterationRuleData*);
michael@0: 
michael@0:     /**
michael@0:      * ICU "poor man's RTTI", returns a UClassID for this class.
michael@0:      *
michael@0:      * @stable ICU 2.2
michael@0:      */
michael@0:     static UClassID U_EXPORT2 getStaticClassID();
michael@0: 
michael@0: protected:
michael@0: 
michael@0:     /*
michael@0:      * Since this class has pure virtual functions,
michael@0:      * a constructor can't be used.
michael@0:      * @stable ICU 2.0
michael@0:      */
michael@0: /*    UnicodeFilter();*/
michael@0: };
michael@0: 
michael@0: /*inline UnicodeFilter::UnicodeFilter() {}*/
michael@0: 
michael@0: U_NAMESPACE_END
michael@0: 
michael@0: #endif