michael@0: /*
michael@0: **********************************************************************
michael@0: *   Copyright (c) 2001-2007, International Business Machines
michael@0: *   Corporation and others.  All Rights Reserved.
michael@0: **********************************************************************
michael@0: *   Date        Name        Description
michael@0: *   11/20/2001  aliu        Creation.
michael@0: **********************************************************************
michael@0: */
michael@0: #ifndef ESCTRN_H
michael@0: #define ESCTRN_H
michael@0: 
michael@0: #include "unicode/utypes.h"
michael@0: 
michael@0: #if !UCONFIG_NO_TRANSLITERATION
michael@0: 
michael@0: #include "unicode/translit.h"
michael@0: 
michael@0: U_NAMESPACE_BEGIN
michael@0: 
michael@0: /**
michael@0:  * A transliterator that converts Unicode characters to an escape
michael@0:  * form.  Examples of escape forms are "U+4E01" and "".
michael@0:  * Escape forms have a prefix and suffix, either of which may be
michael@0:  * empty, a radix, typically 16 or 10, a minimum digit count,
michael@0:  * typically 1, 4, or 8, and a boolean that specifies whether
michael@0:  * supplemental characters are handled as 32-bit code points or as two
michael@0:  * 16-bit code units.  Most escape forms handle 32-bit code points,
michael@0:  * but some, such as the Java form, intentionally break them into two
michael@0:  * surrogate pairs, for backward compatibility.
michael@0:  *
michael@0:  * <p>Some escape forms actually have two different patterns, one for
michael@0:  * BMP characters (0..FFFF) and one for supplements (>FFFF).  To
michael@0:  * handle this, a second EscapeTransliterator may be defined that
michael@0:  * specifies the pattern to be produced for supplementals.  An example
michael@0:  * of a form that requires this is the C form, which uses "\\uFFFF"
michael@0:  * for BMP characters and "\\U0010FFFF" for supplementals.
michael@0:  *
michael@0:  * <p>This class is package private.  It registers several standard
michael@0:  * variants with the system which are then accessed via their IDs.
michael@0:  *
michael@0:  * @author Alan Liu
michael@0:  */
michael@0: class EscapeTransliterator : public Transliterator {
michael@0: 
michael@0:  private:
michael@0: 
michael@0:     /**
michael@0:      * The prefix of the escape form; may be empty, but usually isn't.
michael@0:      */
michael@0:     UnicodeString prefix;
michael@0: 
michael@0:     /**
michael@0:      * The prefix of the escape form; often empty.
michael@0:      */
michael@0:     UnicodeString suffix;
michael@0: 
michael@0:     /**
michael@0:      * The radix to display the number in.  Typically 16 or 10.  Must
michael@0:      * be in the range 2 to 36.
michael@0:      */
michael@0:     int32_t radix;
michael@0: 
michael@0:     /**
michael@0:      * The minimum number of digits.  Typically 1, 4, or 8.  Values
michael@0:      * less than 1 are equivalent to 1.
michael@0:      */
michael@0:     int32_t minDigits;
michael@0: 
michael@0:     /**
michael@0:      * If true, supplementals are handled as 32-bit code points.  If
michael@0:      * false, they are handled as two 16-bit code units.
michael@0:      */
michael@0:     UBool grokSupplementals;
michael@0: 
michael@0:     /**
michael@0:      * The form to be used for supplementals.  If this is null then
michael@0:      * the same form is used for BMP characters and supplementals.  If
michael@0:      * this is not null and if grokSupplementals is true then the
michael@0:      * prefix, suffix, radix, and minDigits of this object are used
michael@0:      * for supplementals.  This pointer is owned.
michael@0:      */
michael@0:     EscapeTransliterator* supplementalHandler;
michael@0: 
michael@0:  public:
michael@0: 
michael@0:     /**
michael@0:      * Registers standard variants with the system.  Called by
michael@0:      * Transliterator during initialization.
michael@0:      */
michael@0:     static void registerIDs();
michael@0: 
michael@0:     /**
michael@0:      * Constructs an escape transliterator with the given ID and
michael@0:      * parameters.  See the class member documentation for details.
michael@0:      */
michael@0:     EscapeTransliterator(const UnicodeString& ID,
michael@0:                          const UnicodeString& prefix, const UnicodeString& suffix,
michael@0:                          int32_t radix, int32_t minDigits,
michael@0:                          UBool grokSupplementals,
michael@0:                          EscapeTransliterator* adoptedSupplementalHandler);
michael@0: 
michael@0:     /**
michael@0:      * Copy constructor.
michael@0:      */
michael@0:     EscapeTransliterator(const EscapeTransliterator&);
michael@0: 
michael@0:     /**
michael@0:      * Destructor.
michael@0:      */
michael@0:     virtual ~EscapeTransliterator();
michael@0: 
michael@0:     /**
michael@0:      * Transliterator API.
michael@0:      */
michael@0:     virtual Transliterator* clone() const;
michael@0: 
michael@0:     /**
michael@0:      * ICU "poor man's RTTI", returns a UClassID for the actual class.
michael@0:      */
michael@0:     virtual UClassID getDynamicClassID() const;
michael@0: 
michael@0:     /**
michael@0:      * ICU "poor man's RTTI", returns a UClassID for this class.
michael@0:      */
michael@0:     U_I18N_API static UClassID U_EXPORT2 getStaticClassID();
michael@0: 
michael@0:  protected:
michael@0: 
michael@0:     /**
michael@0:      * Implements {@link Transliterator#handleTransliterate}.
michael@0:      */
michael@0:     virtual void handleTransliterate(Replaceable& text, UTransPosition& offset,
michael@0:                              UBool isIncremental) const;
michael@0: 
michael@0: };
michael@0: 
michael@0: U_NAMESPACE_END
michael@0: 
michael@0: #endif /* #if !UCONFIG_NO_TRANSLITERATION */
michael@0: 
michael@0: #endif