caniter.h

Go to the documentation of this file.

// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
 *******************************************************************************
 * Copyright (C) 1996-2014, International Business Machines Corporation and
 * others. All Rights Reserved.
 *******************************************************************************
 */

#ifndef CANITER_H
#define CANITER_H

#include "unicode/utypes.h"

#if !UCONFIG_NO_NORMALIZATION

#include "unicode/uobject.h"
#include "unicode/unistr.h"

#ifndef CANITER_SKIP_ZEROES
#define CANITER_SKIP_ZEROES TRUE
#endif

U_NAMESPACE_BEGIN

class Hashtable;
class Normalizer2;
class Normalizer2Impl;

class U_COMMON_API CanonicalIterator U_FINAL : public UObject {
public:
    CanonicalIterator(const UnicodeString &source, UErrorCode &status);

    virtual ~CanonicalIterator();

    UnicodeString getSource();

    void reset();

    UnicodeString next();

    void setSource(const UnicodeString &newSource, UErrorCode &status);

#ifndef U_HIDE_INTERNAL_API

    static void U_EXPORT2 permute(UnicodeString &source, UBool skipZeros, Hashtable *result, UErrorCode &status);
#endif  /* U_HIDE_INTERNAL_API */

    static UClassID U_EXPORT2 getStaticClassID();

    virtual UClassID getDynamicClassID() const;

private:
    // ===================== PRIVATES ==============================
    // private default constructor
    CanonicalIterator();


    CanonicalIterator(const CanonicalIterator& other);

    CanonicalIterator& operator=(const CanonicalIterator& other);

    // fields
    UnicodeString source;
    UBool done;

    // 2 dimensional array holds the pieces of the string with
    // their different canonically equivalent representations
    UnicodeString **pieces;
    int32_t pieces_length;
    int32_t *pieces_lengths;

    // current is used in iterating to combine pieces
    int32_t *current;
    int32_t current_length;

    // transient fields
    UnicodeString buffer;

    const Normalizer2 &nfd;
    const Normalizer2Impl &nfcImpl;

    // we have a segment, in NFD. Find all the strings that are canonically equivalent to it.
    UnicodeString *getEquivalents(const UnicodeString &segment, int32_t &result_len, UErrorCode &status); //private String[] getEquivalents(String segment)

    //Set getEquivalents2(String segment);
    Hashtable *getEquivalents2(Hashtable *fillinResult, const char16_t *segment, int32_t segLen, UErrorCode &status);
    //Hashtable *getEquivalents2(const UnicodeString &segment, int32_t segLen, UErrorCode &status);

    //Set extract(int comp, String segment, int segmentPos, StringBuffer buffer);
    Hashtable *extract(Hashtable *fillinResult, UChar32 comp, const char16_t *segment, int32_t segLen, int32_t segmentPos, UErrorCode &status);
    //Hashtable *extract(UChar32 comp, const UnicodeString &segment, int32_t segLen, int32_t segmentPos, UErrorCode &status);

    void cleanPieces();

};

U_NAMESPACE_END

#endif /* #if !UCONFIG_NO_NORMALIZATION */

#endif