Home | History | Annotate | Download | only in unicode
      1 /*
      2  *******************************************************************************
      3  * Copyright (C) 1996-2010, International Business Machines Corporation and    *
      4  * others. All Rights Reserved.                                                *
      5  *******************************************************************************
      6  */
      7 
      8 #ifndef CANITER_H
      9 #define CANITER_H
     10 
     11 #include "unicode/utypes.h"
     12 
     13 #if !UCONFIG_NO_NORMALIZATION
     14 
     15 #include "unicode/uobject.h"
     16 #include "unicode/unistr.h"
     17 
     18 /**
     19  * \file
     20  * \brief C++ API: Canonical Iterator
     21  */
     22 
     23 /** Should permutation skip characters with combining class zero
     24  *  Should be either TRUE or FALSE. This is a compile time option
     25  *  @stable ICU 2.4
     26  */
     27 #ifndef CANITER_SKIP_ZEROES
     28 #define CANITER_SKIP_ZEROES TRUE
     29 #endif
     30 
     31 U_NAMESPACE_BEGIN
     32 
     33 class Hashtable;
     34 class Normalizer2;
     35 class Normalizer2Impl;
     36 
     37 /**
     38  * This class allows one to iterate through all the strings that are canonically equivalent to a given
     39  * string. For example, here are some sample results:
     40 Results for: {LATIN CAPITAL LETTER A WITH RING ABOVE}{LATIN SMALL LETTER D}{COMBINING DOT ABOVE}{COMBINING CEDILLA}
     41 1: \\u0041\\u030A\\u0064\\u0307\\u0327
     42  = {LATIN CAPITAL LETTER A}{COMBINING RING ABOVE}{LATIN SMALL LETTER D}{COMBINING DOT ABOVE}{COMBINING CEDILLA}
     43 2: \\u0041\\u030A\\u0064\\u0327\\u0307
     44  = {LATIN CAPITAL LETTER A}{COMBINING RING ABOVE}{LATIN SMALL LETTER D}{COMBINING CEDILLA}{COMBINING DOT ABOVE}
     45 3: \\u0041\\u030A\\u1E0B\\u0327
     46  = {LATIN CAPITAL LETTER A}{COMBINING RING ABOVE}{LATIN SMALL LETTER D WITH DOT ABOVE}{COMBINING CEDILLA}
     47 4: \\u0041\\u030A\\u1E11\\u0307
     48  = {LATIN CAPITAL LETTER A}{COMBINING RING ABOVE}{LATIN SMALL LETTER D WITH CEDILLA}{COMBINING DOT ABOVE}
     49 5: \\u00C5\\u0064\\u0307\\u0327
     50  = {LATIN CAPITAL LETTER A WITH RING ABOVE}{LATIN SMALL LETTER D}{COMBINING DOT ABOVE}{COMBINING CEDILLA}
     51 6: \\u00C5\\u0064\\u0327\\u0307
     52  = {LATIN CAPITAL LETTER A WITH RING ABOVE}{LATIN SMALL LETTER D}{COMBINING CEDILLA}{COMBINING DOT ABOVE}
     53 7: \\u00C5\\u1E0B\\u0327
     54  = {LATIN CAPITAL LETTER A WITH RING ABOVE}{LATIN SMALL LETTER D WITH DOT ABOVE}{COMBINING CEDILLA}
     55 8: \\u00C5\\u1E11\\u0307
     56  = {LATIN CAPITAL LETTER A WITH RING ABOVE}{LATIN SMALL LETTER D WITH CEDILLA}{COMBINING DOT ABOVE}
     57 9: \\u212B\\u0064\\u0307\\u0327
     58  = {ANGSTROM SIGN}{LATIN SMALL LETTER D}{COMBINING DOT ABOVE}{COMBINING CEDILLA}
     59 10: \\u212B\\u0064\\u0327\\u0307
     60  = {ANGSTROM SIGN}{LATIN SMALL LETTER D}{COMBINING CEDILLA}{COMBINING DOT ABOVE}
     61 11: \\u212B\\u1E0B\\u0327
     62  = {ANGSTROM SIGN}{LATIN SMALL LETTER D WITH DOT ABOVE}{COMBINING CEDILLA}
     63 12: \\u212B\\u1E11\\u0307
     64  = {ANGSTROM SIGN}{LATIN SMALL LETTER D WITH CEDILLA}{COMBINING DOT ABOVE}
     65  *<br>Note: the code is intended for use with small strings, and is not suitable for larger ones,
     66  * since it has not been optimized for that situation.
     67  * Note, CanonicalIterator is not intended to be subclassed.
     68  * @author M. Davis
     69  * @author C++ port by V. Weinstein
     70  * @stable ICU 2.4
     71  */
     72 class U_COMMON_API CanonicalIterator : public UObject {
     73 public:
     74     /**
     75      * Construct a CanonicalIterator object
     76      * @param source    string to get results for
     77      * @param status    Fill-in parameter which receives the status of this operation.
     78      * @stable ICU 2.4
     79      */
     80     CanonicalIterator(const UnicodeString &source, UErrorCode &status);
     81 
     82     /** Destructor
     83      *  Cleans pieces
     84      * @stable ICU 2.4
     85      */
     86     virtual ~CanonicalIterator();
     87 
     88     /**
     89      * Gets the NFD form of the current source we are iterating over.
     90      * @return gets the source: NOTE: it is the NFD form of source
     91      * @stable ICU 2.4
     92      */
     93     UnicodeString getSource();
     94 
     95     /**
     96      * Resets the iterator so that one can start again from the beginning.
     97      * @stable ICU 2.4
     98      */
     99     void reset();
    100 
    101     /**
    102      * Get the next canonically equivalent string.
    103      * <br><b>Warning: The strings are not guaranteed to be in any particular order.</b>
    104      * @return the next string that is canonically equivalent. A bogus string is returned when
    105      * the iteration is done.
    106      * @stable ICU 2.4
    107      */
    108     UnicodeString next();
    109 
    110     /**
    111      * Set a new source for this iterator. Allows object reuse.
    112      * @param newSource     the source string to iterate against. This allows the same iterator to be used
    113      *                     while changing the source string, saving object creation.
    114      * @param status        Fill-in parameter which receives the status of this operation.
    115      * @stable ICU 2.4
    116      */
    117     void setSource(const UnicodeString &newSource, UErrorCode &status);
    118 
    119     /**
    120      * Dumb recursive implementation of permutation.
    121      * TODO: optimize
    122      * @param source     the string to find permutations for
    123      * @param skipZeros  determine if skip zeros
    124      * @param result     the results in a set.
    125      * @param status       Fill-in parameter which receives the status of this operation.
    126      * @internal
    127      */
    128     static void U_EXPORT2 permute(UnicodeString &source, UBool skipZeros, Hashtable *result, UErrorCode &status);
    129 
    130     /**
    131      * ICU "poor man's RTTI", returns a UClassID for this class.
    132      *
    133      * @stable ICU 2.2
    134      */
    135     static UClassID U_EXPORT2 getStaticClassID();
    136 
    137     /**
    138      * ICU "poor man's RTTI", returns a UClassID for the actual class.
    139      *
    140      * @stable ICU 2.2
    141      */
    142     virtual UClassID getDynamicClassID() const;
    143 
    144 private:
    145     // ===================== PRIVATES ==============================
    146     // private default constructor
    147     CanonicalIterator();
    148 
    149 
    150     /**
    151      * Copy constructor. Private for now.
    152      * @internal
    153      */
    154     CanonicalIterator(const CanonicalIterator& other);
    155 
    156     /**
    157      * Assignment operator. Private for now.
    158      * @internal
    159      */
    160     CanonicalIterator& operator=(const CanonicalIterator& other);
    161 
    162     // fields
    163     UnicodeString source;
    164     UBool done;
    165 
    166     // 2 dimensional array holds the pieces of the string with
    167     // their different canonically equivalent representations
    168     UnicodeString **pieces;
    169     int32_t pieces_length;
    170     int32_t *pieces_lengths;
    171 
    172     // current is used in iterating to combine pieces
    173     int32_t *current;
    174     int32_t current_length;
    175 
    176     // transient fields
    177     UnicodeString buffer;
    178 
    179     const Normalizer2 &nfd;
    180     const Normalizer2Impl &nfcImpl;
    181 
    182     // we have a segment, in NFD. Find all the strings that are canonically equivalent to it.
    183     UnicodeString *getEquivalents(const UnicodeString &segment, int32_t &result_len, UErrorCode &status); //private String[] getEquivalents(String segment)
    184 
    185     //Set getEquivalents2(String segment);
    186     Hashtable *getEquivalents2(Hashtable *fillinResult, const UChar *segment, int32_t segLen, UErrorCode &status);
    187     //Hashtable *getEquivalents2(const UnicodeString &segment, int32_t segLen, UErrorCode &status);
    188 
    189     /**
    190      * See if the decomposition of cp2 is at segment starting at segmentPos
    191      * (with canonical rearrangment!)
    192      * If so, take the remainder, and return the equivalents
    193      */
    194     //Set extract(int comp, String segment, int segmentPos, StringBuffer buffer);
    195     Hashtable *extract(Hashtable *fillinResult, UChar32 comp, const UChar *segment, int32_t segLen, int32_t segmentPos, UErrorCode &status);
    196     //Hashtable *extract(UChar32 comp, const UnicodeString &segment, int32_t segLen, int32_t segmentPos, UErrorCode &status);
    197 
    198     void cleanPieces();
    199 
    200 };
    201 
    202 U_NAMESPACE_END
    203 
    204 #endif /* #if !UCONFIG_NO_NORMALIZATION */
    205 
    206 #endif
    207