i18npool/inc/xdictionary.hxx

   1 /*************************************************************************
   2  *
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * Copyright 2000, 2010 Oracle and/or its affiliates.
   6  *
   7  * OpenOffice.org - a multi-platform office productivity suite
   8  *
   9  * This file is part of OpenOffice.org.
  10  *
  11  * OpenOffice.org is free software: you can redistribute it and/or modify
  12  * it under the terms of the GNU Lesser General Public License version 3
  13  * only, as published by the Free Software Foundation.
  14  *
  15  * OpenOffice.org is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18  * GNU Lesser General Public License version 3 for more details
  19  * (a copy is included in the LICENSE file that accompanied this code).
  20  *
  21  * You should have received a copy of the GNU Lesser General Public License
  22  * version 3 along with OpenOffice.org.  If not, see
  23  * <http://www.openoffice.org/license.html>
  24  * for a copy of the LGPLv3 License.
  25  *
  26  ************************************************************************/
  27 #ifndef _XDICTIONARY_H_
  28 #define _XDICTIONARY_H_
  29
  30 #include <sal/types.h>
  31 #include <osl/module.h>
  32
  33 #include <com/sun/star/i18n/Boundary.hpp>
  34
  35 namespace com { namespace sun { namespace star { namespace i18n {
  36
  37 // Whether to use cell boundary code, currently unused but prepared.
  38 #define USE_CELL_BOUNDARY_CODE 0
  39
  40 #define CACHE_MAX 32        // max cache structure number
  41 #define DEFAULT_SIZE 256    // for boundary size, to avoid alloc and release memory
  42
  43 // cache structure.
  44 struct WordBreakCache {
  45     sal_Int32 length;       // contents length saved here.
  46     sal_Unicode *contents;      // seperated segment contents.
  47     sal_Int32* wordboundary;        // word boundaries in segments.
  48     sal_Int32 size;         // size of wordboundary
  49
  50     WordBreakCache();
  51     sal_Bool equals(const sal_Unicode *str, Boundary& boundary);    // checking cached string
  52 };
  53
  54 class xdictionary
  55 {
  56 private:
  57     const sal_uInt8  * existMark;
  58     const sal_Int16  * index1;
  59     const sal_Int32 * index2;
  60     const sal_Int32 * lenArray;
  61     const sal_Unicode* dataArea;
  62     oslModule hModule;
  63     Boundary boundary;
  64     sal_Bool japaneseWordBreak;
  65
  66 #if USE_CELL_BOUNDARY_CODE
  67     // For CTL breakiterator, where the word boundary should not be inside cell.
  68     sal_Bool useCellBoundary;
  69     sal_Int32* cellBoundary;
  70 #endif
  71
  72 public:
  73     xdictionary(const sal_Char *lang);
  74     ~xdictionary();
  75     Boundary nextWord( const rtl::OUString& rText, sal_Int32 nPos, sal_Int16 wordType);
  76     Boundary previousWord( const rtl::OUString& rText, sal_Int32 nPos, sal_Int16 wordType);
  77     Boundary getWordBoundary( const rtl::OUString& rText, sal_Int32 nPos, sal_Int16 wordType, sal_Bool bDirection );
  78     void setJapaneseWordBreak();
  79
  80 #if USE_CELL_BOUNDARY_CODE
  81     void setCellBoundary(sal_Int32* cellArray);
  82 #endif
  83
  84 private:
  85     WordBreakCache cache[CACHE_MAX];
  86
  87     sal_Bool        seekSegment(const rtl::OUString& rText, sal_Int32 pos, Boundary& boundary);
  88     WordBreakCache& getCache(const sal_Unicode *text, Boundary& boundary);
  89     sal_Bool        exists(const sal_uInt32 u);
  90     sal_Int32       getLongestMatch(const sal_Unicode *text, sal_Int32 len);
  91 };
  92
  93 } } } }
  94
  95 #endif
  96