Line data Source code
1 : /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
2 : /*
3 : * This file is part of the LibreOffice project.
4 : *
5 : * This Source Code Form is subject to the terms of the Mozilla Public
6 : * License, v. 2.0. If a copy of the MPL was not distributed with this
7 : * file, You can obtain one at http://mozilla.org/MPL/2.0/.
8 : *
9 : * This file incorporates work covered by the following license notice:
10 : *
11 : * Licensed to the Apache Software Foundation (ASF) under one or more
12 : * contributor license agreements. See the NOTICE file distributed
13 : * with this work for additional information regarding copyright
14 : * ownership. The ASF licenses this file to you under the Apache
15 : * License, Version 2.0 (the "License"); you may not use this file
16 : * except in compliance with the License. You may obtain a copy of
17 : * the License at http://www.apache.org/licenses/LICENSE-2.0 .
18 : */
19 : #ifndef INCLUDED_I18NPOOL_INC_XDICTIONARY_HXX
20 : #define INCLUDED_I18NPOOL_INC_XDICTIONARY_HXX
21 :
22 : #include <sal/types.h>
23 : #include <osl/module.h>
24 :
25 : #include <com/sun/star/i18n/Boundary.hpp>
26 :
27 : namespace com { namespace sun { namespace star { namespace i18n {
28 :
29 : #define CACHE_MAX 32 // max cache structure number
30 : #define DEFAULT_SIZE 256 // for boundary size, to avoid alloc and release memory
31 :
32 : // cache structure.
33 : struct WordBreakCache {
34 : sal_Int32 length; // contents length saved here.
35 : sal_Unicode *contents; // separated segment contents.
36 : sal_Int32* wordboundary; // word boundaries in segments.
37 : sal_Int32 size; // size of wordboundary
38 :
39 : WordBreakCache();
40 : bool equals(const sal_Unicode *str, Boundary& boundary); // checking cached string
41 : };
42 :
43 : struct xdictionarydata
44 : {
45 : const sal_uInt8 * existMark;
46 : const sal_Int16 * index1;
47 : const sal_Int32 * index2;
48 : const sal_Int32 * lenArray;
49 : const sal_Unicode* dataArea;
50 120 : xdictionarydata() :
51 : existMark( NULL ),
52 : index1( NULL ),
53 : index2( NULL ),
54 : lenArray( NULL ),
55 120 : dataArea( NULL )
56 : {
57 120 : }
58 : };
59 :
60 : class xdictionary
61 : {
62 : private:
63 : xdictionarydata data;
64 : void initDictionaryData(const sal_Char *lang);
65 :
66 : Boundary boundary;
67 : bool japaneseWordBreak;
68 :
69 : public:
70 : xdictionary(const sal_Char *lang);
71 : ~xdictionary();
72 : Boundary nextWord( const OUString& rText, sal_Int32 nPos, sal_Int16 wordType);
73 : Boundary previousWord( const OUString& rText, sal_Int32 nPos, sal_Int16 wordType);
74 : Boundary getWordBoundary( const OUString& rText, sal_Int32 nPos, sal_Int16 wordType, bool bDirection );
75 : void setJapaneseWordBreak();
76 :
77 : private:
78 : WordBreakCache cache[CACHE_MAX];
79 : OUString segmentCachedString;
80 : Boundary segmentCachedBoundary;
81 :
82 : bool seekSegment(const OUString& rText, sal_Int32 pos, Boundary& boundary);
83 : WordBreakCache& getCache(const sal_Unicode *text, Boundary& boundary);
84 : bool exists(const sal_uInt32 u);
85 : sal_Int32 getLongestMatch(const sal_Unicode *text, sal_Int32 len);
86 : };
87 :
88 : } } } }
89 :
90 : #endif
91 :
92 : /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|