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_EDITENG_HANGULHANJA_HXX
20 : #define INCLUDED_EDITENG_HANGULHANJA_HXX
21 :
22 : #include <vcl/window.hxx>
23 : #include <memory>
24 : #include <com/sun/star/lang/XMultiServiceFactory.hpp>
25 : #include <com/sun/star/lang/Locale.hpp>
26 : #include <com/sun/star/uno/Sequence.hxx>
27 : #include <com/sun/star/uno/XComponentContext.hpp>
28 : #include <editeng/editengdllapi.h>
29 : #include <i18nlangtag/mslangid.hxx>
30 :
31 :
32 : namespace editeng
33 : {
34 :
35 :
36 : class HangulHanjaConversion_Impl;
37 :
38 :
39 : //= HangulHanjaConversion
40 :
41 : /** encapsulates Hangul-Hanja conversion functionality
42 :
43 : <p>terminology:
44 : <ul><li>A <b>text <em>portion</em></b> is some (potentially large) piece of text
45 : which is to be analyzed for convertible sub-strings.</li>
46 : <li>A <b>text <em>unit</em></b> is a sub string in a text portion, which is
47 : to be converted as a whole.</li>
48 : </ul>
49 : For instance, you could have two independent selections within your document, which are then
50 : two text portions. A text unit would be single Hangul/Hanja words within a portion, or even
51 : single Hangul syllabills when "replace by character" is enabled.
52 : </p>
53 : */
54 : class EDITENG_DLLPUBLIC HangulHanjaConversion
55 : {
56 : friend class HangulHanjaConversion_Impl;
57 :
58 : public:
59 : enum ReplacementAction
60 : {
61 : eExchange, // simply exchange one text with another
62 : eReplacementBracketed, // keep the original, and put the replacement in brackets after it
63 : eOriginalBracketed, // replace the original text, but put it in brackeds after the replacement
64 : eReplacementAbove, // keep the original, and put the replacement text as ruby text above it
65 : eOriginalAbove, // replace the original text, but put it as ruby text above it
66 : eReplacementBelow, // keep the original, and put the replacement text as ruby text below it
67 : eOriginalBelow // replace the original text, but put it as ruby text below it
68 : };
69 :
70 : enum ConversionType // does not specify direction...
71 : {
72 : eConvHangulHanja, // Korean Hangul/Hanja conversion
73 : eConvSimplifiedTraditional // Chinese simplified / Chinese traditional conversion
74 : };
75 :
76 : // Note: conversion direction for eConvSimplifiedTraditional is
77 : // specified by source language.
78 : // This one is for Hangul/Hanja where source and target language
79 : // are the same.
80 : enum ConversionDirection
81 : {
82 : eHangulToHanja,
83 : eHanjaToHangul
84 : };
85 :
86 : enum ConversionFormat
87 : {
88 : eSimpleConversion, // used for simplified / traditional Chinese as well
89 : eHangulBracketed,
90 : eHanjaBracketed,
91 : eRubyHanjaAbove,
92 : eRubyHanjaBelow,
93 : eRubyHangulAbove,
94 : eRubyHangulBelow
95 : };
96 :
97 : private:
98 : ::std::unique_ptr< HangulHanjaConversion_Impl > m_pImpl;
99 :
100 : // used to set initial values of m_pImpl object from saved ones
101 : static bool m_bUseSavedValues; // defines if the followng two values should be used for initialization
102 : static bool m_bTryBothDirectionsSave;
103 : static ConversionDirection m_ePrimaryConversionDirectionSave;
104 :
105 : HangulHanjaConversion (const HangulHanjaConversion &) SAL_DELETED_FUNCTION;
106 : HangulHanjaConversion & operator= (const HangulHanjaConversion &) SAL_DELETED_FUNCTION;
107 :
108 : public:
109 : HangulHanjaConversion(
110 : vcl::Window* _pUIParent,
111 : const ::com::sun::star::uno::Reference< ::com::sun::star::uno::XComponentContext >& rxContext,
112 : const ::com::sun::star::lang::Locale& _rSourceLocale,
113 : const ::com::sun::star::lang::Locale& _rTargetLocale,
114 : const vcl::Font* _pTargetFont,
115 : sal_Int32 nOptions,
116 : bool _bIsInteractive
117 : );
118 :
119 : virtual ~HangulHanjaConversion( );
120 :
121 : // converts the whole document
122 : void ConvertDocument();
123 :
124 : LanguageType GetSourceLanguage() const;
125 : LanguageType GetTargetLanguage() const;
126 : const vcl::Font* GetTargetFont() const;
127 : sal_Int32 GetConversionOptions() const;
128 : bool IsInteractive() const;
129 :
130 : // chinese text conversion
131 : static inline bool IsSimplified( LanguageType nLang );
132 : static inline bool IsTraditional( LanguageType nLang );
133 : static inline bool IsChinese( LanguageType nLang );
134 :
135 : // used to specify that the conversion direction states from the
136 : // last incarnation should be used as
137 : // initial conversion direction for the next incarnation.
138 : // (A hack used to transport a state information from
139 : // one incarnation to the next. Used in Writers text conversion...)
140 : static void SetUseSavedConversionDirectionState( bool bVal );
141 : static bool IsUseSavedConversionDirectionState();
142 :
143 : protected:
144 : /** retrieves the next text portion which is to be analyzed
145 :
146 : <p>pseudo-abstract, needs to be overridden</p>
147 :
148 : @param _rNextPortion
149 : upon return, this must contain the next text portion
150 : @param _rLangOfPortion
151 : upon return, this must contain the language for the found text portion.
152 : (necessary for Chinese translation since there are 5 language variants
153 : too look for even if the 'source' language usually is only 'simplified'
154 : or 'traditional'.)
155 : */
156 : virtual void GetNextPortion(
157 : OUString& /* [out] */ _rNextPortion,
158 : LanguageType& /* [out] */ _rLangOfPortion,
159 : bool /* [in] */ _bAllowImplicitChangesForNotConvertibleText ) = 0;
160 :
161 : /** announces a new "current unit"
162 :
163 : <p>This will be called whenever it is necessary to interactively ask the user for
164 : a conversion. In such a case, a range within the current portion (see <member>GetNextPortion</member>)
165 : is presented to the user for chosing a substitution. Additionally, this method is called,
166 : so that derived classes can e.g. highlight this text range in a document view.</p>
167 :
168 : <p>Note that the indexes are relative to the most recent replace action. See
169 : <member>ReplaceUnit</member> for details.</p>
170 :
171 : @param _nUnitStart
172 : the start index of the unit
173 :
174 : @param _nUnitEnd
175 : the start index (exclusively!) of the unit.
176 :
177 : @param _bAllowImplicitChangesForNotConvertibleText
178 : allows implicit changes other than the text itself for the
179 : text parts not being convertible.
180 : Used for chinese translation to attribute all not convertible
181 : text (e.g. western text, empty paragraphs, spaces, ...) to
182 : the target language and target font of the conversion.
183 : This is to ensure that after the conversion any new text entered
184 : anywhere in the document will have the target language (of course
185 : CJK Language only) and target font (CJK font only) set.
186 :
187 : @see GetNextPortion
188 : */
189 : virtual void HandleNewUnit( const sal_Int32 _nUnitStart, const sal_Int32 _nUnitEnd ) = 0;
190 :
191 : /** replaces a text unit within a text portion with a new text
192 :
193 : <p>pseudo-abstract, needs to be overridden</p>
194 :
195 : <p>Note an important thing about the indices: They are always relative to the <em>previous
196 : call</em> of ReplaceUnit. This means whe you get a call to ReplaceUnit, and replace some text
197 : in your document, than you have to remember the document position immediately <em>behind</em>
198 : the changed text. In a next call to ReplaceUnit, an index of <em>0</em> will denote exactly
199 : this position behind the previous replacement<br/>
200 : The reaons for this is that this class here does not know anything about your document structure,
201 : so after a replacement took place, it's impossible to address anything in the range from the
202 : beginning of the portion up to the replaced text.<br/>
203 : In the very first call to ReplaceUnit, an index of <em>0</em> denotes the very first position of
204 : the current portion.</p>
205 :
206 : <p>If the language of the text to be replaced is different from
207 : the target language (as given by 'GetTargetLanguage') for example
208 : when converting simplified Chinese from/to traditional Chinese
209 : the language attribute of the new text has to be changed as well,
210 : **and** the font is to be set to the default (document) font for
211 : that language.</p>
212 :
213 : @param _nUnitStart
214 : the start index of the range to replace
215 :
216 : @param _nUnitEnd
217 : the end index (exclusively!) of the range to replace. E.g., an index
218 : pair (4,5) indicates a range of length 1.
219 :
220 : @param _rOrigText
221 : the original text to be replaced (as returned by GetNextPortion).
222 : Since in Chinese conversion the original text is needed as well
223 : in order to only do the minimal necassry text changes and to keep
224 : as much attributes as possible this is supplied here as well.
225 :
226 : @param _rReplaceWith
227 : The replacement text
228 :
229 : @param _rOffsets
230 : An sequence matching the indices (characters) of _rReplaceWith
231 : to the indices of the characters in the original text they are
232 : replacing.
233 : This is necessary since some portions of the text may get
234 : converted in portions of different length than the original.
235 : The sequence will be empty if all conversions in the text are
236 : of equal length. That is if always the character at index i in
237 : _rOffsets is replacing the character at index i in the original
238 : text for all valid index values of i.
239 :
240 : @param _eAction
241 : replacement action to take
242 :
243 : @param pNewUnitLanguage
244 : if the replacement unit is required to have a new language that
245 : is specified here. If the language is to be left unchanged this
246 : is the 0 pointer.
247 : */
248 : virtual void ReplaceUnit(
249 : const sal_Int32 _nUnitStart, const sal_Int32 _nUnitEnd,
250 : const OUString& _rOrigText,
251 : const OUString& _rReplaceWith,
252 : const ::com::sun::star::uno::Sequence< sal_Int32 > &_rOffsets,
253 : ReplacementAction _eAction,
254 : LanguageType *pNewUnitLanguage
255 : ) = 0;
256 :
257 : /** specifies if rubies are supported by the document implementing
258 : this class.
259 :
260 : @return
261 : <TRUE/> if rubies are supported.
262 : */
263 : virtual bool HasRubySupport() const = 0;
264 : };
265 :
266 1 : bool HangulHanjaConversion::IsSimplified( LanguageType nLang )
267 : {
268 1 : return MsLangId::isSimplifiedChinese(nLang);
269 : }
270 :
271 1 : bool HangulHanjaConversion::IsTraditional( LanguageType nLang )
272 : {
273 1 : return MsLangId::isTraditionalChinese(nLang);
274 : }
275 :
276 19 : bool HangulHanjaConversion::IsChinese( LanguageType nLang )
277 : {
278 19 : return MsLangId::isChinese(nLang);
279 : }
280 :
281 : }
282 :
283 :
284 : #endif // INCLUDED_EDITENG_HANGULHANJA_HXX
285 :
286 : /* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|
