android: Update app-specific/MIME type icons
[LibreOffice.git] / sw / source / core / bastyp / breakit.cxx
blob33f55a1d548deaa3a3b9927f0fbb0de5ebcaeb1e
1 /* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
2 /*
3 * This file is part of the LibreOffice project.
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/.
9 * This file incorporates work covered by the following license notice:
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 .
20 #include <breakit.hxx>
21 #include <swtypes.hxx>
23 #include <com/sun/star/i18n/ScriptType.hpp>
24 #include <com/sun/star/i18n/CharacterIteratorMode.hpp>
25 #include <com/sun/star/i18n/BreakIterator.hpp>
26 #include <svl/languageoptions.hxx>
27 #include <unicode/uchar.h>
28 #include <unotools/localedatawrapper.hxx>
29 #include <algorithm>
30 #include <utility>
32 using namespace com::sun::star;
34 SwBreakIt* g_pBreakIt = nullptr;
36 void SwBreakIt::Create_( const uno::Reference<uno::XComponentContext> & rxContext )
38 delete g_pBreakIt;
39 g_pBreakIt = new SwBreakIt( rxContext );
42 void SwBreakIt::Delete_()
44 delete g_pBreakIt;
45 g_pBreakIt = nullptr;
48 SwBreakIt * SwBreakIt::Get()
50 return g_pBreakIt;
53 SwBreakIt::SwBreakIt( uno::Reference<uno::XComponentContext> xContext )
54 : m_xContext(std::move(xContext))
55 , m_xBreak(i18n::BreakIterator::create(m_xContext))
56 , m_aForbiddenLang(LANGUAGE_DONTKNOW)
60 void SwBreakIt::GetLocale_( const LanguageType aLang )
62 if (m_xLanguageTag)
63 m_xLanguageTag->reset(aLang);
64 else
65 m_xLanguageTag.reset(new LanguageTag(aLang));
68 void SwBreakIt::GetLocale_( const LanguageTag& rLanguageTag )
70 if (m_xLanguageTag)
71 *m_xLanguageTag = rLanguageTag;
72 else
73 m_xLanguageTag.reset(new LanguageTag(rLanguageTag));
76 void SwBreakIt::GetForbidden_( const LanguageType aLang )
78 LocaleDataWrapper aWrap(m_xContext, GetLanguageTag(aLang));
80 m_aForbiddenLang = aLang;
81 m_oForbidden.emplace(aWrap.getForbiddenCharacters());
84 sal_uInt16 SwBreakIt::GetRealScriptOfText( const OUString& rText, sal_Int32 nPos ) const
86 sal_uInt16 nScript = i18n::ScriptType::WEAK;
87 if (!rText.isEmpty())
89 if( nPos && nPos == rText.getLength() )
90 --nPos;
91 else if( nPos < 0)
92 nPos = 0;
94 nScript = m_xBreak->getScriptType(rText, nPos);
95 sal_Int32 nChgPos = 0;
96 if (i18n::ScriptType::WEAK == nScript && nPos >= 0 && nPos + 1 < rText.getLength())
98 // A weak character followed by a mark may be meant to combine with
99 // the mark, so prefer the following character's script
100 switch (u_charType(rText[nPos + 1]))
102 case U_NON_SPACING_MARK:
103 case U_ENCLOSING_MARK:
104 case U_COMBINING_SPACING_MARK:
105 nScript = m_xBreak->getScriptType(rText, nPos+1);
106 break;
109 if( i18n::ScriptType::WEAK == nScript && nPos )
111 nChgPos = m_xBreak->beginOfScript(rText, nPos, nScript);
112 if( 0 < nChgPos )
113 nScript = m_xBreak->getScriptType(rText, nChgPos-1);
116 if( i18n::ScriptType::WEAK == nScript )
118 nChgPos = m_xBreak->endOfScript(rText, nPos, nScript);
119 if( rText.getLength() > nChgPos && 0 <= nChgPos )
120 nScript = m_xBreak->getScriptType(rText, nChgPos);
123 if( i18n::ScriptType::WEAK == nScript )
124 nScript = SvtLanguageOptions::GetI18NScriptTypeOfLanguage( GetAppLanguage() );
125 return nScript;
128 SvtScriptType SwBreakIt::GetAllScriptsOfText( const OUString& rText ) const
130 const SvtScriptType coAllScripts = SvtScriptType::LATIN |
131 SvtScriptType::ASIAN |
132 SvtScriptType::COMPLEX;
133 SvtScriptType nRet = SvtScriptType::NONE;
134 sal_uInt16 nScript = 0;
135 if (!rText.isEmpty())
137 for( sal_Int32 n = 0, nEnd = rText.getLength(); n < nEnd;
138 n = m_xBreak->endOfScript(rText, n, nScript) )
140 nScript = m_xBreak->getScriptType(rText, n);
141 switch( nScript )
143 case i18n::ScriptType::LATIN: nRet |= SvtScriptType::LATIN; break;
144 case i18n::ScriptType::ASIAN: nRet |= SvtScriptType::ASIAN; break;
145 case i18n::ScriptType::COMPLEX: nRet |= SvtScriptType::COMPLEX; break;
146 case i18n::ScriptType::WEAK:
147 if( nRet == SvtScriptType::NONE )
148 nRet |= coAllScripts;
149 break;
151 if( coAllScripts == nRet )
152 break;
155 return nRet;
158 sal_Int32 SwBreakIt::getGraphemeCount(const OUString& rText,
159 sal_Int32 nStart, sal_Int32 nEnd) const
161 sal_Int32 nGraphemeCount = 0;
163 sal_Int32 nCurPos = std::max(static_cast<sal_Int32>(0), nStart);
164 while (nCurPos < nEnd)
166 // fdo#49208 cheat and assume that nothing can combine with a space
167 // to form a single grapheme
168 if (rText[nCurPos] == ' ')
170 ++nCurPos;
172 else
174 sal_Int32 nCount2 = 1;
175 nCurPos = m_xBreak->nextCharacters(rText, nCurPos, lang::Locale(),
176 i18n::CharacterIteratorMode::SKIPCELL, nCount2, nCount2);
178 ++nGraphemeCount;
181 return nGraphemeCount;
184 /* vim:set shiftwidth=4 softtabstop=4 expandtab: */