Change-Id: Ic6829542628ed1f5ed7d6b3d6be1d28ed1b5c75b Reviewed-on: https://gerrit.libreoffice.org/37292 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
325 lines
11 KiB
C++
325 lines
11 KiB
C++
/* -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*- */
|
|
/*
|
|
* This file is part of the LibreOffice project.
|
|
*
|
|
* This Source Code Form is subject to the terms of the Mozilla Public
|
|
* License, v. 2.0. If a copy of the MPL was not distributed with this
|
|
* file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
|
*
|
|
* This file incorporates work covered by the following license notice:
|
|
*
|
|
* Licensed to the Apache Software Foundation (ASF) under one or more
|
|
* contributor license agreements. See the NOTICE file distributed
|
|
* with this work for additional information regarding copyright
|
|
* ownership. The ASF licenses this file to you under the Apache
|
|
* License, Version 2.0 (the "License"); you may not use this file
|
|
* except in compliance with the License. You may obtain a copy of
|
|
* the License at http://www.apache.org/licenses/LICENSE-2.0 .
|
|
*/
|
|
|
|
#include <memory>
|
|
#include <string.h>
|
|
|
|
#include <com/sun/star/i18n/UnicodeType.hpp>
|
|
#include <com/sun/star/i18n/WordType.hpp>
|
|
|
|
#include <unotools/charclass.hxx>
|
|
|
|
#include <hintids.hxx>
|
|
#include <doc.hxx>
|
|
#include <IDocumentUndoRedo.hxx>
|
|
#include <IDocumentContentOperations.hxx>
|
|
#include <docary.hxx>
|
|
#include <mvsave.hxx>
|
|
#include <ndtxt.hxx>
|
|
#include <txatbase.hxx>
|
|
#include <rubylist.hxx>
|
|
#include <pam.hxx>
|
|
#include <swundo.hxx>
|
|
#include <breakit.hxx>
|
|
#include <swcrsr.hxx>
|
|
|
|
using namespace ::com::sun::star::i18n;
|
|
|
|
/*
|
|
* Members in the list:
|
|
* - String - the orig text
|
|
* - SwFormatRuby - the ruby attribute
|
|
*/
|
|
sal_uInt16 SwDoc::FillRubyList( const SwPaM& rPam, SwRubyList& rList )
|
|
{
|
|
const SwPaM *_pStartCursor = rPam.GetNext(),
|
|
*_pStartCursor2 = _pStartCursor;
|
|
bool bCheckEmpty = &rPam != _pStartCursor;
|
|
do {
|
|
const SwPosition* pStt = _pStartCursor->Start(),
|
|
* pEnd = pStt == _pStartCursor->GetPoint()
|
|
? _pStartCursor->GetMark()
|
|
: _pStartCursor->GetPoint();
|
|
if( !bCheckEmpty || ( pStt != pEnd && *pStt != *pEnd ))
|
|
{
|
|
SwPaM aPam( *pStt );
|
|
do {
|
|
std::unique_ptr<SwRubyListEntry> pNew(new SwRubyListEntry);
|
|
if( pEnd != pStt )
|
|
{
|
|
aPam.SetMark();
|
|
*aPam.GetMark() = *pEnd;
|
|
}
|
|
if( SelectNextRubyChars( aPam, *pNew ))
|
|
{
|
|
rList.push_back(std::move(pNew));
|
|
aPam.DeleteMark();
|
|
}
|
|
else
|
|
{
|
|
if( *aPam.GetPoint() < *pEnd )
|
|
{
|
|
// goto next paragraph
|
|
aPam.DeleteMark();
|
|
aPam.Move( fnMoveForward, GoInNode );
|
|
}
|
|
else
|
|
break;
|
|
}
|
|
} while( 30 > rList.size() && *aPam.GetPoint() < *pEnd );
|
|
}
|
|
} while( 30 > rList.size() &&
|
|
(_pStartCursor = _pStartCursor->GetNext()) != _pStartCursor2 );
|
|
|
|
return rList.size();
|
|
}
|
|
|
|
void SwDoc::SetRubyList( const SwPaM& rPam, const SwRubyList& rList )
|
|
{
|
|
GetIDocumentUndoRedo().StartUndo( SwUndoId::SETRUBYATTR, nullptr );
|
|
std::set<sal_uInt16> aDelArr;
|
|
aDelArr.insert( RES_TXTATR_CJK_RUBY );
|
|
|
|
SwRubyList::size_type nListEntry = 0;
|
|
|
|
const SwPaM *_pStartCursor = rPam.GetNext(),
|
|
*_pStartCursor2 = _pStartCursor;
|
|
bool bCheckEmpty = &rPam != _pStartCursor;
|
|
do {
|
|
const SwPosition* pStt = _pStartCursor->Start(),
|
|
* pEnd = pStt == _pStartCursor->GetPoint()
|
|
? _pStartCursor->GetMark()
|
|
: _pStartCursor->GetPoint();
|
|
if( !bCheckEmpty || ( pStt != pEnd && *pStt != *pEnd ))
|
|
{
|
|
|
|
SwPaM aPam( *pStt );
|
|
do {
|
|
SwRubyListEntry aCheckEntry;
|
|
if( pEnd != pStt )
|
|
{
|
|
aPam.SetMark();
|
|
*aPam.GetMark() = *pEnd;
|
|
}
|
|
if( SelectNextRubyChars( aPam, aCheckEntry ))
|
|
{
|
|
const SwRubyListEntry* pEntry = rList[ nListEntry++ ].get();
|
|
if( aCheckEntry.GetRubyAttr() != pEntry->GetRubyAttr() )
|
|
{
|
|
// set/reset the attribute
|
|
if( !pEntry->GetRubyAttr().GetText().isEmpty() )
|
|
{
|
|
getIDocumentContentOperations().InsertPoolItem( aPam, pEntry->GetRubyAttr() );
|
|
}
|
|
else
|
|
{
|
|
ResetAttrs( aPam, true, aDelArr );
|
|
}
|
|
}
|
|
|
|
if( !pEntry->GetText().isEmpty() &&
|
|
aCheckEntry.GetText() != pEntry->GetText() )
|
|
{
|
|
// text is changed, so replace the original
|
|
getIDocumentContentOperations().ReplaceRange( aPam, pEntry->GetText(), false );
|
|
}
|
|
aPam.DeleteMark();
|
|
}
|
|
else
|
|
{
|
|
if( *aPam.GetPoint() < *pEnd )
|
|
{
|
|
// goto next paragraph
|
|
aPam.DeleteMark();
|
|
aPam.Move( fnMoveForward, GoInNode );
|
|
}
|
|
else
|
|
{
|
|
const SwRubyListEntry* pEntry = rList[ nListEntry++ ].get();
|
|
|
|
// set/reset the attribute
|
|
if( !pEntry->GetRubyAttr().GetText().isEmpty() &&
|
|
!pEntry->GetText().isEmpty() )
|
|
{
|
|
getIDocumentContentOperations().InsertString( aPam, pEntry->GetText() );
|
|
aPam.SetMark();
|
|
aPam.GetMark()->nContent -= pEntry->GetText().getLength();
|
|
getIDocumentContentOperations().InsertPoolItem(
|
|
aPam, pEntry->GetRubyAttr(), SetAttrMode::DONTEXPAND );
|
|
}
|
|
else
|
|
break;
|
|
aPam.DeleteMark();
|
|
}
|
|
}
|
|
} while( nListEntry < rList.size() && *aPam.GetPoint() < *pEnd );
|
|
}
|
|
} while( 30 > rList.size() &&
|
|
(_pStartCursor = _pStartCursor->GetNext()) != _pStartCursor2 );
|
|
|
|
GetIDocumentUndoRedo().EndUndo( SwUndoId::SETRUBYATTR, nullptr );
|
|
}
|
|
|
|
bool SwDoc::SelectNextRubyChars( SwPaM& rPam, SwRubyListEntry& rEntry )
|
|
{
|
|
// Point must be the startposition, Mark is optional the end position
|
|
SwPosition* pPos = rPam.GetPoint();
|
|
const SwTextNode* pTNd = pPos->nNode.GetNode().GetTextNode();
|
|
OUString const& rText = pTNd->GetText();
|
|
sal_Int32 nStart = pPos->nContent.GetIndex();
|
|
sal_Int32 nEnd = rText.getLength();
|
|
|
|
bool bHasMark = rPam.HasMark();
|
|
if( bHasMark )
|
|
{
|
|
// in the same node?
|
|
if( rPam.GetMark()->nNode == pPos->nNode )
|
|
{
|
|
// then use that end
|
|
const sal_Int32 nTEnd = rPam.GetMark()->nContent.GetIndex();
|
|
if( nTEnd < nEnd )
|
|
nEnd = nTEnd;
|
|
}
|
|
rPam.DeleteMark();
|
|
}
|
|
|
|
// search the start
|
|
// look where a ruby attribute starts
|
|
const SwpHints* pHts = pTNd->GetpSwpHints();
|
|
const SwTextAttr* pAttr = nullptr;
|
|
if( pHts )
|
|
{
|
|
for( size_t nHtIdx = 0; nHtIdx < pHts->Count(); ++nHtIdx )
|
|
{
|
|
const SwTextAttr* pHt = pHts->Get(nHtIdx);
|
|
if( RES_TXTATR_CJK_RUBY == pHt->Which() &&
|
|
*pHt->GetAnyEnd() > nStart )
|
|
{
|
|
if( pHt->GetStart() < nEnd )
|
|
{
|
|
pAttr = pHt;
|
|
if( !bHasMark && nStart > pAttr->GetStart() )
|
|
{
|
|
nStart = pAttr->GetStart();
|
|
pPos->nContent = nStart;
|
|
}
|
|
}
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
if( !bHasMark && nStart && ( !pAttr || nStart != pAttr->GetStart()) )
|
|
{
|
|
// skip to the word begin!
|
|
const sal_Int32 nWordStt = g_pBreakIt->GetBreakIter()->getWordBoundary(
|
|
rText, nStart,
|
|
g_pBreakIt->GetLocale( pTNd->GetLang( nStart )),
|
|
WordType::ANYWORD_IGNOREWHITESPACES,
|
|
true ).startPos;
|
|
if (nWordStt < nStart && nWordStt >= 0)
|
|
{
|
|
nStart = nWordStt;
|
|
pPos->nContent = nStart;
|
|
}
|
|
}
|
|
|
|
bool bAlphaNum = false;
|
|
sal_Int32 nWordEnd = nEnd;
|
|
CharClass& rCC = GetAppCharClass();
|
|
while( nStart < nEnd )
|
|
{
|
|
if( pAttr && nStart == pAttr->GetStart() )
|
|
{
|
|
pPos->nContent = nStart;
|
|
if( !rPam.HasMark() )
|
|
{
|
|
rPam.SetMark();
|
|
pPos->nContent = *pAttr->GetAnyEnd();
|
|
if( pPos->nContent.GetIndex() > nEnd )
|
|
pPos->nContent = nEnd;
|
|
rEntry.SetRubyAttr( pAttr->GetRuby() );
|
|
}
|
|
break;
|
|
}
|
|
|
|
sal_Int32 nChType = rCC.getType(rText, nStart);
|
|
bool bIgnoreChar = false, bIsAlphaNum = false, bChkNxtWrd = false;
|
|
switch( nChType )
|
|
{
|
|
case UnicodeType::UPPERCASE_LETTER:
|
|
case UnicodeType::LOWERCASE_LETTER:
|
|
case UnicodeType::TITLECASE_LETTER:
|
|
case UnicodeType::DECIMAL_DIGIT_NUMBER:
|
|
bChkNxtWrd = bIsAlphaNum = true;
|
|
break;
|
|
|
|
case UnicodeType::SPACE_SEPARATOR:
|
|
case UnicodeType::CONTROL:
|
|
/*??*/ case UnicodeType::PRIVATE_USE:
|
|
case UnicodeType::START_PUNCTUATION:
|
|
case UnicodeType::END_PUNCTUATION:
|
|
bIgnoreChar = true;
|
|
break;
|
|
|
|
case UnicodeType::OTHER_LETTER:
|
|
bChkNxtWrd = true;
|
|
SAL_FALLTHROUGH;
|
|
default:
|
|
bIsAlphaNum = false;
|
|
break;
|
|
}
|
|
|
|
if( rPam.HasMark() )
|
|
{
|
|
if( bIgnoreChar || bIsAlphaNum != bAlphaNum || nStart >= nWordEnd )
|
|
break;
|
|
}
|
|
else if( !bIgnoreChar )
|
|
{
|
|
rPam.SetMark();
|
|
bAlphaNum = bIsAlphaNum;
|
|
if (bChkNxtWrd)
|
|
{
|
|
// search the end of this word
|
|
nWordEnd = g_pBreakIt->GetBreakIter()->getWordBoundary(
|
|
rText, nStart,
|
|
g_pBreakIt->GetLocale( pTNd->GetLang( nStart )),
|
|
WordType::ANYWORD_IGNOREWHITESPACES,
|
|
true ).endPos;
|
|
if( 0 > nWordEnd || nWordEnd > nEnd || nWordEnd == nStart )
|
|
nWordEnd = nEnd;
|
|
}
|
|
}
|
|
pTNd->GoNext( &pPos->nContent, CRSR_SKIP_CHARS );
|
|
nStart = pPos->nContent.GetIndex();
|
|
}
|
|
|
|
nStart = rPam.GetMark()->nContent.GetIndex();
|
|
rEntry.SetText( rText.copy( nStart,
|
|
rPam.GetPoint()->nContent.GetIndex() - nStart ));
|
|
return rPam.HasMark();
|
|
}
|
|
|
|
SwRubyListEntry::~SwRubyListEntry()
|
|
{
|
|
}
|
|
|
|
/* vim:set shiftwidth=4 softtabstop=4 expandtab: */
|