diff options
Diffstat (limited to 'writerfilter/source/rtftok/rtfdispatchsymbol.cxx')
| -rw-r--r-- | writerfilter/source/rtftok/rtfdispatchsymbol.cxx | 442 |
1 files changed, 442 insertions, 0 deletions
diff --git a/writerfilter/source/rtftok/rtfdispatchsymbol.cxx b/writerfilter/source/rtftok/rtfdispatchsymbol.cxx new file mode 100644 index 000000000..3f9ed20bf --- /dev/null +++ b/writerfilter/source/rtftok/rtfdispatchsymbol.cxx @@ -0,0 +1,442 @@ +/* -*- 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/. + */ + +#include "rtfdocumentimpl.hxx" + +#include <com/sun/star/io/WrongFormatException.hpp> +#include <svl/lngmisc.hxx> + +#include <ooxml/resourceids.hxx> + +#include <sal/log.hxx> + +#include "rtfreferenceproperties.hxx" +#include "rtfskipdestination.hxx" + +using namespace com::sun::star; + +namespace writerfilter::rtftok +{ +RTFError RTFDocumentImpl::dispatchSymbol(RTFKeyword nKeyword) +{ + setNeedSect(true); + if (nKeyword != RTFKeyword::HEXCHAR) + checkUnicode(/*bUnicode =*/true, /*bHex =*/true); + else + checkUnicode(/*bUnicode =*/true, /*bHex =*/false); + RTFSkipDestination aSkip(*this); + + if (RTFKeyword::LINE == nKeyword) + { + // very special handling since text() will eat lone '\n' + singleChar('\n', /*bRunProps=*/true); + return RTFError::OK; + } + // Trivial symbols + sal_uInt8 cCh = 0; + switch (nKeyword) + { + case RTFKeyword::TAB: + cCh = '\t'; + break; + case RTFKeyword::BACKSLASH: + cCh = '\\'; + break; + case RTFKeyword::LBRACE: + cCh = '{'; + break; + case RTFKeyword::RBRACE: + cCh = '}'; + break; + case RTFKeyword::EMDASH: + cCh = 151; + break; + case RTFKeyword::ENDASH: + cCh = 150; + break; + case RTFKeyword::BULLET: + cCh = 149; + break; + case RTFKeyword::LQUOTE: + cCh = 145; + break; + case RTFKeyword::RQUOTE: + cCh = 146; + break; + case RTFKeyword::LDBLQUOTE: + cCh = 147; + break; + case RTFKeyword::RDBLQUOTE: + cCh = 148; + break; + default: + break; + } + if (cCh > 0) + { + OUString aStr(OStringToOUString(OStringChar(char(cCh)), RTL_TEXTENCODING_MS_1252)); + text(aStr); + return RTFError::OK; + } + + switch (nKeyword) + { + case RTFKeyword::IGNORE: + { + m_bSkipUnknown = true; + aSkip.setReset(false); + return RTFError::OK; + } + break; + case RTFKeyword::PAR: + { + if (m_aStates.top().getDestination() == Destination::FOOTNOTESEPARATOR) + break; // just ignore it - only thing we read in here is CHFTNSEP + checkFirstRun(); + bool bNeedPap = m_bNeedPap; + checkNeedPap(); + if (bNeedPap) + runProps(); + if (!m_aStates.top().getCurrentBuffer()) + { + parBreak(); + // Not in table? Reset max width. + if (m_nCellxMax) + { + // Was in table, but not anymore -> tblEnd. + RTFSprms aAttributes; + RTFSprms aSprms; + aSprms.set(NS_ooxml::LN_tblEnd, new RTFValue(1)); + writerfilter::Reference<Properties>::Pointer_t pProperties + = new RTFReferenceProperties(std::move(aAttributes), std::move(aSprms)); + Mapper().props(pProperties); + } + m_nCellxMax = 0; + } + else if (m_aStates.top().getDestination() != Destination::SHAPETEXT) + { + RTFValue::Pointer_t pValue; + m_aStates.top().getCurrentBuffer()->push_back(Buf_t(BUFFER_PAR, pValue, nullptr)); + } + // but don't emit properties yet, since they may change till the first text token arrives + m_bNeedPap = true; + if (!m_aStates.top().getFrame().inFrame()) + m_bNeedPar = false; + m_bNeedFinalPar = false; + } + break; + case RTFKeyword::SECT: + { + if (m_bNeedCr) + dispatchSymbol(RTFKeyword::PAR); + + m_bHadSect = true; + if (m_bIgnoreNextContSectBreak) + m_bIgnoreNextContSectBreak = false; + else + { + sectBreak(); + if (m_nResetBreakOnSectBreak != RTFKeyword::invalid) + { + // this should run on _second_ \sect after \page + dispatchSymbol(m_nResetBreakOnSectBreak); // lazy reset + m_nResetBreakOnSectBreak = RTFKeyword::invalid; + m_bNeedSect = false; // dispatchSymbol set it + } + } + } + break; + case RTFKeyword::NOBREAK: + { + OUString aStr(SVT_HARD_SPACE); + text(aStr); + } + break; + case RTFKeyword::NOBRKHYPH: + { + OUString aStr(SVT_HARD_HYPHEN); + text(aStr); + } + break; + case RTFKeyword::OPTHYPH: + { + OUString aStr(SVT_SOFT_HYPHEN); + text(aStr); + } + break; + case RTFKeyword::HEXCHAR: + m_aStates.top().setInternalState(RTFInternalState::HEX); + break; + case RTFKeyword::CELL: + case RTFKeyword::NESTCELL: + { + if (nKeyword == RTFKeyword::CELL) + m_bAfterCellBeforeRow = true; + + checkFirstRun(); + if (m_bNeedPap) + { + // There were no runs in the cell, so we need to send paragraph and character properties here. + auto pPValue = new RTFValue(m_aStates.top().getParagraphAttributes(), + m_aStates.top().getParagraphSprms()); + bufferProperties(m_aTableBufferStack.back(), pPValue, nullptr); + auto pCValue = new RTFValue(m_aStates.top().getCharacterAttributes(), + m_aStates.top().getCharacterSprms()); + bufferProperties(m_aTableBufferStack.back(), pCValue, nullptr); + } + + RTFValue::Pointer_t pValue; + m_aTableBufferStack.back().emplace_back(Buf_t(BUFFER_CELLEND, pValue, nullptr)); + m_bNeedPap = true; + } + break; + case RTFKeyword::NESTROW: + { + tools::SvRef<TableRowBuffer> const pBuffer( + new TableRowBuffer(m_aTableBufferStack.back(), m_aNestedTableCellsSprms, + m_aNestedTableCellsAttributes, m_nNestedCells)); + prepareProperties(m_aStates.top(), pBuffer->GetParaProperties(), + pBuffer->GetFrameProperties(), pBuffer->GetRowProperties(), + m_nNestedCells, m_nNestedCurrentCellX - m_nNestedTRLeft); + + if (m_aTableBufferStack.size() == 1 || !m_aStates.top().getCurrentBuffer()) + { + throw io::WrongFormatException("mismatch between \\itap and number of \\nestrow", + nullptr); + } + assert(m_aStates.top().getCurrentBuffer() == &m_aTableBufferStack.back()); + // note: there may be several states pointing to table buffer! + for (std::size_t i = 0; i < m_aStates.size(); ++i) + { + if (m_aStates[i].getCurrentBuffer() == &m_aTableBufferStack.back()) + { + m_aStates[i].setCurrentBuffer( + &m_aTableBufferStack[m_aTableBufferStack.size() - 2]); + } + } + m_aTableBufferStack.pop_back(); + m_aTableBufferStack.back().emplace_back( + Buf_t(BUFFER_NESTROW, RTFValue::Pointer_t(), pBuffer)); + + m_aNestedTableCellsSprms.clear(); + m_aNestedTableCellsAttributes.clear(); + m_nNestedCells = 0; + m_bNeedPap = true; + } + break; + case RTFKeyword::ROW: + { + m_bAfterCellBeforeRow = false; + if (m_aStates.top().getTableRowWidthAfter() > 0) + { + // Add fake cellx / cell, RTF equivalent of + // OOXMLFastContextHandlerTextTableRow::handleGridAfter(). + auto pXValue = new RTFValue(m_aStates.top().getTableRowWidthAfter()); + m_aStates.top().getTableRowSprms().set(NS_ooxml::LN_CT_TblGridBase_gridCol, pXValue, + RTFOverwrite::NO_APPEND); + dispatchSymbol(RTFKeyword::CELL); + + // Adjust total width, which is done in the \cellx handler for normal cells. + m_nTopLevelCurrentCellX += m_aStates.top().getTableRowWidthAfter(); + + m_aStates.top().setTableRowWidthAfter(0); + } + + bool bRestored = false; + // Ending a row, but no cells defined? + // See if there was an invalid table row reset, so we can restore cell infos to help invalid documents. + if (!m_nTopLevelCurrentCellX && m_nBackupTopLevelCurrentCellX) + { + restoreTableRowProperties(); + bRestored = true; + } + + // If the right edge of the last cell (row width) is smaller than the width of some other row, mimic WW8TabDesc::CalcDefaults(): resize the last cell + const int MINLAY = 23; // sw/inc/swtypes.hxx, minimal possible size of frames. + if ((m_nCellxMax - m_nTopLevelCurrentCellX) >= MINLAY) + { + auto pXValueLast = m_aStates.top().getTableRowSprms().find( + NS_ooxml::LN_CT_TblGridBase_gridCol, false); + const int nXValueLast = pXValueLast ? pXValueLast->getInt() : 0; + auto pXValue = new RTFValue(nXValueLast + m_nCellxMax - m_nTopLevelCurrentCellX); + m_aStates.top().getTableRowSprms().eraseLast(NS_ooxml::LN_CT_TblGridBase_gridCol); + m_aStates.top().getTableRowSprms().set(NS_ooxml::LN_CT_TblGridBase_gridCol, pXValue, + RTFOverwrite::NO_APPEND); + m_nTopLevelCurrentCellX = m_nCellxMax; + } + + if (m_nTopLevelCells) + { + // Make a backup before we start popping elements + m_aTableInheritingCellsSprms = m_aTopLevelTableCellsSprms; + m_aTableInheritingCellsAttributes = m_aTopLevelTableCellsAttributes; + m_nInheritingCells = m_nTopLevelCells; + } + else + { + // No table definition? Then inherit from the previous row + m_aTopLevelTableCellsSprms = m_aTableInheritingCellsSprms; + m_aTopLevelTableCellsAttributes = m_aTableInheritingCellsAttributes; + m_nTopLevelCells = m_nInheritingCells; + } + + while (m_aTableBufferStack.size() > 1) + { + SAL_WARN("writerfilter.rtf", "dropping extra table buffer"); + // note: there may be several states pointing to table buffer! + for (std::size_t i = 0; i < m_aStates.size(); ++i) + { + if (m_aStates[i].getCurrentBuffer() == &m_aTableBufferStack.back()) + { + m_aStates[i].setCurrentBuffer(&m_aTableBufferStack.front()); + } + } + m_aTableBufferStack.pop_back(); + } + + replayRowBuffer(m_aTableBufferStack.back(), m_aTopLevelTableCellsSprms, + m_aTopLevelTableCellsAttributes, m_nTopLevelCells); + + // The scope of the table cell defaults is one row. + m_aDefaultState.getTableCellSprms().clear(); + m_aStates.top().getTableCellSprms() = m_aDefaultState.getTableCellSprms(); + m_aStates.top().getTableCellAttributes() = m_aDefaultState.getTableCellAttributes(); + + writerfilter::Reference<Properties>::Pointer_t paraProperties; + writerfilter::Reference<Properties>::Pointer_t frameProperties; + writerfilter::Reference<Properties>::Pointer_t rowProperties; + prepareProperties(m_aStates.top(), paraProperties, frameProperties, rowProperties, + m_nTopLevelCells, m_nTopLevelCurrentCellX - m_nTopLevelTRLeft); + sendProperties(paraProperties, frameProperties, rowProperties); + + m_bNeedPap = true; + m_bNeedFinalPar = true; + m_aTableBufferStack.back().clear(); + m_nTopLevelCells = 0; + + if (bRestored) + // We restored cell definitions, clear these now. + // This is necessary, as later cell definitions want to overwrite the restored ones. + resetTableRowProperties(); + } + break; + case RTFKeyword::COLUMN: + { + bool bColumns = false; // If we have multiple columns + RTFValue::Pointer_t pCols + = m_aStates.top().getSectionSprms().find(NS_ooxml::LN_EG_SectPrContents_cols); + if (pCols) + { + RTFValue::Pointer_t pNum = pCols->getAttributes().find(NS_ooxml::LN_CT_Columns_num); + if (pNum && pNum->getInt() > 1) + bColumns = true; + } + checkFirstRun(); + if (bColumns) + { + sal_uInt8 const sBreak[] = { 0xe }; + Mapper().startCharacterGroup(); + Mapper().text(sBreak, 1); + Mapper().endCharacterGroup(); + } + else + dispatchSymbol(RTFKeyword::PAGE); + } + break; + case RTFKeyword::CHFTN: + { + if (m_aStates.top().getCurrentBuffer() == &m_aSuperBuffer) + // Stop buffering, there will be no custom mark for this footnote or endnote. + m_aStates.top().setCurrentBuffer(nullptr); + break; + } + case RTFKeyword::PAGE: + { + // Ignore page breaks inside tables. + if (m_aStates.top().getCurrentBuffer() == &m_aTableBufferStack.back()) + break; + + // If we're inside a continuous section, we should send a section break, not a page one. + RTFValue::Pointer_t pBreak + = m_aStates.top().getSectionSprms().find(NS_ooxml::LN_EG_SectPrContents_type); + // Unless we're on a title page. + RTFValue::Pointer_t pTitlePg + = m_aStates.top().getSectionSprms().find(NS_ooxml::LN_EG_SectPrContents_titlePg); + if (((pBreak + && pBreak->getInt() + == static_cast<sal_Int32>(NS_ooxml::LN_Value_ST_SectionMark_continuous)) + || m_nResetBreakOnSectBreak == RTFKeyword::SBKNONE) + && !(pTitlePg && pTitlePg->getInt())) + { + if (m_bWasInFrame) + { + dispatchSymbol(RTFKeyword::PAR); + m_bWasInFrame = false; + } + sectBreak(); + // note: this will not affect the following section break + // but the one just pushed + dispatchFlag(RTFKeyword::SBKPAGE); + if (m_bNeedPar) + dispatchSymbol(RTFKeyword::PAR); + m_bIgnoreNextContSectBreak = true; + // arrange to clean up the synthetic RTFKeyword::SBKPAGE + m_nResetBreakOnSectBreak = RTFKeyword::SBKNONE; + } + else + { + bool bFirstRun = m_bFirstRun; + checkFirstRun(); + checkNeedPap(); + sal_uInt8 const sBreak[] = { 0xc }; + Mapper().text(sBreak, 1); + if (bFirstRun || m_bNeedCr) + { + // If we don't have content in the document yet (so the break-before can't move + // to a second layout page) or we already have characters sent (so the paragraph + // properties are already finalized), then continue inserting a fake paragraph. + if (!m_bNeedPap) + { + parBreak(); + m_bNeedPap = true; + } + } + m_bNeedCr = true; + } + } + break; + case RTFKeyword::CHPGN: + { + OUString aStr("PAGE"); + singleChar(cFieldStart); + text(aStr); + singleChar(cFieldSep, true); + singleChar(cFieldEnd); + } + break; + case RTFKeyword::CHFTNSEP: + { + static const sal_Unicode uFtnEdnSep = 0x3; + Mapper().utext(reinterpret_cast<const sal_uInt8*>(&uFtnEdnSep), 1); + } + break; + default: + { + SAL_INFO("writerfilter.rtf", + "TODO handle symbol '" << keywordToString(nKeyword) << "'"); + aSkip.setParsed(false); + } + break; + } + return RTFError::OK; +} + +} // namespace writerfilter::rtftok + +/* vim:set shiftwidth=4 softtabstop=4 expandtab: */ |