/* -*- 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 #include #include #include #include #include "style.hxx" #include #include #include #include #include #include #include #include using namespace com::sun::star; namespace pdfi { PDFIProcessor::PDFIProcessor( const uno::Reference< task::XStatusIndicator >& xStat , css::uno::Reference< css::uno::XComponentContext > const & xContext) : m_xContext(xContext), prevCharWidth(0), m_pDocument( ElementFactory::createDocumentElement() ), m_pCurPage(nullptr), m_pCurElement(nullptr), m_nNextFontId( 1 ), m_nNextGCId( 1 ), m_nPages(0), m_nNextZOrder( 1 ), m_xStatusIndicator( xStat ) { FontAttributes aDefFont; aDefFont.familyName = "Helvetica"; aDefFont.fontWeight = u"normal"; aDefFont.isItalic = false; aDefFont.size = 10*PDFI_OUTDEV_RESOLUTION/72; m_aIdToFont.insert({0, aDefFont}); m_aFontToId.insert({aDefFont, 0}); GraphicsContext aDefGC; m_aGCStack.push_back( aDefGC ); m_aGCToId.insert({aDefGC, 0}); m_aIdToGC.insert({0, aDefGC}); } void PDFIProcessor::setPageNum( sal_Int32 nPages ) { m_nPages = nPages; } void PDFIProcessor::pushState() { GraphicsContextStack::value_type const a(m_aGCStack.back()); m_aGCStack.push_back(a); } void PDFIProcessor::popState() { m_aGCStack.pop_back(); } void PDFIProcessor::setFlatness( double value ) { getCurrentContext().Flatness = value; } void PDFIProcessor::setTransformation( const geometry::AffineMatrix2D& rMatrix ) { basegfx::unotools::homMatrixFromAffineMatrix( getCurrentContext().Transformation, rMatrix ); } void PDFIProcessor::setLineDash( const uno::Sequence& dashes, double /*start*/ ) { // TODO(F2): factor in start offset GraphicsContext& rContext( getCurrentContext() ); comphelper::sequenceToContainer(rContext.DashArray,dashes); } void PDFIProcessor::setLineJoin(sal_Int8 nJoin) { getCurrentContext().LineJoin = nJoin; } void PDFIProcessor::setLineCap(sal_Int8 nCap) { getCurrentContext().LineCap = nCap; } void PDFIProcessor::setMiterLimit(double) { SAL_WARN("sdext.pdfimport", "PDFIProcessor::setMiterLimit(): not supported by ODF"); } void PDFIProcessor::setLineWidth(double nWidth) { getCurrentContext().LineWidth = nWidth; } void PDFIProcessor::setFillColor( const rendering::ARGBColor& rColor ) { getCurrentContext().FillColor = rColor; } void PDFIProcessor::setStrokeColor( const rendering::ARGBColor& rColor ) { getCurrentContext().LineColor = rColor; } void PDFIProcessor::setFont( const FontAttributes& i_rFont ) { FontAttributes aChangedFont( i_rFont ); GraphicsContext& rGC=getCurrentContext(); // for text render modes, please see PDF reference manual if (rGC.TextRenderMode == 1) { aChangedFont.isOutline = true; } else if (rGC.TextRenderMode == 2) { // tdf#81484: faux bold is represented as "stroke+fill" (while using the same color for both stroke and fill) in pdf. // Convert to bold instead if the stroke color is the same as the fill color, // otherwise it should be outline. if (getCurrentContext().LineColor == getCurrentContext().FillColor) aChangedFont.fontWeight = u"bold"; else aChangedFont.isOutline = true; } FontToIdMap::const_iterator it = m_aFontToId.find( aChangedFont ); if( it != m_aFontToId.end() ) rGC.FontId = it->second; else { m_aFontToId[ aChangedFont ] = m_nNextFontId; m_aIdToFont[ m_nNextFontId ] = aChangedFont; rGC.FontId = m_nNextFontId; m_nNextFontId++; } } void PDFIProcessor::setTextRenderMode( sal_Int32 i_nMode ) { GraphicsContext& rGC=getCurrentContext(); rGC.TextRenderMode = i_nMode; IdToFontMap::iterator it = m_aIdToFont.find( rGC.FontId ); if( it != m_aIdToFont.end() ) setFont( it->second ); } sal_Int32 PDFIProcessor::getFontId( const FontAttributes& rAttr ) const { const sal_Int32 nCurFont = getCurrentContext().FontId; const_cast(this)->setFont( rAttr ); const sal_Int32 nFont = getCurrentContext().FontId; const_cast(this)->getCurrentContext().FontId = nCurFont; return nFont; } // line diagnose block - start void PDFIProcessor::processGlyphLine() { if (m_GlyphsList.empty()) return; double spaceDetectBoundary = 0.0; // Try to find space glyph and its width for (CharGlyph & i : m_GlyphsList) { OUString& glyph = i.getGlyph(); sal_Unicode ch = '\0'; if (!glyph.isEmpty()) ch = glyph[0]; if ((ch == 0x20) || (ch == 0xa0)) { double spaceWidth = i.getWidth(); spaceDetectBoundary = spaceWidth * 0.5; break; } } // If space glyph is not found, use average glyph width instead if (spaceDetectBoundary == 0.0) { double avgGlyphWidth = 0.0; for (const CharGlyph & i : m_GlyphsList) avgGlyphWidth += i.getWidth(); avgGlyphWidth /= m_GlyphsList.size(); spaceDetectBoundary = avgGlyphWidth * 0.2; } FrameElement* frame = ElementFactory::createFrameElement( m_GlyphsList[0].getCurElement(), getGCId(m_GlyphsList[0].getGC())); frame->ZOrder = m_nNextZOrder++; frame->IsForText = true; frame->FontSize = getFont(m_GlyphsList[0].getGC().FontId).size; ParagraphElement* para = ElementFactory::createParagraphElement(frame); for (size_t i = 0; i < m_GlyphsList.size(); i++) { bool prependSpace = false; TextElement* text = ElementFactory::createTextElement( para, getGCId(m_GlyphsList[i].getGC()), m_GlyphsList[i].getGC().FontId); if (i == 0) { text->x = m_GlyphsList[0].getGC().Transformation.get(0, 2); text->y = m_GlyphsList[0].getGC().Transformation.get(1, 2); text->w = 0; text->h = 0; para->updateGeometryWith(text); frame->updateGeometryWith(para); } else { double spaceSize = m_GlyphsList[i].getPrevSpaceWidth(); prependSpace = spaceSize > spaceDetectBoundary; } if (prependSpace) text->Text.append(" "); text->Text.append(m_GlyphsList[i].getGlyph()); } m_GlyphsList.clear(); } void PDFIProcessor::drawGlyphs( const OUString& rGlyphs, const geometry::RealRectangle2D& rRect, const geometry::Matrix2D& rFontMatrix, double fontSize) { double ascent = getFont(getCurrentContext().FontId).ascent; basegfx::B2DHomMatrix fontMatrix( rFontMatrix.m00, rFontMatrix.m01, 0.0, rFontMatrix.m10, rFontMatrix.m11, 0.0); fontMatrix.scale(fontSize, fontSize); basegfx::B2DHomMatrix totalTextMatrix1(fontMatrix); basegfx::B2DHomMatrix totalTextMatrix2(fontMatrix); totalTextMatrix1.translate(rRect.X1, rRect.Y1); totalTextMatrix2.translate(rRect.X2, rRect.Y2); basegfx::B2DHomMatrix corrMatrix; corrMatrix.scale(1.0, -1.0); corrMatrix.translate(0.0, ascent); totalTextMatrix1 = totalTextMatrix1 * corrMatrix; totalTextMatrix2 = totalTextMatrix2 * corrMatrix; totalTextMatrix1 *= getCurrentContext().Transformation; totalTextMatrix2 *= getCurrentContext().Transformation; basegfx::B2DHomMatrix invMatrix(totalTextMatrix1); basegfx::B2DHomMatrix invPrevMatrix(prevTextMatrix); invMatrix.invert(); invPrevMatrix.invert(); basegfx::B2DHomMatrix offsetMatrix1(totalTextMatrix1); basegfx::B2DHomMatrix offsetMatrix2(totalTextMatrix2); offsetMatrix1 *= invPrevMatrix; offsetMatrix2 *= invMatrix; double charWidth = offsetMatrix2.get(0, 2); double prevSpaceWidth = offsetMatrix1.get(0, 2) - prevCharWidth; if ((totalTextMatrix1.get(0, 0) != prevTextMatrix.get(0, 0)) || (totalTextMatrix1.get(0, 1) != prevTextMatrix.get(0, 1)) || (totalTextMatrix1.get(1, 0) != prevTextMatrix.get(1, 0)) || (totalTextMatrix1.get(1, 1) != prevTextMatrix.get(1, 1)) || (offsetMatrix1.get(0, 2) < 0.0) || (prevSpaceWidth > prevCharWidth * 1.3) || (!basegfx::fTools::equalZero(offsetMatrix1.get(1, 2), 0.0001))) { processGlyphLine(); } CharGlyph aGlyph(m_pCurElement, getCurrentContext(), charWidth, prevSpaceWidth, rGlyphs); aGlyph.getGC().Transformation = totalTextMatrix1; m_GlyphsList.push_back(aGlyph); prevCharWidth = charWidth; prevTextMatrix = totalTextMatrix1; } void PDFIProcessor::endText() { TextElement* pText = m_pCurElement->dynCastAsTextElement(); if( pText ) m_pCurElement = pText->Parent; } void PDFIProcessor::setupImage(ImageId nImage) { const GraphicsContext& rGC(getCurrentContext()); basegfx::B2DTuple aScale, aTranslation; double fRotate, fShearX; rGC.Transformation.decompose(aScale, aTranslation, fRotate, fShearX); const sal_Int32 nGCId = getGCId(rGC); FrameElement* pFrame = ElementFactory::createFrameElement( m_pCurElement, nGCId ); ImageElement* pImageElement = ElementFactory::createImageElement( pFrame, nGCId, nImage ); pFrame->x = pImageElement->x = aTranslation.getX(); pFrame->y = pImageElement->y = aTranslation.getY(); pFrame->w = pImageElement->w = aScale.getX(); pFrame->h = pImageElement->h = aScale.getY(); pFrame->ZOrder = m_nNextZOrder++; // Poppler wrapper takes into account that vertical axes of PDF and ODF are opposite, // and it flips matrix vertically (see poppler's GfxState::GfxState()). // But image internal vertical axis is independent of PDF vertical axis direction, // so arriving matrix is extra-flipped relative to image. // We force vertical flip here to compensate that. pFrame->MirrorVertical = true; } void PDFIProcessor::drawMask(const uno::Sequence& xBitmap, bool /*bInvert*/ ) { // TODO(F3): Handle mask and inversion setupImage( m_aImages.addImage(xBitmap) ); } void PDFIProcessor::drawImage(const uno::Sequence& xBitmap ) { setupImage( m_aImages.addImage(xBitmap) ); } void PDFIProcessor::drawColorMaskedImage(const uno::Sequence& xBitmap, const uno::Sequence& /*xMaskColors*/ ) { // TODO(F3): Handle mask colors setupImage( m_aImages.addImage(xBitmap) ); } void PDFIProcessor::drawMaskedImage(const uno::Sequence& xBitmap, const uno::Sequence& /*xMask*/, bool /*bInvertMask*/) { // TODO(F3): Handle mask and inversion setupImage( m_aImages.addImage(xBitmap) ); } void PDFIProcessor::drawAlphaMaskedImage(const uno::Sequence& xBitmap, const uno::Sequence& /*xMask*/) { // TODO(F3): Handle mask setupImage( m_aImages.addImage(xBitmap) ); } void PDFIProcessor::strokePath( const uno::Reference< rendering::XPolyPolygon2D >& rPath ) { basegfx::B2DPolyPolygon aPoly=basegfx::unotools::b2DPolyPolygonFromXPolyPolygon2D(rPath); aPoly.transform(getCurrentContext().Transformation); PolyPolyElement* pPoly = ElementFactory::createPolyPolyElement( m_pCurElement, getGCId(getCurrentContext()), aPoly, PATH_STROKE ); pPoly->updateGeometry(); pPoly->ZOrder = m_nNextZOrder++; } void PDFIProcessor::fillPath( const uno::Reference< rendering::XPolyPolygon2D >& rPath ) { basegfx::B2DPolyPolygon aPoly=basegfx::unotools::b2DPolyPolygonFromXPolyPolygon2D(rPath); aPoly.transform(getCurrentContext().Transformation); PolyPolyElement* pPoly = ElementFactory::createPolyPolyElement( m_pCurElement, getGCId(getCurrentContext()), aPoly, PATH_FILL ); pPoly->updateGeometry(); pPoly->ZOrder = m_nNextZOrder++; } void PDFIProcessor::eoFillPath( const uno::Reference< rendering::XPolyPolygon2D >& rPath ) { basegfx::B2DPolyPolygon aPoly=basegfx::unotools::b2DPolyPolygonFromXPolyPolygon2D(rPath); aPoly.transform(getCurrentContext().Transformation); PolyPolyElement* pPoly = ElementFactory::createPolyPolyElement( m_pCurElement, getGCId(getCurrentContext()), aPoly, PATH_EOFILL ); pPoly->updateGeometry(); pPoly->ZOrder = m_nNextZOrder++; } void PDFIProcessor::intersectClip(const uno::Reference< rendering::XPolyPolygon2D >& rPath) { // TODO(F3): interpret fill mode basegfx::B2DPolyPolygon aNewClip = basegfx::unotools::b2DPolyPolygonFromXPolyPolygon2D(rPath); aNewClip.transform(getCurrentContext().Transformation); basegfx::B2DPolyPolygon aCurClip = getCurrentContext().Clip; if( aCurClip.count() ) // #i92985# adapted API from (..., false, false) to (..., true, false) aNewClip = basegfx::utils::clipPolyPolygonOnPolyPolygon( aCurClip, aNewClip, true, false ); getCurrentContext().Clip = aNewClip; } void PDFIProcessor::intersectEoClip(const uno::Reference< rendering::XPolyPolygon2D >& rPath) { // TODO(F3): interpret fill mode basegfx::B2DPolyPolygon aNewClip = basegfx::unotools::b2DPolyPolygonFromXPolyPolygon2D(rPath); aNewClip.transform(getCurrentContext().Transformation); basegfx::B2DPolyPolygon aCurClip = getCurrentContext().Clip; if( aCurClip.count() ) // #i92985# adapted API from (..., false, false) to (..., true, false) aNewClip = basegfx::utils::clipPolyPolygonOnPolyPolygon( aCurClip, aNewClip, true, false ); getCurrentContext().Clip = aNewClip; } void PDFIProcessor::hyperLink( const geometry::RealRectangle2D& rBounds, const OUString& rURI ) { if( !rURI.isEmpty() ) { HyperlinkElement* pLink = ElementFactory::createHyperlinkElement( &m_pCurPage->Hyperlinks, rURI ); pLink->x = rBounds.X1; pLink->y = rBounds.Y1; pLink->w = rBounds.X2-rBounds.X1; pLink->h = rBounds.Y2-rBounds.Y1; } } const FontAttributes& PDFIProcessor::getFont( sal_Int32 nFontId ) const { IdToFontMap::const_iterator it = m_aIdToFont.find( nFontId ); if( it == m_aIdToFont.end() ) it = m_aIdToFont.find( 0 ); return it->second; } sal_Int32 PDFIProcessor::getGCId( const GraphicsContext& rGC ) { sal_Int32 nGCId = 0; auto it = m_aGCToId.find( rGC ); if( it != m_aGCToId.end() ) nGCId = it->second; else { m_aGCToId.insert({rGC, m_nNextGCId}); m_aIdToGC.insert({m_nNextGCId, rGC}); nGCId = m_nNextGCId; m_nNextGCId++; } return nGCId; } const GraphicsContext& PDFIProcessor::getGraphicsContext( sal_Int32 nGCId ) const { auto it = m_aIdToGC.find( nGCId ); if( it == m_aIdToGC.end() ) it = m_aIdToGC.find( 0 ); return it->second; } void PDFIProcessor::endPage() { processGlyphLine(); // draw last line if( m_xStatusIndicator.is() && m_pCurPage && m_pCurPage->PageNumber == m_nPages ) m_xStatusIndicator->end(); } void PDFIProcessor::startPage( const geometry::RealSize2D& rSize ) { // initial clip is to page bounds getCurrentContext().Clip = basegfx::B2DPolyPolygon( basegfx::utils::createPolygonFromRect( basegfx::B2DRange( 0, 0, rSize.Width, rSize.Height ))); sal_Int32 nNextPageNr = m_pCurPage ? m_pCurPage->PageNumber+1 : 1; if( m_xStatusIndicator.is() ) { if( nNextPageNr == 1 ) startIndicator( " " ); m_xStatusIndicator->setValue( nNextPageNr ); } m_pCurPage = ElementFactory::createPageElement(m_pDocument.get(), nNextPageNr); m_pCurElement = m_pCurPage; m_pCurPage->w = rSize.Width; m_pCurPage->h = rSize.Height; m_nNextZOrder = 1; } void PDFIProcessor::emit( XmlEmitter& rEmitter, const TreeVisitorFactory& rVisitorFactory ) { #if OSL_DEBUG_LEVEL > 0 m_pDocument->emitStructure( 0 ); #endif ElementTreeVisitorSharedPtr optimizingVisitor( rVisitorFactory.createOptimizingVisitor(*this)); // FIXME: localization startIndicator( " " ); m_pDocument->visitedBy( *optimizingVisitor, std::list>::const_iterator()); #if OSL_DEBUG_LEVEL > 0 m_pDocument->emitStructure( 0 ); #endif // get styles StyleContainer aStyles; ElementTreeVisitorSharedPtr finalizingVisitor( rVisitorFactory.createStyleCollectingVisitor(aStyles,*this)); // FIXME: localization m_pDocument->visitedBy( *finalizingVisitor, std::list>::const_iterator() ); EmitContext aContext( rEmitter, aStyles, m_aImages, *this, m_xStatusIndicator, m_xContext ); ElementTreeVisitorSharedPtr aEmittingVisitor( rVisitorFactory.createEmittingVisitor(aContext)); PropertyMap aProps; // document prolog #define OASIS_STR "urn:oasis:names:tc:opendocument:xmlns:" aProps[ "xmlns:office" ] = OASIS_STR "office:1.0" ; aProps[ "xmlns:style" ] = OASIS_STR "style:1.0" ; aProps[ "xmlns:text" ] = OASIS_STR "text:1.0" ; aProps[ "xmlns:svg" ] = OASIS_STR "svg-compatible:1.0" ; aProps[ "xmlns:table" ] = OASIS_STR "table:1.0" ; aProps[ "xmlns:draw" ] = OASIS_STR "drawing:1.0" ; aProps[ "xmlns:fo" ] = OASIS_STR "xsl-fo-compatible:1.0" ; aProps[ "xmlns:xlink"] = "http://www.w3.org/1999/xlink"; aProps[ "xmlns:dc"] = "http://purl.org/dc/elements/1.1/"; aProps[ "xmlns:number"] = OASIS_STR "datastyle:1.0" ; aProps[ "xmlns:presentation"] = OASIS_STR "presentation:1.0" ; aProps[ "xmlns:math"] = "http://www.w3.org/1998/Math/MathML"; aProps[ "xmlns:form"] = OASIS_STR "form:1.0" ; aProps[ "xmlns:script"] = OASIS_STR "script:1.0" ; aProps[ "xmlns:dom"] = "http://www.w3.org/2001/xml-events"; aProps[ "xmlns:xforms"] = "http://www.w3.org/2002/xforms"; aProps[ "xmlns:xsd"] = "http://www.w3.org/2001/XMLSchema"; aProps[ "xmlns:xsi"] = "http://www.w3.org/2001/XMLSchema-instance"; aProps[ "office:version" ] = "1.0"; aContext.rEmitter.beginTag( "office:document", aProps ); // emit style list aStyles.emit( aContext, *aEmittingVisitor ); m_pDocument->visitedBy( *aEmittingVisitor, std::list>::const_iterator() ); aContext.rEmitter.endTag( "office:document" ); endIndicator(); } void PDFIProcessor::startIndicator( const OUString& rText ) { sal_Int32 nElements = m_nPages; if( !m_xStatusIndicator.is() ) return; sal_Int32 nLength = rText.getLength(); OUStringBuffer aStr( nLength*2 ); const sal_Unicode* pText = rText.getStr(); for( int i = 0; i < nLength; i++ ) { if( nLength-i > 1&& pText[i] == '%' && pText[i+1] == 'd' ) { aStr.append( nElements ); i++; } else aStr.append( pText[i] ); } m_xStatusIndicator->start( aStr.makeStringAndClear(), nElements ); } void PDFIProcessor::endIndicator() { if( m_xStatusIndicator.is() ) m_xStatusIndicator->end(); } static bool lr_tb_sort( std::unique_ptr const & pLeft, std::unique_ptr const & pRight ) { // Ensure irreflexivity (which could be compromised if h or w is negative): if (pLeft == pRight) return false; // first: top-bottom sorting // Note: allow for 10% overlap on text lines since text lines are usually // of the same order as font height whereas the real paint area // of text is usually smaller double fudge_factor_left = 0.0, fudge_factor_right = 0.0; if( pLeft->dynCastAsTextElement() ) fudge_factor_left = 0.1; if( pRight->dynCastAsTextElement() ) fudge_factor_right = 0.1; // Allow negative height double lower_boundary_left = pLeft->y + std::max(pLeft->h, 0.0) - fabs(pLeft->h) * fudge_factor_left; double lower_boundary_right = pRight->y + std::max(pRight->h, 0.0) - fabs(pRight->h) * fudge_factor_right; double upper_boundary_left = pLeft->y + std::min(pLeft->h, 0.0); double upper_boundary_right = pRight->y + std::min(pRight->h, 0.0); // if left's lower boundary is above right's upper boundary // then left is smaller if( lower_boundary_left < upper_boundary_right ) return true; // if right's lower boundary is above left's upper boundary // then left is definitely not smaller if( lower_boundary_right < upper_boundary_left ) return false; // Allow negative width double left_boundary_left = pLeft->y + std::min(pLeft->w, 0.0); double left_boundary_right = pRight->y + std::min(pRight->w, 0.0); double right_boundary_left = pLeft->y + std::max(pLeft->w, 0.0); double right_boundary_right = pRight->y + std::max(pRight->w, 0.0); // by now we have established that left and right are inside // a "line", that is they have vertical overlap // second: left-right sorting // if left's right boundary is left to right's left boundary // then left is smaller if( right_boundary_left < left_boundary_right ) return true; // if right's right boundary is left to left's left boundary // then left is definitely not smaller if( right_boundary_right < left_boundary_left ) return false; // here we have established vertical and horizontal overlap // so sort left first, top second if( pLeft->x < pRight->x ) return true; if( pRight->x < pLeft->x ) return false; if( pLeft->y < pRight->y ) return true; return false; } void PDFIProcessor::sortElements(Element* pEle) { if( pEle->Children.empty() ) return; // sort method from std::list is equivalent to stable_sort // See S Meyers, Effective STL pEle->Children.sort(lr_tb_sort); } /* Produce mirrored-image for each code point which has the Bidi_Mirrored property, within a string. This need to be done in forward order. */ OUString PDFIProcessor::SubstituteBidiMirrored(const OUString& rString) { const sal_Int32 nLen = rString.getLength(); OUStringBuffer aMirror(nLen); for (sal_Int32 i = 0; i < nLen;) { const sal_uInt32 nCodePoint = rString.iterateCodePoints(&i); aMirror.appendUtf32(unicode::GetMirroredChar(nCodePoint)); } return aMirror.makeStringAndClear(); } } /* vim:set shiftwidth=4 softtabstop=4 expandtab: */