org.sejda.impl.sambox.component.optimization.OptimizationRuler Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of sejda-sambox Show documentation
Show all versions of sejda-sambox Show documentation
Package containing tasks implemented using sambox.
/*
* Created on 03 feb 2016
* Copyright 2015 by Andrea Vacondio ([email protected]).
* This file is part of Sejda.
*
* Sejda is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Sejda is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Sejda. If not, see {
private static final Logger LOG = LoggerFactory.getLogger(OptimizationRuler.class);
private OptimizationPolicy policy;
public OptimizationRuler(OptimizationPolicy policy) {
requireNotNullArg(policy, "Optimization policy cannot be null");
this.policy = policy;
}
@Override
public Boolean apply(PDDocument document) {
if (policy == OptimizationPolicy.YES) {
return true;
}
if (policy == OptimizationPolicy.AUTO) {
return willNeedOptimization(document);
}
return false;
}
private boolean willNeedOptimization(PDDocument document) {
return hasSharedNameDictionaries(document) || hasInheritedResources(document);
}
/**
* @param document
* @return true if the document page tree has non leaf nodes with fonts or images resources, inherited by page leaves.
*/
private boolean hasInheritedResources(PDDocument document) {
// we take all the resource dictionaries in non-leaf nodes (i.e. inherited by pages) and count fonts and the xobjects of subtype Image, so basically we try to determine if
// pages are going to inherit images or fonts, potentially unused in which case we want to optimize
List resources = document.getPages().streamNodes().filter(PDPageTree::isPageTreeNode)
.map(d -> d.getDictionaryObject(COSName.RESOURCES, COSDictionary.class)).filter(Objects::nonNull)
.distinct().toList();
long inheritedImage = resources.stream().map(d -> d.getDictionaryObject(COSName.XOBJECT, COSDictionary.class))
.filter(Objects::nonNull).flatMap(d -> d.getValues().stream()).map(COSBase::getCOSObject)
.filter(d -> d instanceof COSDictionary).map(d -> (COSDictionary) d)
.map(d -> d.getNameAsString(COSName.SUBTYPE)).filter(Objects::nonNull)
.filter(COSName.IMAGE.getName()::equals).count();
long inheritedFonts = resources.stream().map(d -> d.getDictionaryObject(COSName.FONT, COSDictionary.class))
.filter(Objects::nonNull).flatMap(d -> d.getValues().stream()).map(COSBase::getCOSObject)
.filter(d -> d instanceof COSDictionary).count();
long inheritedExtGState = resources.stream()
.map(d -> d.getDictionaryObject(COSName.EXT_G_STATE, COSDictionary.class)).filter(Objects::nonNull)
.flatMap(d -> d.getValues().stream()).map(COSBase::getCOSObject).filter(d -> d instanceof COSDictionary)
.count();
LOG.debug("Found {} inherited images, {} inherited fonts and {} inherited graphic states potentially unused",
inheritedImage, inheritedFonts, inheritedExtGState);
return (inheritedImage + inheritedFonts + inheritedExtGState) > 0;
}
/**
*
* @param document
* @return true if we detect optimizable name resource dictionaries that are shared among pages. Being shared they likely contains resources used by multiple pages so we need
* to optimize in case of split/extract to avoid carrying unused resources to the extracted pages
*/
private boolean hasSharedNameDictionaries(PDDocument document) {
// we get from all the pages resource dictionaries, all the optimizable resources name dictionaries (fonts, xobject, extgstate)
List optimizableDictionaries = document.getPages().stream().map(PDPage::getCOSObject)
.filter(Objects::nonNull).map(d -> d.getDictionaryObject(COSName.RESOURCES, COSDictionary.class))
.filter(Objects::nonNull).flatMap(
d -> Stream.of(d.getDictionaryObject(COSName.EXT_G_STATE, COSDictionary.class),
d.getDictionaryObject(COSName.XOBJECT, COSDictionary.class),
d.getDictionaryObject(COSName.FONT, COSDictionary.class))).filter(Objects::nonNull)
.filter(x -> x.size() > 0).toList();
long distinctFontDictionaries = optimizableDictionaries.stream().distinct().count();
if (optimizableDictionaries.size() > distinctFontDictionaries) {
// if the distinct count is different it means one or more name dictionaries is shared among pages
LOG.debug("Found shared named dictionaries");
return true;
}
return false;
}
}