Examples with PDOutlineItem com.tom_roush.pdfbox.pdmodel.interactive.documentnavigation.outline.PDOutlineItem used on opensource projects

Example 1 with PDOutlineItem

use of com.tom_roush.pdfbox.pdmodel.interactive.documentnavigation.outline.PDOutlineItem in project PdfBox-Android by TomRoush.

the class TestPDPageTree method indexOfPageFromOutlineDestination.

@Test
public void indexOfPageFromOutlineDestination() throws IOException {
    doc = PDDocument.load(TestPDPageTree.class.getResourceAsStream("/pdfbox/com/tom_roush/pdfbox/pdmodel/with_outline.pdf"));
    PDDocumentOutline outline = doc.getDocumentCatalog().getDocumentOutline();
    for (PDOutlineItem current : outline.children()) {
        if (current.getTitle().contains("Second")) {
            assertEquals(2, doc.getPages().indexOf(current.findDestinationPage(doc)));
        }
    }
}

Example 2 with PDOutlineItem

use of com.tom_roush.pdfbox.pdmodel.interactive.documentnavigation.outline.PDOutlineItem in project PdfBox-Android by TomRoush.

the class PDFMergerUtility method appendDocument.

/**
 * append all pages from source to destination.
 *
 * @param destination the document to receive the pages
 * @param source the document originating the new pages
 *
 * @throws IOException If there is an error accessing data from either
 * document.
 */
public void appendDocument(PDDocument destination, PDDocument source) throws IOException {
    if (source.getDocument().isClosed()) {
        throw new IOException("Error: source PDF is closed.");
    }
    if (destination.getDocument().isClosed()) {
        throw new IOException("Error: destination PDF is closed.");
    }
    PDDocumentCatalog destCatalog = destination.getDocumentCatalog();
    PDDocumentCatalog srcCatalog = source.getDocumentCatalog();
    if (isDynamicXfa(srcCatalog.getAcroForm())) {
        throw new IOException("Error: can't merge source document containing dynamic XFA form content.");
    }
    PDDocumentInformation destInfo = destination.getDocumentInformation();
    PDDocumentInformation srcInfo = source.getDocumentInformation();
    mergeInto(srcInfo.getCOSObject(), destInfo.getCOSObject(), Collections.<COSName>emptySet());
    // use the highest version number for the resulting pdf
    float destVersion = destination.getVersion();
    float srcVersion = source.getVersion();
    if (destVersion < srcVersion) {
        destination.setVersion(srcVersion);
    }
    int pageIndexOpenActionDest = -1;
    if (destCatalog.getOpenAction() == null) {
        // PDFBOX-3972: get local dest page index, it must be reassigned after the page cloning
        PDDestinationOrAction openAction = null;
        try {
            openAction = srcCatalog.getOpenAction();
        } catch (IOException ex) {
            // PDFBOX-4223
            Log.e("PdfBox-Android", "Invalid OpenAction ignored", ex);
        }
        PDDestination openActionDestination = null;
        if (openAction instanceof PDActionGoTo) {
            openActionDestination = ((PDActionGoTo) openAction).getDestination();
        } else if (openAction instanceof PDDestination) {
            openActionDestination = (PDDestination) openAction;
        }
        if (openActionDestination instanceof PDPageDestination) {
            PDPage page = ((PDPageDestination) openActionDestination).getPage();
            if (page != null) {
                pageIndexOpenActionDest = srcCatalog.getPages().indexOf(page);
            }
        }
        destCatalog.setOpenAction(openAction);
    }
    PDFCloneUtility cloner = new PDFCloneUtility(destination);
    mergeAcroForm(cloner, destCatalog, srcCatalog);
    COSArray destThreads = (COSArray) destCatalog.getCOSObject().getDictionaryObject(COSName.THREADS);
    COSArray srcThreads = (COSArray) cloner.cloneForNewDocument(destCatalog.getCOSObject().getDictionaryObject(COSName.THREADS));
    if (destThreads == null) {
        destCatalog.getCOSObject().setItem(COSName.THREADS, srcThreads);
    } else {
        destThreads.addAll(srcThreads);
    }
    PDDocumentNameDictionary destNames = destCatalog.getNames();
    PDDocumentNameDictionary srcNames = srcCatalog.getNames();
    if (srcNames != null) {
        if (destNames == null) {
            destCatalog.getCOSObject().setItem(COSName.NAMES, cloner.cloneForNewDocument(srcNames));
        } else {
            cloner.cloneMerge(srcNames, destNames);
        }
    }
    if (destNames != null) {
        // found in 054080.pdf from PDFBOX-4417 and doesn't belong there
        destNames.getCOSObject().removeItem(COSName.ID_TREE);
        Log.w("PdfBox-Android", "Removed /IDTree from /Names dictionary, doesn't belong there");
    }
    PDDocumentNameDestinationDictionary destDests = destCatalog.getDests();
    PDDocumentNameDestinationDictionary srcDests = srcCatalog.getDests();
    if (srcDests != null) {
        if (destDests == null) {
            destCatalog.getCOSObject().setItem(COSName.DESTS, cloner.cloneForNewDocument(srcDests));
        } else {
            cloner.cloneMerge(srcDests, destDests);
        }
    }
    PDDocumentOutline destOutline = destCatalog.getDocumentOutline();
    PDDocumentOutline srcOutline = srcCatalog.getDocumentOutline();
    if (srcOutline != null) {
        if (destOutline == null || destOutline.getFirstChild() == null) {
            PDDocumentOutline cloned = new PDDocumentOutline((COSDictionary) cloner.cloneForNewDocument(srcOutline));
            destCatalog.setDocumentOutline(cloned);
        } else {
            // search last sibling for dest, because /Last entry is sometimes wrong
            PDOutlineItem destLastOutlineItem = destOutline.getFirstChild();
            while (destLastOutlineItem.getNextSibling() != null) {
                destLastOutlineItem = destLastOutlineItem.getNextSibling();
            }
            for (PDOutlineItem item : srcOutline.children()) {
                // get each child, clone its dictionary, remove siblings info,
                // append outline item created from there
                COSDictionary clonedDict = (COSDictionary) cloner.cloneForNewDocument(item);
                clonedDict.removeItem(COSName.PREV);
                clonedDict.removeItem(COSName.NEXT);
                PDOutlineItem clonedItem = new PDOutlineItem(clonedDict);
                destLastOutlineItem.insertSiblingAfter(clonedItem);
                destLastOutlineItem = destLastOutlineItem.getNextSibling();
            }
        }
    }
    PageMode destPageMode = destCatalog.getPageMode();
    PageMode srcPageMode = srcCatalog.getPageMode();
    if (destPageMode == null) {
        destCatalog.setPageMode(srcPageMode);
    }
    COSDictionary destLabels = destCatalog.getCOSObject().getCOSDictionary(COSName.PAGE_LABELS);
    COSDictionary srcLabels = srcCatalog.getCOSObject().getCOSDictionary(COSName.PAGE_LABELS);
    if (srcLabels != null) {
        int destPageCount = destination.getNumberOfPages();
        COSArray destNums;
        if (destLabels == null) {
            destLabels = new COSDictionary();
            destNums = new COSArray();
            destLabels.setItem(COSName.NUMS, destNums);
            destCatalog.getCOSObject().setItem(COSName.PAGE_LABELS, destLabels);
        } else {
            destNums = (COSArray) destLabels.getDictionaryObject(COSName.NUMS);
        }
        COSArray srcNums = (COSArray) srcLabels.getDictionaryObject(COSName.NUMS);
        if (srcNums != null) {
            int startSize = destNums.size();
            for (int i = 0; i < srcNums.size(); i += 2) {
                COSBase base = srcNums.getObject(i);
                if (!(base instanceof COSNumber)) {
                    Log.e("PdfBox-Android", "page labels ignored, index " + i + " should be a number, but is " + base);
                    // remove what we added
                    while (destNums.size() > startSize) {
                        destNums.remove(startSize);
                    }
                    break;
                }
                COSNumber labelIndex = (COSNumber) base;
                long labelIndexValue = labelIndex.intValue();
                destNums.add(COSInteger.get(labelIndexValue + destPageCount));
                destNums.add(cloner.cloneForNewDocument(srcNums.getObject(i + 1)));
            }
        }
    }
    COSStream destMetadata = destCatalog.getCOSObject().getCOSStream(COSName.METADATA);
    COSStream srcMetadata = srcCatalog.getCOSObject().getCOSStream(COSName.METADATA);
    if (destMetadata == null && srcMetadata != null) {
        try {
            PDStream newStream = new PDStream(destination, srcMetadata.createInputStream(), (COSName) null);
            mergeInto(srcMetadata, newStream.getCOSObject(), new HashSet<COSName>(Arrays.asList(COSName.FILTER, COSName.LENGTH)));
            destCatalog.getCOSObject().setItem(COSName.METADATA, newStream);
        } catch (IOException ex) {
            // PDFBOX-4227 cleartext XMP stream with /Flate
            Log.e("PdfBox-Android", "Metadata skipped because it could not be read", ex);
        }
    }
    COSDictionary destOCP = destCatalog.getCOSObject().getCOSDictionary(COSName.OCPROPERTIES);
    COSDictionary srcOCP = srcCatalog.getCOSObject().getCOSDictionary(COSName.OCPROPERTIES);
    if (destOCP == null && srcOCP != null) {
        destCatalog.getCOSObject().setItem(COSName.OCPROPERTIES, cloner.cloneForNewDocument(srcOCP));
    } else if (destOCP != null && srcOCP != null) {
        cloner.cloneMerge(srcOCP, destOCP);
    }
    mergeOutputIntents(cloner, srcCatalog, destCatalog);
    // merge logical structure hierarchy
    boolean mergeStructTree = false;
    int destParentTreeNextKey = -1;
    Map<Integer, COSObjectable> srcNumberTreeAsMap = null;
    Map<Integer, COSObjectable> destNumberTreeAsMap = null;
    PDStructureTreeRoot srcStructTree = srcCatalog.getStructureTreeRoot();
    PDStructureTreeRoot destStructTree = destCatalog.getStructureTreeRoot();
    if (destStructTree == null && srcStructTree != null) {
        // create a dummy structure tree in the destination, so that the source
        // tree is cloned. (We can't just copy the tree reference due to PDFBOX-3999)
        destStructTree = new PDStructureTreeRoot();
        destCatalog.setStructureTreeRoot(destStructTree);
        destStructTree.setParentTree(new PDNumberTreeNode(PDParentTreeValue.class));
        // PDFBOX-4429: remove bogus StructParent(s)
        for (PDPage page : destCatalog.getPages()) {
            page.getCOSObject().removeItem(COSName.STRUCT_PARENTS);
            for (PDAnnotation ann : page.getAnnotations()) {
                ann.getCOSObject().removeItem(COSName.STRUCT_PARENT);
            }
        }
    }
    if (destStructTree != null) {
        PDNumberTreeNode destParentTree = destStructTree.getParentTree();
        destParentTreeNextKey = destStructTree.getParentTreeNextKey();
        if (destParentTree != null) {
            destNumberTreeAsMap = getNumberTreeAsMap(destParentTree);
            if (destParentTreeNextKey < 0) {
                if (destNumberTreeAsMap.isEmpty()) {
                    destParentTreeNextKey = 0;
                } else {
                    destParentTreeNextKey = Collections.max(destNumberTreeAsMap.keySet()) + 1;
                }
            }
            if (destParentTreeNextKey >= 0 && srcStructTree != null) {
                PDNumberTreeNode srcParentTree = srcStructTree.getParentTree();
                if (srcParentTree != null) {
                    srcNumberTreeAsMap = getNumberTreeAsMap(srcParentTree);
                    if (!srcNumberTreeAsMap.isEmpty()) {
                        mergeStructTree = true;
                    }
                }
            }
        }
    }
    Map<COSDictionary, COSDictionary> objMapping = new HashMap<COSDictionary, COSDictionary>();
    int pageIndex = 0;
    for (PDPage page : srcCatalog.getPages()) {
        PDPage newPage = new PDPage((COSDictionary) cloner.cloneForNewDocument(page.getCOSObject()));
        if (!mergeStructTree) {
            // PDFBOX-4429: remove bogus StructParent(s)
            newPage.getCOSObject().removeItem(COSName.STRUCT_PARENTS);
            for (PDAnnotation ann : newPage.getAnnotations()) {
                ann.getCOSObject().removeItem(COSName.STRUCT_PARENT);
            }
        }
        newPage.setCropBox(page.getCropBox());
        newPage.setMediaBox(page.getMediaBox());
        newPage.setRotation(page.getRotation());
        PDResources resources = page.getResources();
        if (resources != null) {
            // this is smart enough to just create references for resources that are used on multiple pages
            newPage.setResources(new PDResources((COSDictionary) cloner.cloneForNewDocument(resources)));
        } else {
            newPage.setResources(new PDResources());
        }
        if (mergeStructTree) {
            // add the value of the destination ParentTreeNextKey to every source element
            // StructParent(s) value so that these don't overlap with the existing values
            updateStructParentEntries(newPage, destParentTreeNextKey);
            objMapping.put(page.getCOSObject(), newPage.getCOSObject());
            List<PDAnnotation> oldAnnots = page.getAnnotations();
            List<PDAnnotation> newAnnots = newPage.getAnnotations();
            for (int i = 0; i < oldAnnots.size(); i++) {
                objMapping.put(oldAnnots.get(i).getCOSObject(), newAnnots.get(i).getCOSObject());
            }
        // TODO update mapping for XObjects
        }
        destination.addPage(newPage);
        if (pageIndex == pageIndexOpenActionDest) {
            // PDFBOX-3972: reassign the page.
            // The openAction is either a PDActionGoTo or a PDPageDestination
            PDDestinationOrAction openAction = destCatalog.getOpenAction();
            PDPageDestination pageDestination;
            if (openAction instanceof PDActionGoTo) {
                pageDestination = (PDPageDestination) ((PDActionGoTo) openAction).getDestination();
            } else {
                pageDestination = (PDPageDestination) openAction;
            }
            pageDestination.setPage(newPage);
        }
        ++pageIndex;
    }
    if (mergeStructTree) {
        updatePageReferences(cloner, srcNumberTreeAsMap, objMapping);
        int maxSrcKey = -1;
        for (Map.Entry<Integer, COSObjectable> entry : srcNumberTreeAsMap.entrySet()) {
            int srcKey = entry.getKey();
            maxSrcKey = Math.max(srcKey, maxSrcKey);
            destNumberTreeAsMap.put(destParentTreeNextKey + srcKey, cloner.cloneForNewDocument(entry.getValue()));
        }
        destParentTreeNextKey += maxSrcKey + 1;
        PDNumberTreeNode newParentTreeNode = new PDNumberTreeNode(PDParentTreeValue.class);
        // Note that all elements are stored flatly. This could become a problem for large files
        // when these are opened in a viewer that uses the tagging information.
        // If this happens, then ​PDNumberTreeNode should be improved with a convenience method that
        // stores the map into a B+Tree, see https://en.wikipedia.org/wiki/B+_tree
        newParentTreeNode.setNumbers(destNumberTreeAsMap);
        destStructTree.setParentTree(newParentTreeNode);
        destStructTree.setParentTreeNextKey(destParentTreeNextKey);
        mergeKEntries(cloner, srcStructTree, destStructTree);
        mergeRoleMap(srcStructTree, destStructTree);
        mergeIDTree(cloner, srcStructTree, destStructTree);
        mergeMarkInfo(destCatalog, srcCatalog);
        mergeLanguage(destCatalog, srcCatalog);
        mergeViewerPreferences(destCatalog, srcCatalog);
    }
}

Example 3 with PDOutlineItem

use of com.tom_roush.pdfbox.pdmodel.interactive.documentnavigation.outline.PDOutlineItem in project PdfBox-Android by TomRoush.

the class TestTextStripper method testStripByOutlineItems.

/**
 * Test whether stripping controlled by outline items works properly. The test file has 4
 * outline items at the top level, that point to 0-based pages 0, 2, 3 and 4. We are testing
 * text stripping by outlines pointing to 0-based pages 2 and 3, and also text stripping of the
 * 0-based page 2. The test makes sure that the output is different to a complete strip, not
 * empty, different to each other when different bookmark intervals are used, but identical from
 * bookmark intervals to strips with page intervals. When fed with orphan bookmarks, stripping
 * must be empty.
 *
 * @throws IOException
 */
@Test
public void testStripByOutlineItems() throws IOException, URISyntaxException {
    PDDocument doc = PDDocument.load(new File(TestPDPageTree.class.getResource("/pdfbox/com/tom_roush/pdfbox/pdmodel/with_outline.pdf").toURI()));
    PDDocumentOutline outline = doc.getDocumentCatalog().getDocumentOutline();
    Iterable<PDOutlineItem> children = outline.children();
    Iterator<PDOutlineItem> it = children.iterator();
    PDOutlineItem oi0 = it.next();
    PDOutlineItem oi2 = it.next();
    PDOutlineItem oi3 = it.next();
    PDOutlineItem oi4 = it.next();
    assertEquals(0, findOutlineItemDestPageNum(doc, oi0));
    assertEquals(2, findOutlineItemDestPageNum(doc, oi2));
    assertEquals(3, findOutlineItemDestPageNum(doc, oi3));
    assertEquals(4, findOutlineItemDestPageNum(doc, oi4));
    String textFull = stripper.getText(doc);
    assertFalse(textFull.isEmpty());
    String expectedTextFull = "First level 1\n" + "First level 2\n" + "Fist level 3\n" + "Some content\n" + "Some other content\n" + "Second at level 1\n" + "Second level 2\n" + "Content\n" + "Third level 1\n" + "Third level 2\n" + "Third level 3\n" + "Content\n" + "Fourth level 1\n" + "Content\n" + "Content\n";
    assertEquals(expectedTextFull, textFull.replaceAll("\r", ""));
    // this should grab 0-based pages 2 and 3, i.e. 1-based pages 3 and 4
    // by their bookmarks
    stripper.setStartBookmark(oi2);
    stripper.setEndBookmark(oi3);
    String textoi23 = stripper.getText(doc);
    assertFalse(textoi23.isEmpty());
    assertFalse(textoi23.equals(textFull));
    String expectedTextoi23 = "Second at level 1\n" + "Second level 2\n" + "Content\n" + "Third level 1\n" + "Third level 2\n" + "Third level 3\n" + "Content\n";
    assertEquals(expectedTextoi23, textoi23.replaceAll("\r", ""));
    // this should grab 0-based pages 2 and 3, i.e. 1-based pages 3 and 4
    // by their page numbers
    stripper.setStartBookmark(null);
    stripper.setEndBookmark(null);
    stripper.setStartPage(3);
    stripper.setEndPage(4);
    String textp34 = stripper.getText(doc);
    assertFalse(textp34.isEmpty());
    assertFalse(textoi23.equals(textFull));
    assertTrue(textoi23.equals(textp34));
    // this should grab 0-based page 2, i.e. 1-based page 3
    // by the bookmark
    stripper.setStartBookmark(oi2);
    stripper.setEndBookmark(oi2);
    String textoi2 = stripper.getText(doc);
    assertFalse(textoi2.isEmpty());
    assertFalse(textoi2.equals(textoi23));
    assertFalse(textoi23.equals(textFull));
    String expectedTextoi2 = "Second at level 1\n" + "Second level 2\n" + "Content\n";
    assertEquals(expectedTextoi2, textoi2.replaceAll("\r", ""));
    // this should grab 0-based page 2, i.e. 1-based page 3
    // by the page number
    stripper.setStartBookmark(null);
    stripper.setEndBookmark(null);
    stripper.setStartPage(3);
    stripper.setEndPage(3);
    String textp3 = stripper.getText(doc);
    assertFalse(textp3.isEmpty());
    assertFalse(textp3.equals(textp34));
    assertFalse(textoi23.equals(textFull));
    assertTrue(textoi2.equals(textp3));
    // Test with orphan bookmark
    PDOutlineItem oiOrphan = new PDOutlineItem();
    stripper.setStartBookmark(oiOrphan);
    stripper.setEndBookmark(oiOrphan);
    String textOiOrphan = stripper.getText(doc);
    assertTrue(textOiOrphan.isEmpty());
}