[
https://issues.apache.org/jira/browse/PDFBOX-6283?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18128105#comment-18128105
]
Tilman Hausherr commented on PDFBOX-6283:
-----------------------------------------
This morning I tested with file 000671.pdf. It works nicely, however it isn't
really needed at all, because in annotations, the splitter replaces the named
destinations with their pages. This was done in PDFBOX-5840 because of problems
with file 410609.pdf which has names that are pure hex strings. I thought about
an optimization of yesterdays code and will do that, but at the end I might
just abandon the whole thing like I did in PDFBOX-5840.
> Splitting a pdf does not keep the Catalog names dict
> ----------------------------------------------------
>
> Key: PDFBOX-6283
> URL: https://issues.apache.org/jira/browse/PDFBOX-6283
> Project: PDFBox
> Issue Type: Bug
> Components: Utilities
> Affects Versions: 2.0.37, 3.0.8 PDFBox
> Reporter: Simon Steiner
> Priority: Major
> Labels: split
>
> When I split a pdf, i would expect the Catalog to have a names entry with the
> names that refer to the current page of the split
> {code:java}
> /**
> * Verifies that named destinations (/Names /Dests) pointing to a page
> are copied to the split
> * document that contains that page, remapped to the imported page, and
> that destinations
> * pointing to other pages are not carried over.
> */
> @Test
> public void testSplitCopiesNamedDestinations() throws Exception {
> PDDocument document = new PDDocument();
> PDPage page0 = new PDPage();
> PDPage page1 = new PDPage();
> PDPage page2 = new PDPage();
> document.addPage(page0);
> document.addPage(page1);
> document.addPage(page2);
> PDPageFitDestination dest0 = new PDPageFitDestination();
> dest0.setPage(page0);
> PDPageFitDestination dest1 = new PDPageFitDestination();
> dest1.setPage(page1);
> PDPageFitDestination dest2 = new PDPageFitDestination();
> dest2.setPage(page2);
> Map<String, COSObjectable> destNames = new LinkedHashMap<String,
> COSObjectable>();
> destNames.put("dest0", dest0);
> destNames.put("dest1", dest1);
> destNames.put("dest2", dest2);
> PDDocumentNameDictionary names = new
> PDDocumentNameDictionary(document.getDocumentCatalog());
> PDDestinationNameTreeNode dests = new PDDestinationNameTreeNode();
> dests.setNames(destNames);
> names.setDests(dests);
> document.getDocumentCatalog().setNames(names);
> List<PDDocument> split = new Splitter().split(document);
> Assert.assertEquals(3, split.size());
> for (int i = 0; i < split.size(); i++) {
> PDDocument splitDocument = split.get(i);
> PDDocumentNameDictionary splitNames =
> splitDocument.getDocumentCatalog().getNames();
> Assert.assertNotNull("split document " + i + " should have a name
> dictionary", splitNames);
> Assert.assertNotNull("split document " + i + " should have a
> dests node", splitNames.getDests());
> Map<String, COSObjectable> copiedNames =
> splitNames.getDests().getNames();
> Assert.assertNotNull(copiedNames);
> // only the destination pointing to this page should be present
> Assert.assertEquals(1, copiedNames.size());
> String expectedName = "dest" + i;
> COSObjectable copied = copiedNames.get(expectedName);
> Assert.assertTrue(copied instanceof PDPageFitDestination);
> // the destination must point to the imported page of the split
> document
> PDPage importedPage = ((PDPageFitDestination) copied).getPage();
> Assert.assertSame(splitDocument.getPage(0).getCOSObject(),
> importedPage.getCOSObject());
> }
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
