import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
/**
 * <a href="https://stackoverflow.com/questions/45812696/pdfbox-delete-comment-maintain-strikethrough">
 * PDFBox delete comment maintain strikethrough
 * </a>
 * <br/>
 * <a href="https://expirebox.com/files/3d955e6df4ca5874c38dbf92fc43b5af.pdf">
 * only_fields.pdf
 * </a>
 * <a href="https://file.io/DTvqhC">
 * (alternative download)
 * </a>
 * <p>
 * The OP only wanted the comment removed, not the strike-through. Thus, we must
 * not remove the annotation but merely the comment building attributes.
 * </p>
 */
@Test
public void testRemoveLikeStephanImproved() throws IOException {
    final COSName POPUP = COSName.getPDFName("Popup");
    try (InputStream resource = getClass().getResourceAsStream("only_fields.pdf")) {
        PDDocument document = PDDocument.load(resource);
        List<PDAnnotation> annotations = new ArrayList<>();
        PDPageTree allPages = document.getDocumentCatalog().getPages();

        List<COSObjectable> objectsToRemove = new ArrayList<>();

        for (int i = 0; i < allPages.getCount(); i++) {
            PDPage page = allPages.get(i);
            annotations = page.getAnnotations();

            for (PDAnnotation annotation : annotations) {
                if ("StrikeOut".equals(annotation.getSubtype()))
                {
                    COSDictionary annotationDict = annotation.getCOSObject();
                    COSBase popup = annotationDict.getItem(POPUP);
                    annotationDict.removeItem(POPUP);
                    annotationDict.removeItem(COSName.CONTENTS); // plain text comment
                    annotationDict.removeItem(COSName.RC);       // rich text comment
                    annotationDict.removeItem(COSName.T);        // author

                    if (popup != null)
                        objectsToRemove.add(popup);
                }
            }

            annotations.removeAll(objectsToRemove);
        }

        document.save(new File(RESULT_FOLDER, "only_fields-removeImproved.pdf"));
    }
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
/**
 * This will process all of the pages and the text that is in them.
 *
 * @param pages The pages object in the document.
 *
 * @throws IOException If there is an error parsing the text.
 */
protected void processPages( List<COSObjectable> pages ) throws IOException
{
    if( startBookmark != null )
    {
        startBookmarkPageNumber = getPageNumber( startBookmark, pages );
    }
    if( endBookmark != null )
    {
        endBookmarkPageNumber = getPageNumber( endBookmark, pages );
    }

    if( startBookmarkPageNumber == -1 && startBookmark != null &&
            endBookmarkPageNumber == -1 && endBookmark != null &&
            startBookmark.getCOSObject() == endBookmark.getCOSObject() )
    {
        //this is a special case where both the start and end bookmark
        //are the same but point to nothing.  In this case
        //we will not extract any text.
        startBookmarkPageNumber = 0;
        endBookmarkPageNumber = 0;
    }
    Iterator<COSObjectable> pageIter = pages.iterator();
    while( pageIter.hasNext() )
    {
        PDPage nextPage = (PDPage)pageIter.next();
        PDStream contentStream = nextPage.getContents();
        currentPageNo++;
        if( contentStream != null )
        {
            COSStream contents = contentStream.getStream();
            processPage( nextPage, contents );
        }
    }
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
protected void processPages(List<COSObjectable> pages) throws IOException
{
    if (startBookmark != null)
    {
        startBookmarkPageNumber = getPageNumber(startBookmark, pages);
    }

    if (endBookmark != null)
    {
        endBookmarkPageNumber = getPageNumber(endBookmark, pages);
    }

    if (startBookmarkPageNumber == -1 && startBookmark != null &&
            endBookmarkPageNumber == -1 && endBookmark != null &&
            startBookmark.getCOSObject() == endBookmark.getCOSObject())
    {
        //this is a special case where both the start and end bookmark
        //are the same but point to nothing.  In this case
        //we will not extract any getText.
        startBookmarkPageNumber = 0;
        endBookmarkPageNumber = 0;
    }

    for (COSObjectable page : pages)
    {
        PDPage nextPage = (PDPage) page;
        PDStream contentStream = nextPage.getContents();
        currentPageNo++;
        if (contentStream != null)
        {
            COSStream contents = contentStream.getStream();
            processPage(nextPage, contents);
        }
    }
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
private <T extends COSObjectable> void sanitizeRecursiveNameTree(PDNameTreeNode<T> efTree, Consumer<T> callback) {
    if (efTree == null)
        return;

    Map<String, T> _names;
    try {
        _names = efTree.getNames();
    } catch (IOException e) {
        LOGGER.error("Error in sanitizeRecursiveNameTree", e);
        return;
    }

    if (_names != null) {
        _names.values().forEach(callback);
    }
    if (efTree.getKids() == null)
        return;
    for (PDNameTreeNode<T> node : efTree.getKids()) {
        sanitizeRecursiveNameTree(node, callback);
    }
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
private int getPageNumber( PDOutlineItem bookmark, List<COSObjectable> allPages ) throws IOException
{
    int pageNumber = -1;
    PDPage page = bookmark.findDestinationPage( document );
    if( page != null )
    {
        pageNumber = allPages.indexOf( page )+1;//use one based indexing
    }
    return pageNumber;
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
private void processEmbeddedDocNames(
  Map<String, COSObjectable> embeddedFileNames,
  EmbeddedDocumentExtractor embeddedExtractor) throws IOException,
  SAXException, TikaException {
  if (embeddedFileNames == null) {
    return;
  }
  for (Map.Entry<String, COSObjectable> ent : embeddedFileNames.entrySet()) {
    PDComplexFileSpecification spec = (PDComplexFileSpecification) ent
                                      .getValue();
    PDEmbeddedFile file = spec.getEmbeddedFile();

    Metadata metadata = new Metadata();
    // TODO: other metadata?
    metadata.set(Metadata.RESOURCE_NAME_KEY, ent.getKey());
    metadata.set(Metadata.CONTENT_TYPE, file.getSubtype());
    metadata.set(Metadata.CONTENT_LENGTH, Long.toString(file.getSize()));

    if (embeddedExtractor.shouldParseEmbedded(metadata)) {
      TikaInputStream stream = TikaInputStream.get(file.createInputStream());
      try {
        embeddedExtractor.parseEmbedded(stream, new EmbeddedContentHandler(
                                          handler), metadata, false);
      } finally {
        stream.close();
      }
    }
  }
}
 

import org.apache.pdfbox.pdmodel.common.COSObjectable; //导入依赖的package包/类
private int getPageNumber(PDOutlineItem bookmark, List<COSObjectable> allPages) throws IOException
{
    int pageNumber = -1;
    PDPage page = bookmark.findDestinationPage(document);
    if (page != null)
    {
        pageNumber = allPages.indexOf(page) + 1;//use one based indexing
    }
    return pageNumber;
}