Apache Poi - how to remove all the links from Word Documents

Question

I want to remove all the hyperlinks of a Word document and keep the text. I have these two methods to read word documents with doc and docx extensions.

private void readDocXExtensionDocument(){
    File inputFile = new File(inputFolderDir, "test.docx");
    try {
        XWPFDocument document = new XWPFDocument(OPCPackage.open(new   FileInputStream(inputFile)));
        XWPFWordExtractor extractor = new XWPFWordExtractor(document);
        extractor.setFetchHyperlinks(true);
        String context =  extractor.getText();
        System.out.println(context);
    } catch (InvalidFormatException e) {
        e.printStackTrace();
    } catch (FileNotFoundException e) {
        e.printStackTrace();
    } catch (IOException e) {
        e.printStackTrace();
    }

}

private void readDocExtensionDocument(){
    File inputFile = new File(inputFolderDir, "test.doc");
    POIFSFileSystem fs;
    try {
        fs = new POIFSFileSystem(new FileInputStream(inputFile));
        HWPFDocument document = new HWPFDocument(fs);
        WordExtractor wordExtractor = new WordExtractor(document);
        String[] paragraphs = wordExtractor.getParagraphText();
        System.out.println("Word document has " + paragraphs.length + " paragraphs");
        for(int i=0; i<paragraphs.length; i++){
            paragraphs[i] = paragraphs[i].replaceAll("\\cM?\r?\n", "");
            System.out.println(paragraphs[i]);
        }
    } catch (IOException e) {
        e.printStackTrace();
    }
}

Is it possible to remove all the links of a word document with using apache poi library? If it is not, are there any other libraries that can provide this?

Answer 1

My solution, at least for the .docx category, would be to use regular expressions. Check this one out

private void readDocXExtensionDocument(){
   Pattern p = Pattern.compile("\\<(.+?)\\>");
   File inputFile = new File(inputFolderDir, "test.docx");
   try {
      XWPFDocument document = new XWPFDocument(OPCPackage.open(new   FileInputStream(inputFile)));
      XWPFWordExtractor extractor = new XWPFWordExtractor(document);
      extractor.setFetchHyperlinks(true);
      String context =  extractor.getText();
      Matcher m = p.matcher(context);
      while (m.find()) {
         String link = m.group(0); // the bracketed part
         String textString = m.group(1); // the text of the link without the brackets
         context = context.replaceAll(link, ""); // ordering important.  Link then textString
         context = context.replaceAll(textString, "");
      }
      System.out.println(context);
   } catch (InvalidFormatException e) {
    e.printStackTrace();
   } catch (FileNotFoundException e) {
    e.printStackTrace();
   } catch (IOException e) {
    e.printStackTrace();
   }
  }

The only caveat to this approach is that if there is material with these angled brackets that is not a link, that too could be removed. If you have a better idea of what kind of links might appear, you might try a more specific regular expression instead of the one I provided.