Parse HTML with jsoup and preserve original content

Question

I want to replace some elements in HTML files, keeping all the other content unchanged.

Document doc = Jsoup.parse("<div id=title>Old</div >\n" +
        "<p>1<p>2\n" +
        "<table><tr><td>1</td></tr></table>");
doc.getElementById("title").text("New");
System.out.println(doc.toString());

I expect to have the following output:

<div id=title>New</span></div >
<p>1<p>2
<table><tr><td>1</td></tr></table>

Instead, I have:

<html>
 <head></head>
 <body>
  <div id="title">New</div>
  <p>1</p>
  <p>2 </p>
  <table>
   <tbody>
    <tr>
     <td>1</td>
    </tr>
   </tbody>
  </table>
 </body>
</html>

Jsoup added:

closing p tags
double-quotes to attribute values
tbody
html, head and body elements

Can I serialise modified HTML back to original? Jericho does that but it doesn’t provide slick DOM manipulation methods as Jsoup does.

You could do it if JSoup would give you the character offsets of the elements. See http://stackoverflow.com/questions/11387458/get-character-offsets-for-elements-in-jsoup. We need this feature. Unfortunately, JSoup's author is no longer responding to questions posted to the jsoup google group. — ccleve, Aug 26 '12 at 19:53

score 1 · Answer 1 · answered Aug 27 '12 at 20:40

Is there a reason why attribute values shouldn't get quoted? See here and here.

For the other points try this:

final String html = "<div id=title>Old</div >\n"
            + "<p>1<p>2\n"
            + "<table><tr><td>1</td></tr></table>";

Document doc = Jsoup.parse(html);
doc.select("[id=title]").first().text("New");
doc.select("body, head, html, tbody").unwrap();
doc.outputSettings().prettyPrint(false);

System.out.println(doc);

Parse HTML with jsoup and preserve original content

1 Answers1