JDOM and Parsing

Question

I have XML data in the form of a String as follows:
<root>
<section1>
... some data/tags
</section1>
<section2>
... some data/tags
</section2>
<section3>
... some data/tags
</section3>
</root>
What I would like to do is slightly tweak this XML to remove the whole of <section3>. I would like to use the latest JDOM-beta8 in order to do this. Once I remove this <section3> tag, then I am going to need the new XML String.
So, I guess I have a few questions:
1) How do I JDOM to read in a string to parse this data?
2) Could I use SAX or do I need to use DOM behind JDOM?
3) How do I remove this one node <section3> from the JDOM?
4) How do I get the String of this XML data back from JDOM?
Any help would be most appreciated. Thanks.
Tom

Answer 1

First, here is a snippet of what I do with JDOM to read/parse xml:
import org.jdom.*;
import org.jdom.input.*;
import org.jdom.output.*;
... some class definiton
private final SAXBuilder builder = new SAXBuilder(false);
private Document doc = null;
.. some method defintion
try
builder.setValidation(false); // seems to do nothing
doc = builder.build(new File("<path to xml file to read>"));
Element root = doc.getRootElement();
// see below for insertion that goes here
catch (JDOMException jdome)
jdome.printStackTrace(System.out);
Now, the above opens the xml file, gets you the root node. Now you are ready to do some parsing. In this case, you want to remove Section 3 from the XML and I assume spit the XML file back out? You say you need the new XML String, but not sure how you mean.
So, since Section 3 in your example is one node down, you need to do:
root.removeElement("Section3");
However, in the JDOM api it says that is one level deep, so I am not sure if it just removes the <Section3></Section3> tags, and not any children of that element. There is also
Element section3 = root.getChild("Element3");
if (null != section3)
root.removeContent(section3);
I have never done this before, but the removeContent says it removes the Element, so maybe it removes all children (nested elements) of this element? If not, you can just do something like:
Element section3 = root.getChild("Element3");
if (null != section3)
section3.removeChildren(); // removes all child elements
root.removeContent(section3);
One of those 3 should do the job, but to be honest I am not sure.
I think SAX or DOM would be fine. I think the thing with using SAX is, if you only have to go through the XML one time, it works fine and saves bit time on memory, is faster, etc. If you save the doc in memory for future use, you may need DOM to keep it in memory. Not for sure though. Watching this topic to see what others may say.
I am also not sure how to get the whole XML as a string, but there is a toString() method in the Element class. Since getRootElement() returns the root doc element, I assume toString() will print out the entire XML string. If you remove the Element first, then do a toString() on the root, you should get what you are looking for.