I've got a news page which is being sent out as an email to our organisation from a java application (which I'm coding). But we don't want the top and left navigation divs.

My boss muttered something about using an XSL transform to select the divs we do want, but I reckon he thinks I'm starting with a load of XML data, (which I'm not, the app is dealing with existing html code). So does applying a transform to an html page to produce another html page make sense? Or do I have to do a load of string manipulation in Java (potentially messy). Or maybe there's a better way?

Thanks in advance!
matt

XSL cannot be applied to HTML.
It can only work on XML, but can output HTML.

I have no past experience in Java, but try to look for a DOM parser.
It allows you to instantiate a tree of the page, and then you should be able to exrtact what is useful for you, and re-create a simple html with an XSL processor.
Here is a PHP dom parser reference:
http://www.php.net/manual/en/domdocument.loadhtml.php

PS:
I just found this via google:
http://java.sun.com/j2se/1.5.0/docs/.../Document.html

Ah, thanks tripy. A Dom passer sounds more appropriate for this

Originally Posted by tripy

XSL cannot be applied to HTML.

But, it can be applied to XHTML

Originally Posted by tripy

I have no past experience in Java, but try to look for a DOM parser.
It allows you to instantiate a tree of the page, and then you should be able to exrtact what is useful for you, and re-create a simple html with an XSL processor.

Since version 1.4 all standard java installs come with both a parser and xsl transformer.

For HTML, one option could be to use jTidy to get a DOM object. This can be fed to a transformer, manipulated or just written out.

For your particular case, try jTidy to create a DOM, then remove the nodes corresponding to the divs you don't want, then just output the DOM object as XHTML.

Of note, jTidy is also pretty good at dealing with bad markup.