TWiki Spider

This is a new project I've just kicked off today. At the moment, it's probably of academic interest only, but for those who've been intrigued by the topic here's a quick overview of my plans:

1. Explore how much of TWiki's metadata can be easily accessed via HTTP as valid XHTML.
2. Write XSLT transforms that turn this "low-hanging fruit" into structured XML documents that can be more readily processed by add-on programs.
3. Define and publish XML schemas for these structured documents.
4. Write a Java class library that lets programs access TWiki as a living XML document repository.

To start out, I've cleaned up the TWikiUsers page on my internal TWiki so that it's clean, well-formed XML. All this required was properly quoting the href attributes on the hand-entered HTML index links. With this trivial barrier out of the way, the twiki-users.xslt that I've attached generates a nicely structured document from the standard registration page.

-- KaelinColclasure - 07 Jan 2003