We thought it'd be a good idea to auto-update the www.apache.org homepage
based on the foundation blog.

See the main README for information about how to run the tool (hint: there's
some snippets in ../build.xml for it).

This implementation uses Rome (which is licensed under Apache License v2.0) to
fetch and parse the feed, and then uses JDOM to look for the

  <section id="news">...</section>

part of index.xml. It then replaces the contents of that section with the
latest entries from the feed.

Since Anakia requires more-or-less valid XML input, there's some blog content
that we just can't really deal with properly given the chosen tools. We make a
reasonable attempt to do a reasonable thing and skip over the mess. Using the
python Universal FeedParser could've helped, but I chose not to do that because
then you'd need python installed properly in order to run the rebuild.

The additional copy of jdom and xerces in ./lib is needed to run Rome; these
libraries are _not_ compatible with the ancient ones that Anakia uses.

The generated classes are all checked in; remember to check in your changes too
if you make any! You can run

  ./build.sh

(yes, really! No ant required :)) to re-generate the classes after changing the
source code.