Hi,
I noticed that Informa (our RSS parser) loads DTDs for each feed. In =
plain
english that means that Informa loads RSS 0.91 DTD for each subscribed =
RSS
0.91 feed! This is an unnecessarily waste of system resources. The =
problem
with the Informa Parser classes is that the whole parser subsystem is
realized with static methods that makes extending difficult to =
impossible. I
got no reaction from Informa team for my email :(
In the moment I see two approaches:
* Use Informa and manipulate source
+ Already works with Rss Viewer
- Merge changes to Informa into our adapted source
* Switch to a pure RSS Parser Project
+ Less (unnecessary) code (Informa library comprises classes for
storage etc.)
+ Better performance (No DTD resoles) and (possibly) better
architecture
- Some work for adaption
- Testing with some feeds once more
I've already looked at some pure Parser projects:
Apache Commons FeedParser (in sandbox)
http://jakarta.apache.org/commons/sandbox/feedparser/index.html
Sandler
http://sourceforge.net/projects/sandler
Rome
http://wiki.java.net/bin/view/Javawsxml/Rome
RssLib4J
http://sourceforge.net/projects/rsslib4j/
RssLib J
http://enigmastation.com/rsslibj/
My personal favorit is Rome because of features and api documentation =
and
feed parser because of the apache commons reputation.
I'm awaiting your feedback.
Cheers
Christoph Walcher
|
