Phillip Pearson - web + electronics notes

tech notes and web hackery from a new zealander who was vaguely useful on the web back in 2002 (see: python community server, the blogging ecosystem, the new zealand coffee review, the internet topic exchange).

2003-4-27

TrackBack in Radio and community servers

Dave Winer writes that inbound TrackBack in Radio would require server-side support.

Georg Bauer implemented the server side for this for the Python Community Server project a while back ... it works in a similar way to the comment system. Not many people are using it right now, but here's one blog that I could find with it enabled.

He added an 'flCanHostTrackback' member to the 'community' struct returned by xmlStorageSystem.getServerCapabilities, which returns a boolean true value if trackback is supported.

Also radioCommunityServer.getInitialResources now returns a 'trackbackPageUrl' member that returns the URL of the script that implements trackback. It works the same way as the comment server, e.g.:

 http://www.pycs.net/system/trackback.py?u=0000081&p=P37

That displays the ping URL and a list of all trackbacks.

It also generates a JavaScript summary of trackbacks in the same way that the comment script generates a comment summary. That way, you can show trackback counts by including the summary:

 <script src="http://www.pycs.net/system/trackback.py?u=0000081&c=counts" type="text/javascript">

and then displaying individual counts like this:

 trackback [<script type="text/javascript" language="JavaScript">trackbackCounter('P37')</script>]

It would be cool if it were implemented the same way in RCS. Note that the PyCS license allows you to use the code pretty much how you like, so anyone doing Radio server stuff is welcome to take the trackback.py script and port it to UserTalk to get a quick solution!

Update: See also Phil Ringnalda: Quibbling about TrackBack
... more like this: [, , , ]

Cutlery

Charles Miller: "You know what? I own a knife and fork. In fact, I own several!"

It'd just been so long since I'd used them that I'd forgotten...

... more like this: []

Milestone

(topic: search engine)

My Java web app now responds to XML-RPC requests, verifies user logins from a table in an HSQL database, verifies blog ownership from another table, finds URL grouping from a third table, and adds / updates records in a fourth one.

Time to hook in Lucene and get it searching. I wonder if the indexer is quick enough to run inside the XML-RPC handler. Right now it's handling about 50 hits/sec, running in debug mode under Eclipse (Eclipse, Tomcat and the Python test script are taking about 1/3 of the CPU each), which isn't that bad, especially compared to the Tomcat benchmarks I've seen so far.

Update: It'll be nice when I can switch to running HSQLDB in-process, because I seem to be hitting this Windows box's socket limit rather easily right now. Or will servlets do connection pooling by themselves?
... more like this: [, ]