The main open source project I've been involved in is 4Suite, which provides Python facilities for XML processing. It's a venerable project that's been useful to many, but it's also showing it's age. Time for a reboot. In a joint project of my own company Zepheira and Cleveland Clinic where 4Suite is used in a patient database program. The result of this joint effort is called Akara. Part of Akara is Amara, the low-level XML interface. Progress on Amara has been pleasing of late, and I've started to arm folks up to the opportunities it presents with an e-mail article series called "Seven days of Amara". So far there are two installments out:
- Day 1: "XML output for the desperate hacker" - a neat approach to writing out XML
- Day 2: "Watering the trees of markup" - outline of Amara's tree APIs for XML and HTML processing
And while I'm at it I cooked up a little toy for Posterous combining techniques from the above. The following Python code generated a super-simple summary of my posts to date on Posterous:
-- %< --
from amara import bindery
from amara.writers.struct import *
feed = bindery.parse(PFEED) w = structwriter(indent=u"yes")
w.feed(
ROOT(
E(u'div',
( E(u'a', {u'href': unicode(it.link)}, unicode(it.title))
for it in feed.rss.channel.item ),
)))
-- %< -- In use:
-- %< --
$ python posterous.py
<?xml version="1.0" encoding="utf-8"?>
<div>
I go chop ya dollar
The Hollow Men
Wikimedia plus search
Wenger knows. You don't.
Physics in its essence is real (or is it?)
IOUNote redesign
Multi-process! Yes! (re: Google Chrome, BTW)
Interview with Cesc Fabregas in The Guardian
MoinMoin on Apache (using CGI) with multiple wikis
I'm not usually a sucker for the newest hotness...
</div>
-- %< --
This isn't really Posterous-specific at all. It's works on any RSS 2.0. I do wish posterous would support Atom instead of (or at least in addition to) RSS 2.0, but that's topic for another day.
I hope this gets the developers among you interested in my latest work. Get it hot off the plate, and free/libre! (Apache-style license).