02:47: Hello, PyBlosxom

I switched the weblogs from Blosxom to PyBlosxom, today.

Now I guess I can get started on integrating Bazaar--I just need to read-up on (and read through) the architecture a little more.

I spent a while trying to figure out how to get the effect of Blosxom's `ISP plugin', until I realised that, at least when running in CGI mode, I can just query the CGI environment-variables in config.py and set datadir, flavourdir, base_url, and even blog_title accordingly.

I don't know how to handle this in WSGI mode (since I haven't really learnt anything about WSGI yet), but it's not clear that CGI is expensive enough for me to care, at this point.

Aside from that, the only real issue with the conversion was that PyBlosxom groups entries made on the same date, and Blosxom doesn't; so, timestamps work differently--I had to move the VisualIDs for my entries out of the timestamp-column and into the `story' header, and we still need to update Pam's templates so that the time is similarly moved (from being attached to the date, to being attached to the story per se).

Oh--we also (finally) got a way of walking back through historical posts, beyond the `front-page limit' (num_entries), courtesy of PyBlosxom's wbgpager plugin.

[Reply]