Trying out Pelican
I really like Jekyll, as I’ve mentioned before. Unfortunately it’s written in Ruby, and I really just don’t get on well with Ruby - and I’d like to extend the CMS so I can add features to my blog.
Hyde is a Python alternative to Jekyll, however the documentation is utterly terrible and it just doesn’t really seem to be maintained anymore. Today, I stumbled onto Pelican, and thought I’d give it a go.
So far out of the box it seems to do pretty much exactly what I want, even as far as helping to automate pushes to S3.
Update: It took a bit of smacking around to get it to work - the skinning is particularly easy because I’m very familiar with Jinja templates. However quite a few other things took some mucking about with:
- I had to chop the shit out of the import script I was using (once I work out what license it’s under I’ll include it in the sources of this site), mainly to cope with the georss objects, but also to include more granularity in the time keeping and to organize the articles into directories by month/year.
- Pelican claims to support ISO8601 dates in the metadata, yet it does no
such thing: it chokes when presented with a date taken directly from
strftime()took care of that bug though.
- I still have a lot of style fixes to go on the site, but it’s quite close. I’m rather excited - as this is the closest I’ve been in years to ready to move off of Blogger.