Have made significant progress yesterday and today in a merb app that takes a bunch of yamlfiles and generates my static story archive from them. The data makes a stop midway in an sql database, which I'm not in fact using for anything. Though I will; the intent is to have the yaml as archive, the sql as the working data copy, and re-generation of the site to happen on any data edit. rsync would then push the files live.
From this mouthful of jargon you'd never guess I'm a Pythonista in my private life. But sigh. The world of Ruby web app frameworks is too much better right now.
This is an exercise in exploring the problem space of a fic archive. My first observation so far is about how complex the data is, especially since I'm adding tags to the mess. (My second is that I've written way more than I'd thought: 59 files to process. Lord.) Storylines, multi-part stories within a larger series, banners, awards, commentary, ratings, pairings, warnings: just assembling the test data is taking some time. But if I invest the time in getting a large, complex body of test data to start with, I'll have fewer surprises later.