| | 203 | |
| | 204 | == Thoughts on Improvements == |
| | 205 | I know this thing has a lot of shortcomings, but I think it is well on its way to being what I want. Here are some of the things I have in mind at the moment. |
| | 206 | |
| | 207 | * There should be a web interface for manipulating templates, files, and fetching. Editing the DB contents by hand is far from ideal, and a web interface could show results very clearly. |
| | 208 | * Pages that change haven't been accounted for. For example, the page for the latest comic might not have a ''next'' link until the next comic is available. The fetcher needs to know to re-fetch that page in those circumstances. |
| | 209 | * Comic images should be fetched and referenced locally. |
| | 210 | * The fetcher should be rate limited. I am currently running it only periodically in a way that it only fetches one or two pages per site, but something should be built in so that it doesn't hammer sites. |
| | 211 | * The scripts should have a common configuration instead of hardcoded DB connection data in each one. |
| | 212 | * This needs to be tested with many more comics. |
| | 213 | * This needs to be tested with something that is entirely unlike comics. |
| | 214 | * genshi for templating works great for this specific case, but it might be preferable to allow for user-defined templates, which might require a sandboxable template system. |
| | 215 | * I should learn how badly I'm butchering RDF concepts. |
| | 216 | * Document templates need to be decoupled from the program which generates documents from them. |
| | 217 | |
| | 218 | [[AddComment]] |
