audiodude / best-albums Goto Github PK

Instead of #3, I think it would be better to just combine the scripts. So running bester would ingest the original format from the dump file and then go on to process additional new albums.

The slug should be the filename minus ".md". Although these slugs are currently unused, it would be good to have them for the future in case we need to unambiguously refer to a specific album.

Currently I get the album art from Amazon, which I believe is legal because I'm using the amazon affiliate program. However, I am manually uploading it to imgur, in order to avoid hotlinking what might be a volatile URL structure on Amazon. It would be nice if I could just specify "album_art_url" and have the bester script sideload the image to imgur and generate photo_url_sm and photo_url_lg.

This could be useful, even though I've already written a script to parse the old JSON and output the albums.json format. Consider that with something like this, the entire site could be generated by one script, which processes old and new files alike. Currently, if the albums.json file is messed up somehow, you have to regenerate the original albums.json, and then add each "new" album in the right order so that the ordering isn't changed on the site. This is very fragile and error prone.

On the other hand, there are potentially other ways to make the generation process idempotent without having to backport old albums.

This could be coupled with history manipulation possibly, so that clicking the back button shows the last album that was expanded.

Not sure if this timestamp should be the time when the album was added to albums.json, or the mtime of the file. Probably the latter.

Again, this isn't currently used in the interface, but it would be nice to have it in case it's needed one day.

This is part of making the script idempotent.