- cross-posted to:
- torrent_trackers@lemmy.ml
- cross-posted to:
- torrent_trackers@lemmy.ml
The user can make a instance very easily and the design is pretty great. The software only holds torrents with at least one seeder. Codeberg git.
The user can make a instance very easily and the design is pretty great. The software only holds torrents with at least one seeder. Codeberg git.
What was the thought behind using a big flat csv file instead of json or a sql server?
Didn’t use json because it’s not as compact a data store In a csv, you only name the columns once, in json you have to name every field.
Sql isn’t a very good for store for git, unless you have a migration system which should really only be used for data structure, not actual data. This does build an sqlite file from the csv so the server can do smart paging and such.
@dessalines@lemmy.ml How big does the csv file get?
Its currently
78 MB. That’s a list of pretty much every torrent that has at least 1 seeder.