It’s not really a well-kept secret that the search in Jellyfin needs a lot of work. It’s slow, doesn’t deal with typos and commas correctly and doesn’t allow searching multiple fields at once.
I made a quick and dirty proxy to enable a proper full-text search in Jellyfin while the dev team is working on the EFCore migration. It’s not perfect but it’s much better than what Jellyfin currently provides.
If you are running Jellyfin inside of Docker and use a Traefik reverse proxy, check out the image/repo below.
If you know what you’re doing (this is Lemmy after all), the proxy is a simple ASP.NET application and works with pretty much every reverse proxy once configured.
https://gitlab.com/DomiStyle/jellysearch
https://hub.docker.com/r/domistyle/jellysearch
If you tested with any Jellyfin client not in the README, feel free to let me know. If you used any other reverse proxy than Traefik, also let me know.
Looks and sounds cool.
FYI, you may need to change the logo. I didn’t think 3rd party tools are allowed to use the official logo, but someone who knows more about that will need to weigh in.
FYI, you may need to change the logo. I didn’t think 3rd party tools are allowed to use the official logo, but someone who knows more about that will need to weigh in.
Good point, I don’t mind changing it if it’s against their guidelines. It’s more of a crutch until full-text search arrives in Jellyfin proper, no need for a fancy logo either way. :)
Core team here. This is alright as there’s already precedent (see Findroid). You can always have fun with more/different colours though!
Really just need to be able to integrate elastic search like wiki.js allows
This integrates Meilisearch but without having to mess with the Jellyfin search code while Jellyfin is transitioning to EFCore.
Oh man, you read my mind: https://github.com/jellyfin/jellyfin/pull/7663#issuecomment-1758314971. Thanks for jumping on this!
