I had a very weird experience with Jellyfins team.
Recently I decided to give the app a try, and it worked fine, but I have serious issues with their media scanner (nested structures, a lot of misc content like interviews, etc.).
Years ago I wrote my own script that scans my drives, downloads imdb database, and 100% accurately matches media to imdb-id and fetches all auxilary data like cast, synopsis, posters, and so on.
I asked on their forum if I could use that to somehow bypass the scanner, prepopulate the database or write my own scanner based on theirs.
I was repedatley said it would be impossible and to not touch the code. "Seriously, do not do this" said one of the team members.
I mean I do understand they were trying to save me a lot of headache, but at the same time they're running a campaign for developers.
I just found it very odd.
The reason is that, it's basically impossible. The issue comes down to the Library handling code. It's legacy Emby code that we've been slowly but surely trying to replace but it's a huge task. This is huge chunk of Jellyfin. Writing your own parser is a great idea, but directly hooking it into Jellyfin will just be giving yourself a lot of headaches and will NOT be easy to do.
You're best to write out NFOs and then let Jellyfin just import those with all external metadata providers removed.
I can't speak for their developer/support team, but that kind of feedback indeed feels strange. Even more so because this kind of intervention is possible without touching any of Jellyfin's code.
In particular you can create an NFO file (same stem, next to the media, with a .info extension) that contains this data. With stuff like the IMDB ID given, Jellyfin is pretty good at extracting metadata from there; without it, false matches aren't uncommon. It's even documented. [1]
Recently I decided to give the app a try, and it worked fine, but I have serious issues with their media scanner (nested structures, a lot of misc content like interviews, etc.).
Years ago I wrote my own script that scans my drives, downloads imdb database, and 100% accurately matches media to imdb-id and fetches all auxilary data like cast, synopsis, posters, and so on.
I asked on their forum if I could use that to somehow bypass the scanner, prepopulate the database or write my own scanner based on theirs.
I was repedatley said it would be impossible and to not touch the code. "Seriously, do not do this" said one of the team members. I mean I do understand they were trying to save me a lot of headache, but at the same time they're running a campaign for developers. I just found it very odd.