Identification issues

[ototos]

Dear Staff,
The situation seems to worsen every year. Whenever I attempt to switch to Jellyfin, I face insurmountable difficulties, particularly with identification issues.

This applies to TV shows, but it’s even worse with movies. For users like me with large libraries, reorganizing folder structures is out of the question. Jellyfin must achieve at least 99% compatibility with Plex or, ideally, surpass its efficiency.

The attached screenshots demonstrate how Jellyfin fails to merge content into a single entry when folders span multiple drives. With Naruto, for instance, it created a mess of duplicates despite identical files. It lacks intelligence: if episode 1x01 already exists, why create another? It should automatically recognize it as part of the same series.

A native solution is urgent—without relying on broken third-party tools like MergeVersions. Jellyfin must handle these cases flawlessly.

Unfortunately, for the fourth year in a row, I’m forced to abandon Jellyfin. Restructuring over 100TB of media (2000+ TV series and 4000+ movies) is simply impossible.

Moreover, as highlighted across various Discord and Telegram chats, as well as in Reddit threads and forum posts here, many of us are experiencing this issue - which could be fixed with a relatively straightforward solution. Please make this a priority!

I hope you’ll address this. Thank you.

[TheDreadPirate]

The file name parser is a nightmare-ish piece of code.  If it is simple, please submit a PR to fix it.

There are many many tools available to automate renaming your files and folders.  Two things I see are the use of non-english words for "Season", which is a known stumbling point, and no providing both the season number and episode number in the file name.

An example of a compliant structure.

Code:

Gunsmith Cats/
└── Season 01
    ├── s01e01.mkv
    ├── s01e02.mkv
    └── s01e03.mkv

Including the show name and episode name in the file name are acceptable.  I just opted to not include them.

Regarding duplicates, there is a setting you must enable to merge shows across multiple drives.  It is not enabled by default.

   

[ototos]

The file name parser is a nightmare-ish piece of code.  If it is simple, please submit a PR to fix it.

There are many many tools available to automate renaming your files and folders.  Two things I see are the use of non-english words for "Season", which is a known stumbling point, and no providing both the season number and episode number in the file name.

An example of a compliant structure.

Code:

Gunsmith Cats/
└── Season 01
    ├── s01e01.mkv
    ├── s01e02.mkv
    └── s01e03.mkv

Including the show name and episode name in the file name are acceptable.  I just opted to not include them.

Regarding duplicates, there is a setting you must enable to merge shows across multiple drives.  It is not enabled by default.

All my releases have always been renamed exactly as shown in the screenshot.
If even that isn’t good enough, then we’re really in trouble...
It shouldn’t matter which language the library is set to—not all Jellyfin users speak English. It’s possible to have a series with multiple language entries, but Jellyfin should be smart enough to recognize that it’s the same TV show.
I’ve enabled the Merge TV series option—I noticed it was disabled, even though I was sure I had turned it on before. No big deal, though... Right now, I’m scanning another 4TB of data.
Let’s hope for the best, because for the most important series, I always include both 1080p and 2160p versions. Sometimes, due to automation systems or upload timing, the titles might not be properly translated.

[ototos]

How can the identifier be this unreliable?
To start with, about 70% of my content (movies) looks like in the attached screenshot. On top of that, the posters and cover art are completely wrong. What’s Fast and Furious even doing there? It has nothing to do with my file.

My file follows this path:
Film\2016\Sully (2016)\Sully (2016) - [1080p h264 9,90 G].mkv

It’s named in a foolproof way. At least Plex doesn’t have issues with it—and never has.
This absolutely needs to be improved. For anyone considering switching to this software, it’s unthinkable to have to rename everything or start from scratch.
I hope a major fix is rolled out soon and that this isn’t just ignored in favor of other updates.


Thanks.

[TheDreadPirate]

What metadata providers do you have enabled in your library. There are separate lists for text metadata and image providers.

I have never had issues like you're describing since I started ensuring my library structure and naming is compliant with Jellyfin's documentation.

One thing I see with Sully is that you have the movie in a "2016" folder. Is "2016" added as a library root folder? Or do you only have "Film" as a library root? Nesting movies is not supported. You'd have to add "2016" as a library root or move all your movies up to the "Film" folder.

[ototos]

What metadata providers do you have enabled in your library.  There are separate lists for text metadata and image providers.

I have never had issues like you're describing since I started ensuring my library structure and naming is compliant with Jellyfin's documentation.

One thing I see with Sully is that you have the movie in a "2016" folder.  Is "2016" added as a library root folder?  Or do you only have "Film" as a library root?  Nesting movies is not supported.  You'd have to add "2016" as a library root or move all your movies up to the "Film" folder.

I have TMDB and TVDB enabled (all available providers).

I’m considering switching from Plex to Jellyfin, but with Plex, over the last 5 years, I’ve never had any issues with my folder structure. As mentioned in previous posts, converting over 100TB of content is unthinkable—we’re talking about a massive amount of data. Moreover, dumping all movies into a single "Film" folder would be ridiculous: it would make everything harder to manage and slow down directory loading.

That’s why I prefer organizing movies by year—it keeps things at least somewhat orderly.

And I haven’t even started properly with TV series yet, which come in multiple resolutions and are spread across different directories...

[TheDreadPirate]

Plex is Plex, and Jellyfin is Jellyfin. A lot of aspects of Jellyfin's scanner are inherited from Emby when we forked. Not many have been brave enough to attempt to fix the scanner.

I've stated how I think you could ensure consistent identification. We also have documentation here.

https://jellyfin.org/docs/general/server/media/movies/
https://jellyfin.org/docs/general/server/media/shows/

If that is too much of an administrative burden, well......I highly doubt the Jellyfin scanner is going to be changed anytime soon. If the devs knew how to improve it they would have done so already. But my cursory glance has determined that the way the devs have described it to me, "a nightmare piece of code", is an apt description.

[ototos]

Plex is Plex, and Jellyfin is Jellyfin.  A lot of aspects of Jellyfin's scanner are inherited from Emby when we forked.  Not many have been brave enough to attempt to fix the scanner.

I've stated how I think you could ensure consistent identification.  We also have documentation here.

https://jellyfin.org/docs/general/server/media/movies/
https://jellyfin.org/docs/general/server/media/shows/

If that is too much of an administrative burden, well......I highly doubt the Jellyfin scanner is going to be changed anytime soon.  If the devs knew how to improve it they would have done so already.  But my cursory glance has determined that the way the devs have described it to me, "a nightmare piece of code", is an apt description.

Excuse me, but I read online that a few years ago it was possible to manually modify the regular expressions (regex) for content identification. Is this information correct or inaccurate?

If it were correct, I wonder why this functionality was removed. Perhaps it could have solved my problem, and I believe it might be useful for many other users as well.

I understand that taking existing code and reworking it is complex. This is probably one of those cases where certain parts would need to be completely rewritten to create a more robust and high-performing system - potentially even superior to the famous Plex.

Unfortunately, Jellyfin is the newcomer to the market, and those who already have a media server with many imported contents might not consider it as an option, ending up choosing Emby or sticking with Plex. Currently, Jellyfin remains the best choice mainly for those starting from scratch. :-(

As far as you know, are they actively working on this? Or is it currently on hold, waiting for other higher-priority tasks to be completed?

[ototos]

TWO WHOLE DAYS spent manually fixing unrecognized content... and today when I checked again, everything was back to square one! So frustrating! I wasted four hours of my life for nothing!

[TheDreadPirate]

Excuse me, but I read online that a few years ago it was possible to manually modify the regular expressions (regex) for content identification. Is this information correct or inaccurate?

If it were correct, I wonder why this functionality was removed. Perhaps it could have solved my problem, and I believe it might be useful for many other users as well.

If this was ever a thing, it hasn't been since I've been using Jellyfin (Feb 2023 with 10.8.10).

As far as you know, are they actively working on this? Or is it currently on hold, waiting for other higher-priority tasks to be completed?

Not that I'm aware of.  Currently the biggest efforts are the EFCore library database migration, allowing external database providers, and HDR to HDR transcoding.

---

TWO WHOLE DAYS spent manually fixing unrecognized content... and today when I checked again, everything was back to square one! So frustrating! I wasted four hours of my life for nothing!

Do you have NFOs in your library? The only two circumstances that would cause manually changed metadata to revert are having NFOs in your library or "replace all metadata" library scans.