Jump to content

Emby can't recognize episodes or files that have underscores in filename?


TheBasementNerd

Recommended Posts

TheBasementNerd

Most of my downloads for my anime has filenames with underscores, like A_Certain_Magical_Index_02 for example. The series is recognized just fine, and because the episodes are sorted into Season folders the seasons are fine, but with the filename like that it doesn't download any metadata for the episode itself. Once I replace the underscores with spaces it has no problem, but is there a way to have it work with underscores, or is there no way to fix this? Can it be added please?

Link to comment
Share on other sites

Hi there, can we look at a specific example in detail? It is not going to be underscores, but something else. Renaming it causes it to be treated like a new library item so that is most likely the explanation. Thanks.

Link to comment
Share on other sites

TheBasementNerd

Hi there, can we look at a specific example in detail? It is not going to be underscores, but something else. Renaming it causes it to be treated like a new library item so that is most likely the explanation. Thanks.

[DB]Haven't You Heard I'm Sakamoto_-_01_(Dual Audio_10bit_BD1080p_x265).mkv does not recognize the episode, doesn't download metadata

 

[DB]Haven't You Heard I'm Sakamoto - 01 (Dual Audio 10bit BD1080p x265).mkv does recognize the episode, all metadata downloaded

 

The only thing I did was load all the episodes into a renamer, and have it replace any "_" with a " ", so all underscores turned into a space. It took a second to download, but all episodes are accounted for, with descriptions and images, whereas before they had nothing but a screencap from the episode taken by Emby and the default filename, no other associated metadata

Edited by TheBasementNerd
Link to comment
Share on other sites

TheBasementNerd

Hi there, can we look at a specific example in detail? It is not going to be underscores, but something else. Renaming it causes it to be treated like a new library item so that is most likely the explanation. Thanks.

Another example here

Same server, all episodes followed the same naming scheme as the second episode on the right (The End of Evangelion is also named in its file as End of Evangelion, I know why that isn't matching to its episode already)

 

The only thing changed is I removed all the underscores, and it successfully downloaded the metadata for the first episode and placed it within Season 1, whereas the rest have no metadata as you can see

5c4b9849788e3_EvangelionScreen.png

Link to comment
Share on other sites

Happy2Play

I can reproduce this.  It would appear to be the hyphen surrounded by underscores, I removed the "-_" for episode to even appear.  But [DB]Haven't You Heard I'm Sakamoto_01_(Dual Audio 10bit BD1080p x265).mkv still didn't pull metadata.  Renamed [DB]Haven't You Heard I'm Sakamoto - 01 (Dual Audio 10bit BD1080p x265).mkv or [DB]Haven't You Heard I'm Sakamoto 01 (Dual Audio 10bit BD1080p x265).mkv pulled metadata and displayed. 

 

But this original naming scheme deviates from supported naming schemes, I would think underscore alone should have worked though.

 

https://github.com/MediaBrowser/Wiki/wiki/TV-naming#episode-naming-conventions

Edited by Happy2Play
Link to comment
Share on other sites

What folder structure, and how is the library setup in emby library setup? Because i tried that exact filename and it showed up for me. thanks.

Link to comment
Share on other sites

Deathsquirrel

Rather than wrestle with recognition issues for nonstandard file name formats, why not just use the auto organize plugin or something like filebot to rename your files in a way that avoids this entirely?

Link to comment
Share on other sites

TheBasementNerd

What folder structure, and how is the library setup in emby library setup? Because i tried that exact filename and it showed up for me. thanks.

F:\Anime\Haven't You Heard I'm Sakamoto\[DB]Haven't You Heard I'm Sakamoto - 01 (Dual Audio 10bit BD1080p x265).mkv

 

I have all these shows located in my Anime folder, which is what Emby points to for the Anime library. Next is the folder for the name of the show, correctly typed so that TheTVDB will recognize it, and then if it has multiple seasons or a specials season, I create a folder for Season 1 and onward, and then a season for specials either as Season 00 or Specials. For Sakumoto, I have it without multiple seasons

 

Sorry reply took a while, been busy

Link to comment
Share on other sites

TheBasementNerd

@@TheBasementNerd, can you try removing that series from your library, then run a library scan, and then add it back? thanks !

The series does currently load correctly after I did the change to the filenames as I said, what would be the reason for doing this?

Link to comment
Share on other sites

Well I was hoping to just test the originals by removing and adding them back. At this point we don't yet know if there is an ongoing issue or if it is isolated to the 3.5 to 4.0 upgrade. We are investigating though. Thanks.

Link to comment
Share on other sites

  • 2 years later...
Ascentw

Hi, I'm not fully sure about the currently situation about this behavior, but when I tested like 2 years ago it didn't recognize any filename where the underscores are substitutes for spaces and it still don't to this day, at least for what I've noticing. I'll try my best to explain both the examples of the behaviour and also give an answer to the other user on why auto organizing / using filebot is not a great solution to the case (I'm aware of how simple and fast filebot is, but the problem lays not in filebot, but mostly torrent clients)

Firstly, the examples.

image.thumb.png.4f9ed9b97efabe2db82cd7176206686b.png

 

Even though it recognizes the series (or maybe I did identify manually before), the problem is the "_" as substitutes for spaces. Until now, I'm pretty sure it always happens if the episode is numbered like "01", "02", "03", etc.

Second example: 
image.thumb.png.aa0a47af06b0476116d26e76035674dd.png

also pretty much the same thing. In this case the format was "tvshowname"  - "episodenumber", but again with "_" as spaces.

So going a bit further on recognition: If I go and just change the underscores to spaces using any file renamer, it will work and the episode will get the metadata and such.

This comes to the second problem: why don't simply renaming all of them then?
Because the problem isn't renaming files itself, but the way it works for torrent clients. A lot of fansubbed shows who haven't an official release in several languages are fansubbed and therefore posted in trackers or private trackers. From this perspective, when downloading an anime/tvshow that you want to keep seeding even after finishing download, you must have the file name mapped according to the names the torrent client, and if you are familiar with torrent clients you know they pretty much don't have a "find and replace" feature embed into their clients. I'm not saying it's impossible to do it so, but mostly either are completely manual (one by one renaming episodes) or some workaround messing with fastresume files like I come up to with qbittorrent, but it still not pratical.

I did write quite a lot, but it is not that niche as above. I mean, just take any torrent that you find on any torrent website. If the filename has underscores instead of spaces, mostly people would just put into filebot and that's it, but it means that you mostly probably not seeding any of the shows that happens to have that type of filename.

 

So, the conclusion and why I commented in the first place:

Is there a way to make the "scanner" of files to recognize the underscores as spaces when this happens and therefore avoiding a lot of renaming and problems as such?
or maybe, considering that messing with the scanner in this way would break something, make some kind of:

if the file name has no spaces from start to end
          then underscores=space

This is just a dumb example from someone who isn't familiar with coding and how both the scanner and metadata agent works, but it would be cool if this simple underscore thing could be detect normally just by the code being aware that they are supposed to have the same value as spaces or something around this idea.

Grateful, Ascent.

Edit: I think the correct term is underscore, not underline. I edit my text to fix that.

 

 

Edited by Ascentw
Link to comment
Share on other sites

My advise would be to not torrent download into your libraries but instead to a central location where you can properly process, convert, manipulate the files and rename them proper.

If using a headend program to control downloads like Sonarr/Radarr these programs can do that for you and rename the files proper before adding them into your libraries.  So it's just a matter of configuration and the programs do the work for you.

Link to comment
Share on other sites

Ascentw

I gotta be honest, I had to dig a bit into how sonarr/radarr works before replying again in this thread. I was aware of their existence, but never before felt like the need to use something of this sort. After some hours spent I had to say: it's quite great and pretty useful for renaming, tracking and some other stuff. Still, I don't think its either a solution or have something to do with this problem in particular, unless I'm missing something. 

For my few hours into it, I couldn't see how it changes the fact that you still have to rename the underscores and still have to somewhat break the seeding process. Sonarr will, as far as I'm aware, change the name os episodes like any other app would do, and unless you consider having 2 copies of the same thing (which defeats the whole purpose anyway), it would still break the torrent.

Sorry if it fells like I'm just complaining and not open to solutions, but it's just that I really like both emby itself and how great the webOS app really is, and I think some minor things like not being able to recognize eps just because of having underscores instead of spaces is something that, if possible, would be amazing to have a fix implemented on that. Like I said, I'm aware of how easy it's to rename with some apps like filebot or now even sonarr, but it still breaks the torrent. If I get a "no, it wouldn't be implemented because of 'x' reason", I would accept without any complies. I just want to make sure it's not a thing like that could be fixed in minutes by the emby developers before I go all the way into my method of renaming files without breaking torrents in qbittorrent, just because it takes a bit of time to me to do the renames with the fastresume files of qbittorrent and I feel like it would be a waste of time to do it for all the series if it gets implemented like in the next update or something.
 

Edited by Ascentw
Link to comment
Share on other sites

Not sure what you're trying to do but you don't want seeded content in your Emby libraries as that's not a good thing to do.

You want your Emby library media to be the curated content ready for streaming.

You may want to look into/understand NZB (and similar apps) as well as news groups which likely are far better than what you currently are using and have no ratios.
But the Emby forums aren't the place to discuss this type of thing. You can get info on this in both the R/S forums as they both support this.

Link to comment
Share on other sites

Ascentw

When I've come to this ideia of better recognizing episodes, I was partially aware that there wasn't a real big change I could do to just have it better for this specific case. The next paragraph will be an brief explanation onto my motives and how it's unlikely that there is better options for this type of situation, and feel free to skip reading it if necessary, since is not that relevant.

I am aware of much better websites for general torrent when it comes to english stuff, specially the mainstream content, but this is more of a niche case that before I tried to justify using more broad examples, but not the ones I mostly encounter in a daily basis. Going back to what I said before, when I referred to trackers and ratio, I didn't mean for content in english, but mostly for fansubs in my native language. Since it's both hard (cost) to host files in servers and also torrents are a lot easier to batch download seasons, there's a lot of old anime that never got any official subs into my language that people dedicate their time subtitling and stuff, and for that there are the trackers. So, there comes the whole point: it's not like I'm stuck in a ratio jail or something, at this point I have a pretty comfortable one in most trackers, and I would like to keep seeding so people interested in this fansubbed content can still acess their work, and also me when it comes to other stuff I've still didn't watch/download yet. It comes much more to a mentality thing than a "need" of seeding. It's not like the trackers make money over us, or either the fansub who subtitles the anime, it's pretty much purely for love of being able to giving an opportunity to watch this shows who mostly aren't available in their language and keep it seeding so their projects won't die easily.

Going back to the whole purpose of the discussion: this is why I've asked if there was any chance of the "scanner" (is it how it is called?) of files recognize the underscore as a word separator, because it would make it much easier to recognize episodes and would be a great help to me and my interest in keeping the file names as it is and also helping in more broad cases like the creator of this topic with the problems he showed us above.

Edited by Ascentw
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...