Jump to content

Make collections path independent


levander

Recommended Posts

Happy2Play

Would using hashes work?

Makes sense, but, actually, this is a really tough one because there isn't really any other way to uniquely identify something other than its fully qualified path.  You might think hash but one, there is a performance impact to that and two that wouldn't survive any change you made to the file either.

Edited by Happy2Play
Link to comment
Share on other sites

amyhimesama

Please make collections independent of it’s path so relocating or renaming media in folder structures doesn’t affect collections and its items.

 

I often update media files, it’s folder name or/and folder structure for various reasons and every time I do that those items affected disappears from collections I put them in. 

 

 

Lucky Patcher Kodi nox

Collections items would have to be tracked by a universal id (imdb, tmdb, tvdb id), but if one of those don't exist then you are back to where we started, no way to track.

Edited by amyhimesama
Link to comment
Share on other sites

Collections items would have to be tracked by a universal id (imdb, tmdb, tvdb id), but if one of those don't exist then you are back to where we started, no way to track.

 

And that would force ALL instances of that item into the collection which you may not want.

Link to comment
Share on other sites

funwithmedia

Hadn't there been a thought/conversation once upon a time about assigning a unique ID to an item and storing that ID in the NFO? I know it's not a perfect solution, but it seems that there aren't any perfect solutions for this sort of thing. At least if you keep your files and NFO's together this method would allow Collections, Watched Status (eg, for Extras), etc. to persist if files (with accompanying NFO's) are moved.

  • Like 2
Link to comment
Share on other sites

  • 1 month later...
levander

Nfo? once it is put into collections, item’s nfo could store that data (or emby library db). It is same thing as with other custom data like tags. Tags are not affected when scanning after moving/renaming. 

 

 

Theoretically... but managing that and making sure it is all still valid and accurate could very quickly become a difficult task.

 

Plus, not everyone saves to NFOs (in fact, by default, we don't).

 

 

1. at the moment managing custom collections and making sure that the path is still valid is by far more difficult and making custom collections is discouraging.

2. by default everything that is in NFO is also stored in library db, so my proposition is not NFO exclusive

3. if imdb id is used it would still be good compromise until smart view lists is introduced.

Link to comment
Share on other sites

3. if imdb id is used it would still be good compromise until smart view lists is introduced.

 

We already discussed why that isn't really a good compromise...

 

 

And that would force ALL instances of that item into the collection which you may not want.

Link to comment
Share on other sites

levander

Ok, but is there still something you think you may do about it or did you gave up on this request because smart views will solve it one day? 

Link to comment
Share on other sites

crusher11

How are multiple instances handled within collections at present? If they're all linked together how can one be in a collection but another not be?

Link to comment
Share on other sites

levander

How are multiple instances handled within collections at present? If they're all linked together how can one be in a collection but another not be?

 

By path. All items in collections.xml are paths. 

Edited by levander
Link to comment
Share on other sites

crusher11

Right, but if I have the 4K version in a collection but not the 1080p version, for example, but they're done as a multi-version thing...they'd both show up in the collection anyway, surely?

Link to comment
Share on other sites

levander

Every separate media file can be shown separately in collection. That is what ebr tries not to compromise. That wouldn’t be possible if imdb id would be used instead of paths. 

Link to comment
Share on other sites

It is possible to have different versions NOT shown as one item.  Also, everything that can be in a collection doesn't have an IMDb (or other) ID.  

Link to comment
Share on other sites

  • 2 weeks later...
242Movies

Another vote for including collections info in .NFO files.

 

I have collections like "Bahamas (filmed in the Bahamas)", "American Film Institute Winners", "Golden Globe Winners", "Marvel Cinematic Universe", "Bud Spencer and Terrence Hill Movies", et cetera.

 

It would take a very long time to rebuild those collections if I move the files to a new hard drive or if I make a copy of my drive for somebody else.

 

If collection info is stored in the NFO file it could rebuild those collections on the next library scan.

 

Imagine creating a collection of "1001 Movies You Must See Before You Die" and then having to create it all over again when you move the drive to a new PC.

 

If duplicate files is a problem ffmpeg or a similar program can extract the Duration and bitrate from the files so Director's Cut would be a different file than Theatrical and 4k would would be different than 1080p.

No need to do a CPU-intensive hash.

 

EDIT:

An alternative method might be allowing users to share collection lists in .csv or similar files which contain IMDB or themoviedb IDs and emby can scan the user's share and populate any matched items in collections.

Edited by 242Movies
Link to comment
Share on other sites

levander

Another vote for including collections info in .NFO files.

 

I have collections like "Bahamas (filmed in the Bahamas)", "American Film Institute Winners", "Golden Globe Winners", "Marvel Cinematic Universe", "Bud Spencer and Terrence Hill Movies", et cetera.

 

It would take a very long time to rebuild those collections if I move the files to a new hard drive or if I make a copy of my drive for somebody else.

 

If collection info is stored in the NFO file it could rebuild those collections on the next library scan.

 

Imagine creating a collection of "1001 Movies You Must See Before You Die" and then having to create it all over again when you move the drive to a new PC.

 

If duplicate files is a problem ffmpeg or a similar program can extract the Duration and bitrate from the files so Director's Cut would be a different file than Theatrical and 4k would would be different than 1080p.

No need to do a CPU-intensive hash.

 

EDIT:

An alternative method might be allowing users to share collection lists in .csv or similar files which contain IMDB or themoviedb IDs and emby can scan the user's share and populate any matched items in collections.

 

Just to be clear there is SET field in nfos that usually allowed a movie to belong to ONLY one set at a time in for example KODI. And that should avoided of course. But I’m not sure was that KODI only rule or is there something else...

Link to comment
Share on other sites

242Movies

I would be happy with a secondary nfo file called moviename.ct.nfo (collections and tags?)

 

Right now I'm holding off on some of the more time-consuming collections work because I know that the work would only be temporary instead of persistent.

Link to comment
Share on other sites

levander

 

Right now I'm holding off on some of the more time-consuming collections work because I know that the work would only be temporary instead of persistent.

 

I know...

Link to comment
Share on other sites

crusher11

I have way more free time than I do hard drive space, so I've been putting together stubs of everything, making collections, etc. Every time I create an actual movie file, I have to make sure it ends up in all the right collections. It's a right pain.

Link to comment
Share on other sites

242Movies

I used the metadata manager to create a genre called Bahamas and added some movies to it.

 

The server is configured to save nfo files to the media folder and the nfo file contains the genre.

 

When I moved the hard drive to a new server it did find and create the Bahamas genre and added the movies that I had configured with that genre.

 

I can use the same method for Best Picture Winners or any other thing I needed collections to do.

 

This somewhat solves my problem of having to create everything from scratch.

 

Is there a way to add a genre to a collection or to favorites?

Edited by 242Movies
Link to comment
Share on other sites

242Movies

Thank you for the response.

 

Is there any other way to accomplish what i need without having the 'collection' buried?

 

It would be nice to have something on the home screen or in collections or favorites and have it repopulate when I move the drive to a new server.

Link to comment
Share on other sites

242Movies

Okay. Thanks.

I can work with genres for now.

 

TheMovieDB mentioned they will be adding a 'Universe' item to collections that may be useful within Emby.

Link to comment
Share on other sites

levander

I would say smart views/playlists once we add support for those.

 

Is there something special you are waiting for before you can add support for those?

Link to comment
Share on other sites

Is there something special you are waiting for before you can add support for those?

 

Just a few extra hours in the day :).  It will be a huge task.

  • Like 1
Link to comment
Share on other sites

aaronsomek

Makes sense, but, actually, this is a really tough one because there isn't really any other way to uniquely identify something other than its fully qualified path.  You might think hash but one, there is a performance impact to that and two that wouldn't survive any change you made to the file either. 

Could you use the Title of the media to ensure it always goes into a collection? Maybe the Title plus Year to provide better accuracy. This way if I update The Omega Man (1971).avi to a better quality version the library just looks at TitleName.xxx.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...