Hi,
This may have been mentionned in the past (https://forums.plex.tv/discussion/78326 , https://forums.plex.tv/discussion/23649#entry488762), but while there’s a lookup for the ID in nfo file next to the movie, there’s no check for it in the filename (at least in /plexmediaserver/Resources/Plug-ins-a17e99e/Scanners.bundle/Contents/Resources/Movies/Plex Movie Scanner.py, but I guess there’s no other place to check).
I use this convention for naming my files :
\Movies\Title 1 (year) [tt1234567].ext \Movies\Title 2 (year) [tt1234568].ext \Series\Title (year) [tt1234569]\Season 1\01x01 - Episode Title.ext
It keeps filename very clean, and as the IMDB is close to an universal id, it allow to identify clearly what is what, not like the not so reliable (title,year) couple used to match on online db.
A simple regex like (tt[0-9]{7,}) matching on the filename would allow to get this id, which avoids the need to create an almost empty and overloading nfo file (but priority can still be given to the nfo file when present).
Could this be implemented to increase accuracy of media scanner ? For both movies and series ?
Thanks by advance,
Best regards,
Alexandre
While this does not seem totally unreasonable I fail to see the need. With proper naming including the date always for movies and when there are duplicate names for TV series I have not run into more than 1 or 2 out of every 1000 TV shows or movies that do not correctly match.
I also see problems with Plex trying to find IMDB ids in the file names.
It seems that it is mostly problems with the naming or with the order of agents that cause mismatches or failures in the matching of videos.
I believe that Plex has many other things of greater importance to work on without adding the complexities of parsing file names trying to find embedded IDs that may or may not be there.
@Elijah_Baley said:
While this does not seem totally unreasonable I fail to see the need. With proper naming including the date always for movies and when there are duplicate names for TV series I have not run into more than 1 or 2 out of every 1000 TV shows or movies that do not correctly match.
This is still too much when you spend time to name your files correctly. Also, without reviewing each entry, how would you know there’s a mistake in some entries ?
I also see problems with Plex trying to find IMDB ids in the file names.
IMDB ID is very specific, (tt[0-9]{7,}), and it could even be taken into account only when put into bracked [tt1234567] to avoid improbable mistake.
It seems that it is mostly problems with the naming or with the order of agents that cause mismatches or failures in the matching of videos.
Yep, I experienced that with Emby, TOMD isn’t good at all compared to TMDB.
I believe that Plex has many other things of greater importance to work on without adding the complexities of parsing file names trying to find embedded IDs that may or may not be there.
I don’t say there isn’t other priorities, but this is not a very consuming development either. Also, Plex is already parsing filenames, this is ~ a simple regex check to populate the guid field when matching (and parent folder for series, as done to retreive the name).
Depending on your workflow for creating your media and naming them this may or maynot be an easy addition for you, but Plex will already use the IMDB ID if it is in an NFO file next to the Movie.
