And they pre-date my using Plex by quite some number of years. I do not want to change them as Plex is not the only software accessing these files, and because FOR HUMANS, who access the file system directly, the repetitive nature of the Plex scheme is… well, repetitive, unnecessary and - quite frankly - annoying and difficult to read.
Why do I need {series}, {season} and {episode} in the file name when they are already in the path?! Just use the path FFS. It’s not difficult.
To be fair, Plex has correctly identified all but a small number of shows and episodes. But it gets wrong ALL episodes with a number in the episode name. So “01 An episode with a number 123.type” gets tagged as episode 23 (of the season defined in the path BTW, not of season 01 as might be expected of this erroneous parsing), while “02 Another episode.type” is correctly tagged as episode 2 in that same season.
Why in hell can’t we have the ability to add our own regular expressions? Or, at the very least, allow us to FIX the episodes it gets wrong.
I don’t know what kind of answer you expect. My naming scheme was always /Root/Show name/Season (even if there was only one)/Show name - SXXEYY - Title - quality.ext
I’ve been using this scheme long before I was using Plex. Hate it/love it it’s up to you.
What I like about Plex it’s recognizing properly files even if you dump them in wrong place as long as it is withing root folder.
I understand that multiple apps deal with this file naming scheme. I understand it’s kind of a standard. But is that a good reason for naming files this way? Catering for it is one thing. Making it the default is quite another.
As a “human readable” format, it sucks. It is cluttered, and some of the important information (season and episode) are embedded with other characters. But as a computer readable format it’s even worse, what with all the redundant information. Either save the data in the file path, or save it in the file name, not both. You may as well just shove all the files into a single generic “TV” folder (yes, I know, file system limitations may preclude this). But why bother with the sub-folders for series and season if they can be ignored, or worse, overridden, by the file name? That’s the real downfall of this scheme… conflicts.
Anyway, these are rhetorical questions. As was the original really. It’s just wrong.
@“Bartlomiej Baraniec” said:
What I like about Plex it’s recognizing properly files even if you dump them in wrong place as long as it is withing root folder.
I consider that to be a bug, not a feature. I want to know my file system is as accurate as the library application that accesses it.
Now in this situation if Plex were to recognise the file was in the wrong location and move it (optionally of course), THAT would be helpful, and about the only good reason I can think of for having the redundant data.
@HitsVille said:
Yep. Sonarr, Radar, Sickrage, Couch potato all cater for it not to mention Filebot. So it’s definitely just you.
But as a developer and designer I would never be allowed to simply say “that’s how it is”. You may support such pseudo-standards, but there is no compelling reason to demand them.
To be fair, Plex has correctly identified all but a small number of shows and episodes. But it gets wrong ALL episodes with a number in the episode name. So “01 An episode with a number 123.type” gets tagged as episode 23 (of the season defined in the path BTW, not of season 01 as might be expected of this erroneous parsing), while “02 Another episode.type” is correctly tagged as episode 2 in that same season.
Well some real examples would be good. I have 53TB of media including 385 TV shows and around 34000 episodes and haven’t encountered this once.
@Xhaka said:
Well some real examples would be good. I have 53TB of media including 385 TV shows and around 34000 episodes and haven’t encountered this once.
That naming isn’t anywhere near close to what Plex looks for. Plex has a naming standard, if you want your media to be found then you have to follow it. There’s no middle ground, and the Plex devs aren’t going to change their policy. So, if you want to use Plex and have it work the way it should, you will need to name and structure your media according to the Plex naming convention.
@kegobeer-plex said:
That naming isn’t anywhere near close to what Plex looks for. Plex has a naming standard, if you want your media to be found then you have to follow it. There’s no middle ground, and the Plex devs aren’t going to change their policy. So, if you want to use Plex and have it work the way it should, you will need to name and structure your media according to the Plex naming convention.
Seems you miss my point entirely. The Plex preferred naming scheme is just bloody awful and stupid. It is what it is, and I’m not asking them to change it (although that would be nice, at least as an option as most library apps allow), just noting the stupidity and inconsistency. Being able to have one series or season in the path and a different set in the file name is just really, really bad practice. That it still works without either highlighting or rectifying this issue is even worse.
Also, my naming convention predates Plex by many years. Other library systems have worked with it flawlessly, or at least allowed me to rectify bad scans.
BTW all, see how much fun it is trying to access files with the preferred Plex naming scheme through DNLA. As often as not, the file name gets truncated after 20 characters or so due to insufficient display area. This means every episode in the series/season may present with the same name - depending on show name length - as the episode number and name is the data that gets cut off. That does not occur with my naming scheme and is a large part of why I chose it - along with human readability. Not all media players have Plex clients (if only it were so), but almost all support DNLA. Just something to bear in mind.
@HitsVille said:
Well some real examples would be good. I have 53TB of media including 385 TV shows and around 34000 episodes and haven’t encountered this once.
Well when I asked for an example of something with a number in the filename that Plex gets wrong, I actually meant something that you had at least made a slight attempt to name it according to the TVDB/Plex method. That issue is nothing to do with the episode title containing a number.
@HitsVille said:
Yep. Sonarr, Radar, Sickrage, Couch potato all cater for it not to mention Filebot. So it’s definitely just you.
But as a developer and designer I would never be allowed to simply say “that’s how it is”. You may support such pseudo-standards, but there is no compelling reason to demand them.
Well personally as neither a developer or designer, just a simple end user… I tend to look what a particular software does or does not do. If one particular media server software doesn’t suit my needs or there are at least caveats to using it beyond what I’m willing to accept, well it would never have been installed in the first place.
You seem keen to throw words around… stupid, awful, bad practice etc… You compare it unfavourably with a host of as yet unnamed alternates (certainly not Emby, because that requires identical naming to Plex), so why are you wasting your/our time here if you already have something that isn’t so stupid or awful?
As for DLNA (not DNLA), I cant comment too much as I don’t use it. However there are 97 pages here http://forums.plex.tv/categories/dlna/p1 so clearly lots of people do. I wonder how they are all coping with your perceived issue?
The bottom line is Plex insists that TV shows are named according to the TVDB and FileBot has already been mentioned and can rename several terabytes in a matter of minutes. It really isn’t difficult. Adapt to the required naming or just move on.
You could write your own scanner for the recordings. It’s quite simple, there are numerous examples to be found.
Most of my files are from VDR, and since VDR is working on the same directory it’s not possible to change the naming. Paths and filenames have no consistent meaning, all useful information is in “info” file inside each directory. Yet with custom scanner all this is parsed properly, and then TVDB Agent finds the correct metadata just fine.
Oh, and it is really weird that even though most of the metadata can be edited afterwards, you can’t edit the season/episode and request new metadata for that? No way to fix incorrect matches. Why is there “Fix match” only on series top level, not per episode?
With movies, you can choose “Fix match” and type in the movie name. Then it will load metadata for that movie, whatever the original filename is.
@HitsVille said:
Well when I asked for an example of something with a number in the filename that Plex gets wrong, I actually meant something that you had at least made a slight attempt to name it according to the TVDB/Plex method. That issue is nothing to do with the episode title containing a number.
Sorry, I thought you were up to speed on the issue. Perhaps you should read the whole thread?
@HitsVille said:
The bottom line is Plex insists that TV shows are named according to the TVDB and FileBot has already been mentioned and can rename several terabytes in a matter of minutes. It really isn’t difficult. Adapt to the required naming or just move on.
No, the bottom line is that Plex shouldn’t be insisting on anything. Plex is a tool that is supposed to serve its users, not demand things from them. Certainly not when alternate parsing mechanisms are readily available. I could modify the code myself but would have to re-implement the fix after every upgrade. It’s simply not worth it.
As I have already mentioned, other library systems allow users to provide their own matching patterns, or to fix mismatched items. That Plex fails to do either of these is pretty piss poor. Don’t get me wrong, I love Plex, but that doesn’t mean it couldn’t be better. In fact, this is the only issue - thus far - I have with the software.
As far as the TVDB goes, it is nothing more than a database (that’s what the DB in TVDB stands for after all). It does not proscribe any particular file naming convention. It just provides the data. It is Plex that is parsing the name and matching it to the database using an API. The API does not require a particular file format either. It doesn’t work that way. Besides, other library apps that I have used in the past also utilise the TVDB, but they also allowed user defined matching patterns that worked 100% with my file naming scheme. So stop attempting to shift blame for this fiasco onto the TVDB. This is all Plex.
I have looked at FileBot, but given the reviews of that software on the Windows Store (seemingly the only way to get it now), I’m not inclined to shell out money for it. Regardless, renaming my files is a last resort. You have already read my reasons why.
@Zuikkis said:
You could write your own scanner for the recordings. It’s quite simple, there are numerous examples to be found.
Most of my files are from VDR, and since VDR is working on the same directory it’s not possible to change the naming. Paths and filenames have no consistent meaning, all useful information is in “info” file inside each directory. Yet with custom scanner all this is parsed properly, and then TVDB Agent finds the correct metadata just fine.
Thanks. The first truly useful (to me) suggestion. I know everyone else seems happy with the bastardised naming scheme Plex requests, but it’s just really poor. Anyone who has ever designed or maintained a database will get where I’m coming from. (that’s not to say they’d support my position, just that they’s understand it)
@Zuikkis said:
Oh, and it is really weird that even though most of the metadata can be edited afterwards, you can’t edit the season/episode and request new metadata for that? No way to fix incorrect matches. Why is there “Fix match” only on series top level, not per episode?
With movies, you can choose “Fix match” and type in the movie name. Then it will load metadata for that movie, whatever the original filename is.
Exactly! It may be on their to-do list, but it’s a glaring omission.
The reason why Plex works the way it does comes down to matching. There’s no real AI involved, so you have to hold its hand a bit with your structure. The important pieces it needs are Show name, Season Number and Episode Number. While it “can” pick these things up, it normally looks for them in specific places. Hence the reason for structure.
You don’t have to follow the structure. I just wouldn’t expect it to behave as intended if you don’t. Yes, it should serve up media in whatever structure you elect to use, but you’ll sacrifice matching your files to outside databases if you do. That’s your call.
