So after a week or 2 of getting endless reminders/prompts to switch to the new movie/tv agents, I finally agreed and clicked the upgrade button. I forced metadata refresh because I read that it would fix movies being listed as New when they get upgraded to new version. After the refresh finished, 10-20% of 2700 collection were now unmatched and I have to individually go through and match each one. So frustrating and once again making me consider leaving the platform entirely because this has happened multiple times before.
I have nfo files for EVERYTHING in my collection to help match and keep everything consistent. Plex is the only platform that does not have built in abilities to read nfo’s and I honestly do not understand why. Issues like this would be nonexistent.
One thing Plex has yet to do is have it when you click on an item at the top of it’s page list the directory and actual file name like Emby does. This makes it so easy to correctly match since you can see what the item actually is versus what Plex claims it is.
Excluding the issues I have with the agent itself (especially lack of nfo support), just having the ability to filter by unmatched then initiate a bulk match (even if it required manual intervention for each entry) would be nice and extremely helpful in a situation like this.
Plex like any of these do not always match something correctly EVEN when it IS correctly named. Don’t assume the enduser made a mistake it only makes you look bad.
OH, yeah you would think so but having 500+ titles epically fail the matching screams user error.
I have thousands of titles, significantly more than OP and can say I haven’t had even 20 matching fails and it’s ALWAYS because there’s other media that has a very similar title.
Agree not even looking at the numbers but rather the percentage I’d have to agree. In movies out of 1100+ movies in both Emby and Plex I usually have to fix 3 at max. In TV Series I have a couple I have to temporarily move to allow for another named the same but different year to be matched properly and then move the other shows back to be matched correctly, and that’s out of a 167 shows. Movie series I still have to go through and put the year after a lot of the files cause I put them in front so they would play in the correct order in SMPlayer.
Did you miss the part where this happened AFTER I switched to the new agent? Meaning all movies were previously matched and no they didn’t require massive user intervention to get to that point. That was all based off the old scraper. As I said, manually clicking Match then letting it run on each individual movie matches correctly, so no, it’s not my filenaming. Nice reading comprehension and assumptions there chief.
I am however noticing a pattern; it seems as if the scraper would choke in multiple unmatches in a row. There’s very limited amounts of a solo film getting umatched. I see lots of pairs or sets of unmatches one after the other before it would have a string of matches. Clicking on a film then Match brings up the correct film and is an easy fix but doing a couple hundred of these is not how I want to spend my time…
You may be unaware that the file naming scheme has changed to include a metadata ID now. The new Movie Agent uses it to match the movies. There was even a test a ninja/employee did where they named the movie a nonsense name but gave it the proper ID and it found it correctly.
Also the new Agent ignores anything in square brackets. So the examples you gave that included the IMDB tt numbers would be ignored.
… but I would expect it to ~work, with one concern.
Does the Library point to multiple “letter” folders /movies/J and /movies/T etc., or does it point to only the parent /movies?
Each “immediate parent” folder should be added to the Library, not a grandparent.
Are there any “loose” movies in the /movies folder, or loose within the /movies/G and /movies/T folders? (“Loose” meaning that they aren’t in a per-movie folder.)
Arbitrarily deep nesting and mixed levels of nesting can cause problems, particularly for TV Shows.
Tip! : More specifically, the folder you want to specify as the content location for the library is the folder that contains each of the individual show folders. So, if you chose to categorize your children’s content separate from more “adult” content (e.g. /TV Shows/Kids/ShowName vs /TV Shows/Regular/ShowName ), then you would specify /TV Shows/Kids as the source location for a “kids” TV library.
I agree: use {imdb-tt3780500}. That’s what is documented.
In practice I think square brackets currently work. This matches to the IMDB ID:
It could potentially be that there was an issue with the matching service at that specific time but it shouldn’t “unmatch” things, they’ll just remain what they were.
If you just simply refresh metadata on one of these items (if there are any unmatched ones left) does it update and match correctly?
Also if you could do some of these refreshes and then grab your server logs and send them to me it could shed some light on what’s happening in your case.
Now that you mention it, it may indeed be the case that the code which looks for the IMDb ID doesn’t care about square brackets.
It is kinda bogo, because the hash which Plex is using is not computed across the whole video file. That’d take far too much time. Instead the hash is only computed for a small snippet of the video.
It is kinda bogo, because the hash which Plex is using is not computed across the whole video file. That’d take far too much time. Instead the hash is only computed for a small snippet of the video.