Was wondering with the new agent if you were going to look at the actors in a move.
We have a game here… Actors are listed by name, and there are actors with the same name. so If you look close you can fine the wrong actor image with a movie. Also that same actor may list movies he could not have been in.
In one case the actor is currently in his 20’s but is shown playing in a 1940 movie.
The sources used by the agent(s) usually have a database where each actor/artist has it’s own unique id. You can see that when you open Get Info > View XML on a movie of a Plex library of yours.
If you get a wrong picture from an actor, this has probably been mixed up in the database (e.g. a user not considering that there’s a younger actor of the same name and just dumping that picture or picking the 1st match of that name when editing the cast instead of double-checking it). The charm of The Movie Database is that you can actually contribute fixes yourself if you want to 
Any chance of it FINALLY parsing the IMDb ID from the filename? It’s always seemed ridiculous to create the clutter of a separate file just for the IMDb ID when it can fit unobtrusively in the filename.
I would love it if it did this as well - I add the id to my filenames so that I can copy and paste it quickly when I do manual matches to fix errors.
Not sure if anyone has already suggested this but, can he have an option to pull “Content Rating” for a movie from TheMovieDB instead of IMDB?
You have (2019) in the file name Twice & after the first there is a / after it.
I would try deleting from the end of the fist one (2019) to the end of the second one (2019).
At the beginning, it’s the folder name following by file name. Both have year as plex recommand
Then maybe try changing 기생충 to Parastie or maybe try and see if you change Parasite to 기생충 if your trying to get a different language if that is what that is.
Using radarr, the original file name was used instead of the English name. I have started to update file name but take time
Hi @HaldirI tested and it is the 기생충 that is causing the issue in Parasite (2019)/기생충 (2019) [Bluray-1080p] [8bit] [x264] [DTS-5.1] [FR+KO].mkv I will pass this along to our metadata team 
We won’t have this option to choose which one no - do you have an example of where it’s “wrong” for you with IMDB so I can see what our service is providing for it?
I have updated the download links in the first post, the latest version is now 1.19.5.2916-4ebbedbaf
New:
Fixes:
We’ll analyze the files for specific embedded metadata and use these values if we don’t have a value from the agent match, e.g. if your file has an embedded summary but the Plex agent doesn’t have this for the match then we’ll use the file’s metadata.
If you set the advanced library option “Prefer local metadata” then we’ll always prefer embedded metadata from the file over what the agent provides. Please DO NOT enable this if you don’t curate your own embedded tags in your files or if you’re not just testing it out (and thank you if you are). Setting this option is akin to moving the “Local media assets” to the top in the current agent.
Embedded tags are only fully support on MP4 containers (.mp4, .mov, .m4v) but have limited support for embedded Titles only for other containers like MKV.
Hey all,
I’ve just updated from the preview version that was uploaded on 1st June to the one made available today.
I’m using Mac 10.11 and after installing the new version, I can see the application has overwritten the previous preview build, which is what I expected.
My question is, do I need to create another new test library or will the fixes and other changes that come with the new preview build be applied to the test library I already have?
Thanks guys!
I mentioned something similar on to your comment about parsing IMDb IDs from the file name on this thread. My old naming used to have the IMDb ID, but the naming convention I recently adopted (slowly renaming all my files) matches the recommended naming much better and includes either tvdbXXXX or tmdbXXXX. In this case the XXXX is the movie or series ID; for episodes I get the season/episode info from the SXXEYY in the file name.
All that to say, a Plex Team Member (I think) essentially said this makes naming too cumbersome and it is their intention to just have the scanner figure out what the file is based on the information included in file names from their recommended naming.
I don’t see why some regex patterns for IDs, whether IMDb, TVDb, or TMDb, couldn’t be added as a first try. If a pattern hits, use that to match and get other data. If it doesn’t, fall back to ‘normal’ scanning methods. If we are already supposed to follow a fairly strict naming convention and are making sure names match IMDb, TMDb, or TVDb, it’s not that much more work to copy the ID from those sites and place it in the file name. Especially if doing so is going to ensure an exact match to what we want every time.
Edit: Thought my comment was on this thread, was on a different thread.
That is unfortunate.
IMDbs content ratings are worse because:
They are not standardized. Their ratings varies between rating systems and even ways of writing them, leading to many different ratings meaning the same thing, which causes confusion when you filter in Plex. On TheMoviedb there is a set list of standardized ratings, causing a much cleaner and easier to use result.
They can be inaccurate (I’ve seen this many times now), with no way of curating them at the source. On TheMoviedb the community helps curating.
A quick example is the classic movie “Jaws”. On IMDb the content ratings is “PG” which seems inappropriate. On TheMoviedb it is properly set as “R”. But there are many more examples.
Also, since we can already pick cast list source and description source as TheMoviedb, why not this?
What’s so frustrating is that the scanner already includes code to check for the IMDb ID in a “sidecar .nfo file”! Why the Plex devs are so vehemently against adding a check of the filename is just mind-boggling. I have 16,000+ movies, and with one simple line added to a text file, Kodi recognizes every single one of them. There are over 150 unrecognized in Plex.
def checkNfoFile(file, fileCount):
try:
path = None
# Depending on how many media files we have, check differently.
if fileCount == 1:
# Look for any NFO file.
for f in os.listdir(os.path.dirname(file)):
if f[-4:].lower() == '.nfo':
path = os.path.join(os.path.dirname(file), f)
break
else:
# Look for a sidecar NFO file.
path = os.path.splitext(file)[0] + '.nfo'
if path is not None and os.path.exists(path):
nfoText = open(path).read()
m = re.search('(tt[0-9]+)', nfoText)
if m:
return m.groups(1)[0]
except:
print "Warning, couldn't read NFO file."
return None
The entirety of Plex Movie Scanner.py isn’t terribly lengthy. I’m sure someone who knows what they’re doing could add it in minutes. I have literally zero programming experience, but I’m determined to spend a few hours (days? lol) neck deep in Python for Dummies and see if I can modify it myself 
Actually it was rated PG…
A quote from the times…
Los Angeles Times critic Charles Champlin disagreed with the film’s PG rating, saying that " Jaws is too gruesome for children, and likely to turn the stomach of the impressionable at any age.
@WasabiNME the new agent and scanner are written in C++ so adding IMDB ID extraction from filename to the old python agent is a non-starter. I have been discussing the idea of supporting a format for reading the IMDB and other ID’s from filenames, but I am just not sure what standard people use.
At least for IMDB there is a known prefix of tt and then 7-10 numbers that follow. For TMDB and TVDB things are unknown. What would you or others use?
perhaps tmdb:12345 or tmdb-12345 (as Windows does not support : in filenames)? with detection only parsing that if you have it in square brackets or parenthesis like [tmdb:12345] or (tmdb-12345)
I cannot promise anything, but It would certainly help to understand what people currently use.
I see other managers use TT0123456789 for the IMDB
most now require it be in brackets (TT0123456789) when they support the others.
(TM123456) Trakt (TK1234) etc…
I don’t think there is an official format yet… Plex start one…
IMDb is all I use, but more flexibility is better. Any regex shouldn’t care about brackets or a single separator character.
