Misc scripts and tools. Undocumented scripts probably do what I need them to but aren't finished yet.
See the top-level README for setup instructions.
All these scripts use the same .env and requirements.
NOTE: on 06-29 these scripts have changed to using ENV vars to set up the Plex API details. This was done primarily to enable the timeout to apply to all Plex interactions.
If your .env file contains the original PLEX_URL and PLEX_TOKEN entries those will be silently changed for you.
# PLEX API ENV VARS
PLEXAPI_PLEXAPI_TIMEOUT='360'
PLEXAPI_AUTH_SERVER_BASEURL=https://plex.domain.tld
# Just the base URL, no /web or anything at the end.
# i.e. http://192.168.1.11:32400 or the like
PLEXAPI_AUTH_SERVER_TOKEN=PLEX-TOKEN
PLEXAPI_LOG_BACKUP_COUNT='3'
PLEXAPI_LOG_FORMAT='%(asctime)s %(module)12s:%(lineno)-4s %(levelname)-9s %(message)s' # PLEX API ENV VARS
PLEXAPI_LOG_LEVEL='INFO'
PLEXAPI_LOG_PATH='plexapi.log'
PLEXAPI_LOG_ROTATE_BYTES='512000'
PLEXAPI_LOG_SHOW_SECRETS='false'
# GENERAL ENV VARS
TMDB_KEY=TMDB_API_KEY # https://developers.themoviedb.org/3/getting-started/introduction
TVDB_KEY=TVDB_V4_API_KEY # currently not used; https://thetvdb.com/api-information
DELAY=1 # optional delay between items
LIBRARY_NAMES=Movies,TV Shows,Movies 4K # comma-separated list of libraries to act on
# IMAGE DOWNLOAD ENV VARS
## what-to-grab
### collection-related
INCLUDE_COLLECTION_ARTWORK=1 # should get-all-posters retrieve collection posters?
ONLY_COLLECTION_ARTWORK=0 # should get-all-posters retrieve ONLY collection posters?
ONLY_THESE_COLLECTIONS=Bing|Bang|Boing # only grab artwork for these collections and items in them
### tv-related
GRAB_SEASONS=1 # should get-all-posters retrieve season posters?
GRAB_EPISODES=1 # should get-all-posters retrieve episode posters? [requires GRAB_SEASONS]
### background-related
GRAB_BACKGROUNDS=1 # should get-all-posters retrieve backgrounds?
ARTWORK=1 # current background is downloaded with current poster
### quantity-related
ONLY_CURRENT=0 # should get-all-posters retrieve ONLY current artwork?
POSTER_DEPTH=20 # grab this many posters [0 grabs all] [ONLY_CURRENT overrides this]
### what-to-keep
KEEP_JUNK=0 # keep files that script would normally delete [incorrect filetypes, mainly]
FIND_OVERLAID_IMAGES=0 # check all downloaded images for overlays
# RETAIN_OVERLAID_IMAGES=0 # keep images that have an overlay EXIF tag [this will override the following two]
RETAIN_KOMETA_OVERLAID_IMAGES=0 # keep images that have the Kometa overlay EXIF tag
RETAIN_TCM_OVERLAID_IMAGES=0 # keep images that have the TCM overlay EXIF tag
## where-to-put-it
USE_ASSET_NAMING=1 # should grab-all-posters name images to match Kometa's Asset Directory requirements?
USE_ASSET_FOLDERS=1 # should those Kometa-Asset-Directory names use asset folders?
USE_ASSET_SUBFOLDERS=0 # create asset folders in subfolders ["Collections", "Other", or [0-9, A-Z]] ]
ASSETS_BY_LIBRARIES=1 # should those Kometa-Asset-Directory images be sorted into library folders?
ASSET_DIR=assets # top-level directory for those Kometa-Asset-Directory images
# if asset-directory naming is on, the next three are ignored
POSTER_DIR=extracted_posters # put downloaded posters here
CURRENT_POSTER_DIR=current_posters # put downloaded current posters and artwork here
POSTER_CONSOLIDATE=0 # if false, posters are separated into folders by library
## tracking
TRACK_URLS=1 # If set to 1, URLS are tracked and won't be downloaded twice
TRACK_COMPLETION=1 # If set to 1, movies/shows are tracked as complete by rating id
TRACK_IMAGE_SOURCES=1 # keep a file containing file names and source URLs
## general
POSTER_DOWNLOAD=1 # if false, generate a script rather than downloading
FOLDERS_ONLY=0 # Just build out the folder hierarchy; no image downloading
DEFAULT_YEARS_BACK=2 # in absence of a "last run date", grab things added this many years back.
# 0 sets the fallback date to the beginning of time
THREADED_DOWNLOADS=0 # should downloads be done in the background in threads?
RESET_LIBRARIES=Bing,Bang,Boing # reset "last time" count to the fallback date for these libraries
RESET_COLLECTIONS=Bing,Bang,Boing # CURRENTLY UNUSED
ADD_SOURCE_EXIF_COMMENT=1 # CURRENTLY UNUSED
# STATUS ENV VARS
PLEX_OWNER=yournamehere # account name of the server owner
TARGET_PLEX_URL=https://plex.domain2.tld # As above, the target of apply_all_status
TARGET_PLEX_TOKEN=PLEX-TOKEN-TWO # As above, the target of apply_all_status
TARGET_PLEX_OWNER=yournamehere # As above, the target of apply_all_status
LIBRARY_MAP={"LIBRARY_ON_PLEX":"LIBRARY_ON_TARGET_PLEX", ...}
# In apply_all_status, map libraries according to this JSON.
# RESET-POSTERS ENV VARS
TRACK_RESET_STATUS=1 # should reset-posters-* keep track of status and pick up where it left off?
LOCAL_RESET_ARCHIVE=1 # should reset-posters-tmdb keep a local archive of posters?
TARGET_LABELS=this label, that label # comma-separated list of labels to reset posters on
REMOVE_LABELS=0 # attempt to remove the TARGET_LABELs from items after resetting the poster
RESET_SEASONS=1 # reset-posters-* resets season artwork as well in TV libraries
RESET_EPISODES=1 # reset-posters-* resets episode artwork as well in TV libraries [requires RESET_SEASONS=True]
RETAIN_RESET_STATUS_FILE=0 # Don't delete the reset progress file at the end
FLUSH_STATUS_AT_START=0 # Delete the reset progress file at the start instead of reading it
RESET_SEASONS_WITH_SERIES=0 # If there isn't a season poster, use the series poster
DRY_RUN=0 # [currently only works with reset-posters-*]; don't actually do anything, just log
# LIST ITEM IDS ENV VARS
INCLUDE_COLLECTION_MEMBERS=0
ONLY_COLLECTION_MEMBERS=0
# DELETE_COLLECTION ENV VARS
KEEP_COLLECTIONS=bing,bang # List of collections to keep
# REMATCH-ITEMS ENV VARS
UNMATCHED_ONLY=1 # If 1, only rematch things that are currently unmatched
# RESET_ADDED_AT
ADJUST_DATE_FUTURES_ONLY=0 # Only look at items that show up as added in the future
ADJUST_DATE_EPOCH_ONLY=1 # Only adjust items that have "originally available" dates of `1970-01-01`
# REFRESH_METADATA
REFRESH_1970_ONLY=1 # If 1, only refresh things that have an originally-available date of 1970-01-01
# ACTOR ENV VARS
CAST_DEPTH=20 # how deep to go into the cast for actor collections
TOP_COUNT=10 # how many actors to export
KNOWN_FOR_ONLY=0 # ignore cast members who are not primarily known as actors
TRACK_GENDER=1 # Pay attention to actor gender [as recorded on TMDB]
BUILD_COLLECTIONS=0 # build yaml for Kometa config.yml
NUM_COLLECTIONS=20 # this many actors in Kometa yaml
MIN_GENDER_NONE = 5 # include minimum this many "none" gendered actors in the YAML, if possible
MIN_GENDER_FEMALE = 5 # include minimum this many "female" gendered actors in the YAML, if possible
MIN_GENDER_MALE = 5 # include minimum this many "male" gendered actors in the YAML, if possible
MIN_GENDER_NB = 5 # include minimum this many "non-binary" gendered actors in the YAML, if possible
# LOW POSTER COUNT
POSTER_THRESHOLD=10
# CREW COUNT
CREW_DEPTH=20
CREW_COUNT=100
TARGET_JOB=Director
SHOW_JOBS=0- adjust-added-dates.py - fix broken added and perhaps originally available dates in your library
- user-emails.py - extract user emails from your shares
- reset-posters-tmdb.py - reset all artwork in a library to TMDB default
- reset-posters-plex.py - reset all artwork in a library to Plex default
- grab-all-IDs.py - grab [into a sqlite DB] ratingKey, IMDB ID, TMDB ID, TVDB ID for everything in a library from plex
- grab-all-posters.py - grab some or all of the artwork for a library from plex
- image_picker.py - simple web app to make choosing active art from the images downloaded by grab-all-posters simpler
- grab-all-status.py - grab watch status for all users all libraries from plex
- apply-all-status.py - apply watch status for all users all libraries to plex from the file emitted by the previous script
- show-all-playlists.py - Show contents of all user playlists
- delete-collections.py - delete most or all collections from one or more libraries
- refresh-metadata.py - Refresh metadata individually on items in a library
- list-item-ids.py - Generate a list of IDs in libraries and/or collections
- actor-count.py - Generate a list of actor credit counts
- crew-count.py - Generate a list of crew credit counts
- list-low-poster-counts.py - Generate a list of items that have fewer than some number of posters in Plex
You have things in your library that show up added in the future, or way in the past.
This script will set the "added at" date and "originally available" date to match the thing's release date as found on TMDB, if the values set in Plex are more than a day or so off the TMDB release date.
Script-specific variables in .env:
ADJUST_DATE_FUTURES_ONLY=0 # Only look at items that show up as added in the future
ADJUST_DATE_EPOCH_ONLY=1 # Only adjust items that have "originally available" dates of `1970-01-01`
- setup as above
- Run with
python adjust-added-dates.py
You want a list of email addresses for all the people you share with.
Here is a quick and dirty [emphasis on "quick" and "dirty"] way to do that.
- setup as above
- Run with
python user-emails.py
The script will loop through all the shared users on your account and spit out username and email address.
connecting...
getting users...
looping over 26 users...
binguser - [email protected]
mrbang - [email protected]
boingster - [email protected]
...Perhaps you want to reset all the posters in a library
This script will set the poster for every series or movie to the default poster from TMDB/TVDB. It also saves that poster under ./posters/[movies|shows]/<rating_key>.ext in case you want to use them with Kometa's overlay resets.
If there is a file already located at ./posters/[movies|shows]/<rating_key>.ext, the script will use that image instead of retrieving a new one, so if you replace that local one with a poster of your choice, the script will use the custom one rather than the TMDB/TVDB default.
Script-specific variables in .env:
TRACK_RESET_STATUS=1 # pick up where the script left off
LOCAL_RESET_ARCHIVE=1 # keep a local archive of posters
TARGET_LABELS = Bing, Bang, Boing # reset artwork on items with these labels
REMOVE_LABELS=1 # remove labels when done [NOT RECOMMENDED]
RESET_SEASONS=1 # reset-posters-plex resets season artwork as well in TV libraries
RESET_EPISODES=1 # reset-posters-plex resets episode artwork as well in TV libraries [requires RESET_SEASONS=True]
RETAIN_RESET_STATUS_FILE=0 # Don't delete the reset progress file at the end
DRY_RUN=0 # [currently only works with reset-posters-*]; don't actually do anything, just log
FLUSH_STATUS_AT_START=0 # Delete the reset progress file at the start instead of reading them
RESET_SEASONS_WITH_SERIES=0 # If there isn't a season poster, use the series posterIf you set:
TRACK_RESET_STATUS=1
The script will keep track of where it is and will pick up at that point on subsequent runs. This is useful in the event of a lost connection to Plex.
Once it gets to the end of the library successfully, the tracking file is deleted. If you want to disable that for some reason, set RETAIN_RESET_STATUS_FILE to 1
If you want to reset any existing progress tracking and start from the beginning for some reason, set FLUSH_STATUS_AT_START to 1.
If you specify a comma-separated list of labels in the env file:
TARGET_LABELS = This label, That label, Another label
The script will reset posters only on movies with those labels assigned.
If you also set:
REMOVE_LABELS=1
The script will attempt to remove those labels after resetting the poster. I say "attempt" since in testing I have experienced an odd situation where no error occurs but the label is not removed. My test library of 230 4K-Dolby Movies contains 47 that fail in this way; every run it goes through the 47 movies "removing labels" without error yet they still have the labels on the next run.
Besides that Heisenbug, I don't recommend using this [REMOVE_LABELS] since the label removal takes a long time [dozens of seconds per item]. Doing this through the Plex UI is much faster.
If you set:
LOCAL_RESET_ARCHIVE=0
The script will set the artwork by sending the TMDB URL to Plex, without downloading the file locally first. This means a faster run compared to the initial run when downloading.
Example timings on a test library of 1140 TV Shows, resetting artwork for Show-Season-Episode level:
- No downloading: 1 hour 25 minutes
- With downloading: 2 hours 30 minutes
- Second run with downloaded archive: 1 hours 10 minutes
That is on a system with a 1G connection up and down, so values are just relative to each other.
The value of the local archive is that if you want to replace some of those images with your own, it provides a simple way to update all the posters in a library to custom posters of your own. When the script runs, it looks at that archive first, only downloading an image if one doesn't exist in the archive.
In that way it's sort of like Kometa's Asset Directory.
If you're just looking to reset as a one-off, that may not have value.
If no artwork is found at TMDB for a thing, no action is taken.
- setup as above
- Run with
python reset-posters-tmdb.py
tmdb config...
connecting to https://stream.BING.BANG...
getting items from [TV Shows - 4K]...
looping over 876 items...
[=---------------------------------------] 2.7% ... Age of Big Cats
At this time, there is no configuration aside from library name; it replaces all posters. It does not delete any posters from Plex, just grabs a URL and uses the API to set the poster to the URL.
Script-specific variables in .env:
TRACK_RESET_STATUS=1 # pick up where the script left off
LOCAL_RESET_ARCHIVE=1 # keep a local archive of posters
TARGET_LABELS = Bing, Bang, Boing # reset artwork on items with these labels
REMOVE_LABELS=1 # remove labels when done [NOT RECOMMENDED]
RESET_SEASONS=1 # reset-posters-plex resets season artwork as well in TV libraries
RESET_EPISODES=1 # reset-posters-plex resets episode artwork as well in TV libraries [requires RESET_SEASONS=True]
RETAIN_RESET_STATUS_FILE=0 # Don't delete the reset progress file at the end
DRY_RUN=0 # [currently only works with reset-posters-*]; don't actually do anything, just log
FLUSH_STATUS_AT_START=0 # Delete the reset progress file at the start instead of reading them
RESET_SEASONS_WITH_SERIES=0 # If there isn't a season poster, use the series poster
Same as reset-posters-tmdb.py, but it resets the artwork to the first item in Plex's own list of artwork, rather than downloading a new image from TMDB.
With RESET_SEASONS_WITH_SERIES=1, if the season doesn't have artwork the series artwork will be used instead.
Perhaps you want to gather all the IDs for everything in a library.
This script will go through a library and grab PLex RatingKey [which may be unique], IMDB ID, TMDB ID, and TVDB ID for everything in the list of libraries specified in the .env. It stores the data in a sqlite database called ids.sqlite; the repo copy of this file contains that data for 105871 movies and 26699 TV Shows.
Perhaps you want to get local copies of some or all the posters Plex knows about for everything in a library.
Maybe you find it easier to look through a bunch of options in CoverFlow or something.
Maybe you want to grab all the current custom art in a library to put in a Kometa asset directory or back it up for some other purpose.
This script will download some or all the posters for every item in a given set of libraries. It (probably) won't download the same thing more than once, so you can cancel it and restart it if need be. I say "probably" because the script is assuming that the list of posters retrieved from Plex is always in the same order [i.e. that new posters get appended to the end of the list]. On subsequent runs, the script checks only that a target file exists. It doesn't pay any attention to whether the two [the one on disk vs. the one coming from Plex] are the same image. I'll probably add a check to look at the image URL to attempt to ameliorate this at some point.
The script can name these files so that they are ready for use with Kometa's Asset Directory. This only works with ONLY_CURRENT set, since Kometa has no provision for multiple assets for a given thing.
If you have downloaded more than one image for each thing, see image_picker.py for a simpler way to choose which one you want to make active.
If THREADED_DOWNLOADS=1, the script queues downloads so they happen in the background in multiple threads. Once it's gone through the libraries listed in the config, it will then wait until the queue is drained before exiting. If you want to drop out of the library-scanning loop early, create a file stop.dat next to the script, and the library loop will exit at the end of the current show or movie, then go to the "waiting for the downloads" section. This allows you to get out early without flushing the queue [as control-C would do].
You can also skip the current library by creating skip.dat.
Script-specific variables in .env:
# IMAGE DOWNLOAD ENV VARS
## what-to-grab
GRAB_SEASONS=1 # should get-all-posters retrieve season posters?
GRAB_EPISODES=1 # should get-all-posters retrieve episode posters? [requires GRAB_SEASONS]
GRAB_BACKGROUNDS=1 # should get-all-posters retrieve backgrounds?
ONLY_CURRENT=0 # should get-all-posters retrieve ONLY current artwork?
ARTWORK=1 # current background is downloaded with current poster
INCLUDE_COLLECTION_ARTWORK=1 # should get-all-posters retrieve collection posters?
ONLY_COLLECTION_ARTWORK=0 # should get-all-posters retrieve ONLY collection posters?
ONLY_THESE_COLLECTIONS=Bing|Bang|Boing # only grab artwork for these collections and items in them
POSTER_DEPTH=20 # grab this many posters [0 grabs all]
KEEP_JUNK=0 # keep files that script would normally delete [incorrect filetypes, mainly]
FIND_OVERLAID_IMAGES=0 # check all downloaded images for overlays
# RETAIN_OVERLAID_IMAGES=0 # keep images that have an overlay EXIF tag [this will override the following two]
RETAIN_KOMETA_OVERLAID_IMAGES=0 # keep images that have the Kometa overlay EXIF tag
RETAIN_TCM_OVERLAID_IMAGES=0 # keep images that have the TCM overlay EXIF tag
## where-to-put-it
USE_ASSET_NAMING=1 # should grab-all-posters name images to match Kometa's Asset Directory requirements?
USE_ASSET_FOLDERS=1 # should those Kometa-Asset-Directory names use asset folders?
USE_ASSET_SUBFOLDERS=0 # create asset folders in subfolders ["Collections", "Other", or [0-9, A-Z]] ]
ASSETS_BY_LIBRARIES=1 # should those Kometa-Asset-Directory images be sorted into library folders?
ASSET_DIR=assets # top-level directory for those Kometa-Asset-Directory images
# if asset-directory naming is on, the next three are ignored
POSTER_DIR=extracted_posters # put downloaded posters here
CURRENT_POSTER_DIR=current_posters # put downloaded current posters and artwork here
POSTER_CONSOLIDATE=0 # if false, posters are separated into folders by library
## tracking
TRACK_URLS=1 # If set to 1, URLS are tracked and won't be downloaded twice
TRACK_COMPLETION=1 # If set to 1, movies/shows are tracked as complete by rating id
TRACK_IMAGE_SOURCES=1 # keep a file containing file names and source URLs
## general
POSTER_DOWNLOAD=1 # if false, generate a script rather than downloading
FOLDERS_ONLY=0 # Just build out the folder hierarchy; no image downloading
DEFAULT_YEARS_BACK=2 # in absence of a "last run date", grab things added this many years back.
# 0 sets the fallback date to the beginning of time
THREADED_DOWNLOADS=0 # should downloads be done in the background in threads?
RESET_LIBRARIES=Bing,Bang,Boing # reset "last time" count to the fallback date for these libraries
RESET_COLLECTIONS=Bing,Bang,Boing # CURRENTLY UNUSED
ADD_SOURCE_EXIF_COMMENT=1 # CURRENTLY UNUSEDThe point of "POSTER_DEPTH" is that sometimes movies have an insane number of posters, and maybe you don't want all 257 Endgame posters or whatever. Or maybe you want to download them in batches.
If "POSTER_DOWNLOAD" is 0, the script will build a shell/batch script for each library to download the images at your convenience instead of downloading them as it runs, so you can run the downloads overnight or on a different machine with ALL THE DISK SPACE or something.
If "POSTER_CONSOLIDATE" is 1, the script will store all the images in one directory rather than separating them by library name. The idea is that Plex shows the same set of posters for "Star Wars" whether it's in your "Movies" or "Movies - 4K" or whatever other libraries, so there's no reason to pull the same set of posters multiple times. There is an example below.
If "INCLUDE_COLLECTION_ARTWORK" is 1, the script will grab artwork for all the collections in the target library.
If "ONLY_COLLECTION_ARTWORK" is 1, the script will grab artwork for ONLY the collections in the target library; artwork for individual items [movies, shows] will not be grabbed.
If "ONLY_THESE_COLLECTIONS" is not empty, the script will grab artwork for ONLY the collections listed and items contained in those collections. This doesn't affect the sorting or naming, just the filter applied when asking Plex for the items. IF YOU DON'T CHANGE THIS SETTING, NOTHING WILL BE DOWNLOADED.
If "TRACK_URLS" is 1, the script will track every URL it downloads in a sqlite database. On future runs, if a given URL is found in that database it won't be downloaded a second time. This may save time if the same URL appears multiple times in the list of posters from Plex.
If "TRACK_COMPLETION" is 1, the script records collections/movies/shows/seasons/episodes by rating key in a sqlite database. On future runs, if a given rating key is found in that database the thing is considered complete and it will be skipped. This will save time in subsequent runs as the script will not look through all the images for a thing only to determine that it's already downloaded all of them. HOWEVER, this also means that if you increase POSTER_DEPTH, those additional images won't be picked up when you run the script again, since the item will be marked as complete.
The script keeps track of the last date it retrieved items from a library [for show libraries it also tracks seasons and episodes separately], and on each run will only retrieve items added since that date. If there is no "last run date" for a given thing, the script uses a fallback "last run" date of today - DEFAULT_YEARS_BACK.
If DEFAULT_YEARS_BACK = 0, the fallback date is "the beginning of time". There is one other circumstance that will result in a fallback date of "the beginning of time".
If:
- You are running Windows
DEFAULT_YEARS_BACKis > 0- the calculated fallback date is before 01/01/1970
Then the "beginning of time" fallback date will be used. This is a Windows issue.
Normally, this fallback date is used only if there is no last-run date stored, for example the first time you run the script. This means that if you run the script once with DEFAULT_YEARS_BACK=2 then change that to DEFAULT_YEARS_BACK=0, nothing new will be downloaded. The script will read the last run date from its database and will never look at the new fallback date.
You can use RESET_LIBRARIES to force the "last run date" to that fallback date for one or more libraries.
If you want to reset all libraries and clear all history, delete mediascripts.sqlite.
For example:
You run grab-all-posters with DEFAULT_YEARS_BACK=2; it downloads posters for half your "Movies" library. Now you want to grab all the rest. You change that to DEFAULT_YEARS_BACK=0 and run grab-all-posters again. Nothing new will be downloaded since the last run date is now the time of that first run, and nothing has been added to Plex since then. If you want to grab all posters from the beginning of time for that library, set:
DEFAULT_YEARS_BACK=0
RESET_LIBRARIES=Movies
That will set the fallback date to "the beginning of time" and apply that new fallback date to the "Movies" library only.
If "FIND_OVERLAID_IMAGES" is 1, the script checks every imnage it downloads for the EXIF tag that indicates Kometa created it. If found, the image is deleted. You can override the deleting with RETAIN_KOMETA_OVERLAID_IMAGES and/or RETAIN_TCM_OVERLAID_IMAGES.
If "RETAIN_KOMETA_OVERLAID_IMAGES" is 1, those images with the Kometa EXIF tag are not deleted.
If "RETAIN_TCM_OVERLAID_IMAGES" is 1, those images with the Kometa EXIF tag are not deleted.
If "RETAIN_OVERLAID_IMAGES" is 1, the previous two settings will be forced to 0 and all overlaid images will be retained. This is a older deprecated setting.
NOTE: ONLY_CURRENT and POSTER_DEPTH do not take these images into account, meaning that if you have:
ONLY_CURRENT=1
RETAIN_KOMETA_OVERLAID_IMAGES=0
Then nothing will be retained for items with overlaid posters. grab-all-posters will download the current art, find that it has an overlay, delete it, then go to the next movie/show.
Similarly:
ONLY_CURRENT=0
POSTER_DEPTH=20
RETAIN_KOMETA_OVERLAID_IMAGES=0
This won't grab images until you have 20 downloaded. It will grab 20 images, and if ten are found to have overlays, those ten will be deleted and you will end up with 10.
- setup as above
- Run with
python grab-all-posters.py
The posters will be sorted by library [if enabled] with each poster getting an incremented number, like this:
The image names are: title-source-location-INCREMENT.ext
source is where plex reports it got the image: tmdb, fanarttv, gracenote, etc. This will alaways be "None" for collection images since they are provided by the user or generated [the four-poster ones] by Plex.
location will be local or remote depending whether the URL pointed to the plex server or to some other site like tmdb.
The folder structure in which the images are saved is controlled by a combination of settings; please review the examples below to find the format you want and the settings that you need to generate it.
All movies and TV shows in a single folder:
POSTER_CONSOLIDATE=1:
extracted_posters/
└── all_libraries
├── 3 12 Hours-847208
│ ├── 3 12 Hours-tmdb-local-001.jpg
│ ├── 3 12 Hours-tmdb-remote-002.jpg
│ └── backgrounds
│ ├── background-tmdb-local-001.jpg
│ └── background-tmdb-remote-002.jpg
├── 9-1-1 Lone Star-89393
│ ├── 9-1-1 Lone Star-local-local-001.jpg
│ ├── 9-1-1 Lone Star-tmdb-local-002.jpg
│ ├── S01-Season 1
│ │ ├── S01-Season 1-local-local-001.jpg
│ │ ├── S01-Season 1-tmdb-local-002.jpg
│ │ ├── S01E01-Pilot
│ │ │ ├── S01E01-Pilot-local-local-001.jpg
│ │ │ └── S01E01-Pilot-tmdb-remote-002.jpg
│ │ └── backgrounds
│ │ ├── background-fanarttv-remote-001.jpg
│ │ └── background-fanarttv-remote-002.jpg
│ └── backgrounds
│ ├── background-local-local-001.jpg
│ └── background-tmdb-remote-002.jpg
├── collection-ABC
│ ├── ABC-None-local-001.jpg
│ └── ABC-None-local-002.jpg
└── collection-IMDb Top 250
├── IMDb Top 250-None-local-001.jpg
└── IMDb Top 250-None-local-002.pngSplit by Plex library name ['Movies' and 'TV Shows' are Plex library names]:
POSTER_CONSOLIDATE=0:
extracted_posters/
├── Movies
│ ├── 3 12 Hours-847208
│ │ ├── 3 12 Hours-tmdb-local-001.jpg
│ │ ├── 3 12 Hours-tmdb-remote-002.jpg
│ │ └── backgrounds
│ │ ├── background-tmdb-local-001.jpg
│ │ └── background-tmdb-remote-002.jpg
│ └── collection-IMDb Top 250
│ ├── IMDb Top 250-None-local-001.jpg
│ └── IMDb Top 250-None-local-002.png
└── TV Shows
├── 9-1-1 Lone Star-89393
│ ├── 9-1-1 Lone Star-local-local-001.jpg
│ ├── 9-1-1 Lone Star-tmdb-local-002.jpg
│ ├── S01-Season 1
│ │ ├── S01-Season 1-local-local-001.jpg
│ │ ├── S01-Season 1-tmdb-local-002.jpg
│ │ ├── S01E01-Pilot
│ │ │ ├── S01E01-Pilot-local-local-001.jpg
│ │ │ └── S01E01-Pilot-tmdb-remote-002.jpg
│ │ └── backgrounds
│ │ ├── background-fanarttv-remote-001.jpg
│ │ └── background-fanarttv-remote-002.jpg
│ └── backgrounds
│ ├── background-local-local-001.jpg
│ └── background-tmdb-remote-002.jpg
└── collection-ABC
├── ABC-None-local-001.jpg
└── ABC-None-local-002.jpgUse Kometa Asset-directory naming, flat:
USE_ASSET_NAMING=1
USE_ASSET_FOLDERS=0
ASSETS_BY_LIBRARIES=0
ONLY_CURRENT=1
assets
├── Adam-12 (1968) {tvdb-78686}.jpg
├── Adam-12 (1968) {tvdb-78686}_S01E01.jpg
├── Adam-12 (1968) {tvdb-78686}_S01E02.jpg
...
├── Adam-12 (1968) {tvdb-78686}_Season01.jpg
├── Adam-12 (1968) {tvdb-78686}_background.jpg
├── Adam-12 Collection.jpg
├── Star Wars (1977) {imdb-tt0076759} {tmdb-11}.jpg
└── Star Wars (1977) {imdb-tt0076759} {tmdb-11}_background.jpgUse Kometa Asset-directory naming, movies and TV in a single directory, split by item name:
USE_ASSET_NAMING=1
USE_ASSET_FOLDERS=1
ASSETS_BY_LIBRARIES=0
ONLY_CURRENT=1 OR POSTER_DEPTH=1
assets
├── Adam-12 (1968) {tvdb-78686}
│ ├── S01E01.jpg
│ ├── S01E02.jpg
...
│ ├── Season01.jpg
│ ├── background.jpg
│ └── poster.jpg
├── Adam-12 Collection
│ └── poster.jpg
└── Star Wars (1977) {imdb-tt0076759} {tmdb-11}
├── background.jpg
└── poster.jpgUse Kometa Asset-directory naming, split by Plex library name, flat folder:
USE_ASSET_NAMING=1
USE_ASSET_FOLDERS=0
ASSETS_BY_LIBRARIES=1
ONLY_CURRENT=1
assets
├── One Movie
│ ├── Star Wars (1977) {imdb-tt0076759} {tmdb-11}.jpg
│ └── Star Wars (1977) {imdb-tt0076759} {tmdb-11}_background.jpg
└── One Show
├── Adam-12 (1968) {tvdb-78686}.jpg
├── Adam-12 (1968) {tvdb-78686}_S01E01.jpg
├── Adam-12 (1968) {tvdb-78686}_S01E02.jpg
...
├── Adam-12 (1968) {tvdb-78686}_Season01.jpg
├── Adam-12 (1968) {tvdb-78686}_background.jpg
└── Adam-12 Collection.jpgUse Kometa Asset-directory naming, split by Plex library name, split by item name:
USE_ASSET_NAMING=1
USE_ASSET_FOLDERS=1
ASSETS_BY_LIBRARIES=1
ONLY_CURRENT=1 OR POSTER_DEPTH=1
assets
├── One Movie
│ └── Star Wars (1977) {imdb-tt0076759} {tmdb-11}
│ ├── background.jpg
│ └── poster.jpg
└── One Show
├── Adam-12 (1968) {tvdb-78686}
│ ├── S01E01.jpg
│ ├── S01E02.jpg
...
│ ├── Season01.jpg
│ ├── background.jpg
│ └── poster.jpg
└── Adam-12 Collection
└── poster.jpgUse Kometa Asset-directory naming, split by Plex library name, split by first letter, split by item name:
USE_ASSET_NAMING=1
USE_ASSET_FOLDERS=1
ASSETS_BY_LIBRARIES=1
ONLY_CURRENT=1 OR POSTER_DEPTH=1
USE_ASSET_SUBFOLDERS=1
assets
├── One Movie
│ └── S
│ └── Star Wars (1977) {imdb-tt0076759} {tmdb-11}
│ ├── background.jpg
│ └── poster.jpg
└── One Show
├── A
│ ├── Adam-12 (1968) {tvdb-78686}
│ │ ├── S01E01.jpg
│ │ ├── S01E02.jpg
...
│ │ ├── Season01.jpg
│ │ ├── background.jpg
│ │ └── poster.jpg
└── Collections
└── Adam-12 Collection
└── poster.jpgYou've run grab-all-posters and now you want a simpler way to choose which of those hundreds of images you want as your active asset
It presents a web UI that lets you scroll through the images that grab-all-posters downloaded, selecting the one you want by clicking on it.
When you click on an image, it is copied to a parallel file system rooted at active_assets with the correct pathing and naming for the Kometa asset directory.
You can then copy that active_assets directory to the Kometa config dir ready for use.
It keeps track of which images have been chosen on a show/movie basis [is a json file] so that when you come back the current image is highlighted.
This script does not use anything from the .env, but it does make some assumptions:
- You are working with a directory of images produced by
grab-all-posters - That directory is named
assetsand is in this directory next to this script. - You can run a web server on the machine that listens on port 5001
- setup as above
- Run with
python image_picker.py - Go to one of the URLs presented.
In this case I'm running it on a server in my home:
python image-picker.py
* Serving Flask app 'image-picker'
* Debug mode: off
WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
* Running on all addresses (0.0.0.0)
* Running on http://127.0.0.1:5001
* Running on http://192.168.1.11:5001
Since I am running it on a machine remote from the laptop I am on, I go to the second URL.
I see a list of my libraries along with a reminder of where images are being read from and copied to:
Clicking into the library shows the next level of folders [I have mine split by letter here]:
Drilling in displays the movies/shows
Drilling into an individual thing shows the posters for this item, with the filename and image size:
Note that there are checkboxes to show/hide types of images to minimize scrolling, and a slider to control the image size if you need a closer look.
Clicking on an image will copy it to active_assets and it is highlighted in the list.
Now that I've chosen a background, I can uncheck "Show Backgrounds" to hide them from the list. I've also increased the poster size here.
And again, clicking on the image selects and copies.
At this point if you go look in the file system you'll see:
active_assets/Movies/
└── 3
└── 300 (2007) {imdb-tt0416449} {tmdb-1271}
├── background.jpg
└── poster.jpg
TV works the same way except that there are also Season and Episode images.
Perhaps you want to move or restore watch status from one server to another [or to a rebuild]
This script will retrieve all watched items for all libraries on a given plex server. It stores them in a tab-delimited file.
Script-specific variables in .env:
PLEX_OWNER=yournamehere # account name of the server owner- setup as above
- Run with
python grab-all-status.py
Connecting to https://cp1.DOMAIN.TLD...
------------ chazlarson ------------
------------ Movies - 4K ------------
chazlarson movie Movies - 4K It Comes at Night 2017 R
chazlarson movie Movies - 4K Mad Max: Fury Road 2015 R
chazlarson movie Movies - 4K Rio 2011 G
chazlarson movie Movies - 4K Rocky 1976 PG
chazlarson movie Movies - 4K The Witch 2015 R
------------ Movies - 4K DV ------------
chazlarson movie Movies - 4K DV It Comes at Night 2017 R
chazlarson movie Movies - 4K DV Mad Max: Fury Road 2015 R
...The file contains one row per user/library/item:
chazlarson movie Movies - 4K It Comes at Night 2017 R
chazlarson movie Movies - 4K Mad Max: Fury Road 2015 R
chazlarson movie Movies - 4K Rio 2011 G
chazlarson movie Movies - 4K Rocky 1976 PG
chazlarson movie Movies - 4K The Witch 2015 R
chazlarson movie Movies - 4K DV It Comes at Night 2017 R
chazlarson movie Movies - 4K DV Mad Max: Fury Road 2015 R
...This script reads the file produces by the previous script and applies the watched status for each user/library/item
Script-specific variables in .env:
TARGET_PLEX_URL=https://plex.domain2.tld
TARGET_PLEX_TOKEN=PLEX-TOKEN-TWO
TARGET_PLEX_OWNER=yournamehere
LIBRARY_MAP={"LIBRARY_ON_PLEX":"LIBRARY_ON_TARGET_PLEX", ...}These values are for the TARGET of this script; this is easier than requiring you to edit the PLEX_URL, etc, when running the script.
If the target Plex has different library names, you can map one to the other in LIBRARY_MAP.
For example, if the TV library on the source Plex is called "TV - 1080p" and on the target Plex it's "TV Shows on SpoonFlix", you'd map that with:
LIBRARY_MAP={"TV - 1080p":"TV Shows on SpoonFlix"}
And any records from the status.txt file that came from the "TV - 1080p" library on the source Plex would get applied to the "TV Shows on SpoonFlix" library on the target.
- setup as above
- Run with
python apply-all-status.py
connecting to https://cp1.DOMAIN.TLD...
------------ Movies - 4K ------------
Searching for It Comes at Night Marked watched for chazlarson
...There might be a problem with special characters in titles.
Perhaps you want to creep on your users and see what they have on their playlists
This script will list the contents of all playlists users have created [except the owner's, since you already have access to those].
Script-specific variables in .env:
NONE
- setup as above
- Run with
python show-all-playlists.py
connecting to https://cp1.DOMAIN.TLD...
------------ ozzy ------------
------------ ozzy playlist: Abbott Elementary ------------
episode - Abbott Elementary s01e01 Pilot
episode - Abbott Elementary s01e02 Light Bulb
episode - Abbott Elementary s01e03 Wishlist
episode - Abbott Elementary s01e04 New Tech
episode - Abbott Elementary s01e05 Student Transfer
episode - Abbott Elementary s01e06 Gifted Program
episode - Abbott Elementary s01e07 Art Teacher
------------ ozzy playlist: The Bear ------------
episode - The Bear s01e01 System
episode - The Bear s01e02 Hands
...
------------ tony ------------
------------ tony playlist: Specials ------------
movie - Comedy Central Roast of James Franco
movie - Comedy Central Roast of Justin Bieber
movie - Comedy Central Roast of Bruce Willis
------------ tony playlist: Ted ------------
movie - Ted
movie - The Invisible Man
movie - Ace Ventura: When Nature Calls
...Perhaps you want to delete all the collections in one or more libraries
This script will simply delete all collections from the libraries specified in the config, except those listed.
Script-specific variables in .env:
KEEP_COLLECTIONS=bing,bang # comma-separated list of collections to keep- setup as above
- Run with
python delete-collections.py
39 collection(s) retrieved...****
Collection delete - Plex |█████████▎ | ▂▄▆ 9/39 [23%] in 14s (0.6/s, eta: 27s)
-> deleting: 98 Best Action Movies Of All TimePerhaps you want to refresh metadata in one or more libraries; there are situations where refreshing the whole library doesn't work so you have to do it in groups, which can be tiring.
This script will simply loop through the libraries specified in the config, refreshing each item in the library. It waits for the specified DELAY between each.
Script-specific variables in .env:
NONE
- setup as above
- Run with
python refresh-metadata.py
getting items from [TV Shows - 4K]...
looping over 1086 items...
[========================================] 100.1% ... Zoey's Extraordinary Playlist - DONE
getting items from [ TV Shows - Anime]...
looping over 2964 items...
[========================================] 100.0% ... Ōkami Shōnen Ken - DONEPerhaps you want a list of all the IDs of everything in your libraries or collections.
This script wil output this data into its log file:
on 0: INFO: 11/05/2023 05:03:04 PM This collection is called New Episodes
on 0: INFO: 11/05/2023 05:03:05 PM Collection: New Episodes item 1/ 125 | TVDb ID: 411029 | IMDb ID: tt15320362 | All the Light We Cannot See
on 0: INFO: 11/05/2023 05:03:05 PM Collection: New Episodes item 2/ 125 | TVDb ID: 421526 | IMDb ID: tt15475330 | Black Cake
on 0: INFO: 11/05/2023 05:03:05 PM Collection: New Episodes item 3/ 125 | TVDb ID: 419379 | IMDb ID: tt15384586 | Fellow Travelers
on 0: INFO: 11/05/2023 05:03:05 PM Collection: New Episodes item 4/ 125 | TVDb ID: 439494 | IMDb ID: tt10270200 | The Vanishing Triangle
or
on 5782: INFO: 11/05/2023 05:07:49 PM tem 5782/ 5786 | TMDb ID: 9398 | IMDb ID: tt0196229 | Zoolander
on 5783: INFO: 11/05/2023 05:07:49 PM tem 5783/ 5786 | TMDb ID: 329833 | IMDb ID: tt1608290 | Zoolander 2
on 5784: INFO: 11/05/2023 05:07:49 PM tem 5784/ 5786 | TMDb ID: 269149 | IMDb ID: tt2948356 | Zootopia
env vars are the same as grab-all-posters.py for the most part [where they apply], except for:
INCLUDE_COLLECTION_MEMBERS=0
ONLY_COLLECTION_MEMBERS=0
Which probably do about what you'd expect.
- setup as above
- Run with
python list-item-ids.py
Perhaps you want a list of actors with a count of how many movies from your libraries they have been in.
This script connects to a plex library, and grabs all the items. For each item, it then gets the cast from TMDB and keeps track across all items how many times it sees each actor within the list, looking down to a configurable depth. For TV libraries, it's pulling the cast at the show level, and I haven't yet done any research to see if guest stars from individual episodes show up in that list.
At the end, it produces a list of a configurable size in descending order of number of appearances.
Script-specific variables in .env:
CAST_DEPTH=20 ### HOW DEEP TO GO INTO EACH MOVIE CAST
TOP_COUNT=10 ### PUT THIS MANY INTO THE FILE AT THE END
KNOWN_FOR_ONLY=0 ### ONLY CONSIDER CAST MEMBERS "KNOWN FOR" ACTING
BUILD_COLLECTIONS=0 # build yaml for Kometa config.yml
NUM_COLLECTIONS=20 # this many actors in Kometa yaml
MIN_GENDER_NONE = 5 # include minimum this many "none" gendered actors in the YAML, if possible
MIN_GENDER_FEMALE = 5 # include minimum this many "female" gendered actors in the YAML, if possible
MIN_GENDER_MALE = 5 # include minimum this many "male" gendered actors in the YAML, if possible
MIN_GENDER_NB = 5 # include minimum this many "non-binary" gendered actors in the YAML, if possible
CAST_DEPTH is meant to prevent some journeyman character actor from showing up in the top ten; I'm thinking of someone like Clint Howard who's been in the cast of many movies, but I'm guessing when you think of the top ten actors in your library you're not thinking about Clint. Maybe you are, though, in which case set that higher.
TOP_COUNT is the number of actors to show in the list at the end.
Every person in the cast list has a "known_for_department" attribute on TMDB. If you set KNOWN_FOR_ONLY=True, then people who don't have "Acting" in that field will be excluded. Turning this on may slightly distort results. For example, Harold Ramis is the second lead in "Stripes" and "Ghostbusters", but he is primarily known for "Directing" according to TMDB, so if you turn this flag on he doesn't get counted at all.
BUILD_COLLECTIONS will make the script build some YAML to paste into your Kometa config file to generate collections.
NUM_COLLECTIONS controls the number of collections in that YAML
TRACK_GENDER controls whether the script pays attention to actor gender
MIN_GENDER_* control the minimum number of that gender [as recorded by TMDB] actor to include in the list [provided TRACK_GENDER=1]
Actors are sorted into lists by the four genders recorded at TMDB. The top MIN_GENDER_* for each are added to the final list, then if there is space left over the remainder is filled from the master actor list.
If the four MIN_GENDER_* sum to more than NUM_COLLECTIONS, the script exists with an error.
- setup as above
- Run with
python actor-count.py
connecting to https://plex.bing.bang...
getting items from [Movies - 4K DV]...
Completed loading 1996 items from Movies - 4K DV
looping over 1996 items...
[======----------------------------------] 15.0% ... Captain America: Civil War It will go through all your movies, and then at the end print out however many actors you specified in TOP_COUNT along with a bunch of other statistics.
Sample results for the library above:
CAST_DEPTH=20 TOP_COUNT = 10
30 Samuel L. Jackson - 2231
22 Idris Elba - 17605
22 Tom Hanks - 31
21 Woody Harrelson - 57755
21 Gary Oldman - 64
21 Tom Cruise - 500
21 Morgan Freeman - 192
21 Sylvester Stallone - 16483
20 Willem Dafoe - 5293
20 Laurence Fishburne - 2975CAST_DEPTH=40 TOP_COUNT=10
33 Samuel L. Jackson - 2231
24 John Ratzenberger - 7907
23 Tom Hanks - 31
22 Bruce Willis - 62
22 Gary Oldman - 64
22 Idris Elba - 17605
22 Morgan Freeman - 192
21 Woody Harrelson - 57755
21 Fred Tatasciore - 60279
21 Tom Cruise - 500Note that the top ten changed dramatically due to looking deeper into the cast lists.
Perhaps you want a list of crew members with a count of how many movies from your libraries they have been credited in.
This script connects to a plex library, and grabs all the items. For each item, it then gets the crew from TMDB and keeps track across all items how many times it sees each individual with the configured TARGET_JOB within the list, looking down to a configurable depth.
At the end, it produces a list of a configurable size in descending order of number of appearances.
Script-specific variables in .env:
CREW_DEPTH=20 ### HOW DEEP TO GO INTO EACH MOVIE CREW
CREW_COUNT=100 ### HOW MANY INDIVIDUALS TO REPORT AT THE END
TARGET_JOB=Director ### WHAT JOB TO TRACK
SHOW_JOBS=0 ### Display a list of all the jobs the script saw
CREW_DEPTH is meant to allow the script to look deeper into the crew to find all the individuals working as TARGET_JOB.
CREW_COUNT is the number of individuals to show in the list at the end.
If SHOW_JOBS is set to 1, the script will also output a list of all the jobs it saw, if you want a reference to what may be available.
- setup as above
- Run with
python crew-count.py
connecting to Plex...
getting items from [Test-Movies]...
looping over 35 items...
[=========================================] 102.9% ... Wild Gals of the Naked West It will go through all your movies, and then at the end print out however many actors you specified in TOP_COUNT along with a bunch of other statistics.
Sample results for the library above:
CREW_DEPTH=20 CREW_COUNT=100 TARGET_JOB=Director
Top 27 Director in [Test-Movies]:
3 Jules Bass - 16410
3 Arthur Rankin, Jr. - 16411
2 Peyton Reed - 59026
2 Pierre Coffin - 124747
2 Chris Renaud - 124748
2 Robert Wise - 1744
2 Nicholas Meyer - 1788
2 Leonard Nimoy - 1749
2 Jonathan Frakes - 2388
1 Ed Herzog - 219492
1 Zack Snyder - 15217
1 Noam Murro - 78914
1 Edward Berger - 221522
1 Bob Fosse - 66777
1 Jean-Pierre Jeunet - 2419
1 Dario Argento - 4955
1 Hideaki Anno - 77921
1 George Miller - 20629
1 Larry Roemer - 144977
1 Søren Fauli - 110047
1 William Shatner - 1748
1 David Carson - 2380
1 Stuart Baird - 2523
1 Richard Marquand - 19800
1 Jūzō Itami - 69167
1 Michael Boyle - 2754307
1 Russ Meyer - 4590CREW_DEPTH=5 CREW_COUNT=100 TARGET_JOB=Director
Top 22 Director in [Test-Movies]:
3 Jules Bass - 16410
3 Arthur Rankin, Jr. - 16411
2 Peyton Reed - 59026
2 Pierre Coffin - 124747
2 Chris Renaud - 124748
2 Leonard Nimoy - 1749
1 Ed Herzog - 219492
1 Zack Snyder - 15217
1 Noam Murro - 78914
1 Bob Fosse - 66777
1 Dario Argento - 4955
1 Hideaki Anno - 77921
1 George Miller - 20629
1 Larry Roemer - 144977
1 Søren Fauli - 110047
1 Nicholas Meyer - 1788
1 Jonathan Frakes - 2388
1 Stuart Baird - 2523
1 Richard Marquand - 19800
1 Jūzō Itami - 69167
1 Michael Boyle - 2754307
1 Russ Meyer - 4590Note that the list changed due to the different depth; apparently Robert Wise's Director credit is more than 5 entries into the list.
Perhaps you want to know which movies have fewer than 4 posters avaiable in Plex.
Script-specific variables in .env:
POSTER_THRESHOLD=10 # report items with fewer posters than this
- setup as above
- Run with
python list-low-poster-counts.py
Starting list-low-poster-counts 0.1.0 at 2023-12-07 09:35:45 connecting to https://plex.bing.bang... Loading Movies ... Completed loading 6171 of 6171 movie(s) from Movies on 15: 7 Plus Seven has 6 posters on 52: 21 Up has 4 posters on 77: 63 Up has 7 posters on 94: 1962 Halloween Massacre has 8 posters on 119: Ace in the Hole has 8 posters Low poster counts Movies |█ | ▇▇▅ 162/6171 [3%] in 9s (18.6/s, eta: 5:18)