Jellyfin renamer megascript

a true pinnacle of engineering

place all media files in the same directory, should look like:

/path/to/media/
    movie1 (year)/
        movie1.ext
    movie2/
        movie2.ext
        movie2.srt

or for tv shows:

/path/to/media/
    show1 (year)/
        season 01/
            show1 - S01E01.ext
            show1 - S01E02.ext
        season 02/
            subs/
                S02E01.en.ext
                S02E02.spanish.ext
            show1 - S02E01-02.ext
            show1 - S02E02.ext
    show2/
        show2 - S01E01.ext
        show2 - S01E02.ext
        show2 - S02E01-02.ext

the format is very flexible, and should work with most naming conventions subtitles can also all be in a single folder, as long as they are grouped by the season and contain the season/episode number in the full path

as long as the show/movie name is first in the path, and the path contains the season/episode number, this should work out fine.

a TMDB api key is required, and can be set in the environment variable TMDB_API_KEY, or in a file, with the path set in the environment variable TMDB_API_KEY_FILE (defaults to ./.tmdb-api-key)

media resolution will be inferred from the path, or using ffprobe if available

i can't exactly guarantee that this will work, since its only tested on my own (small-ish, 7tb) library, which was already pretty well organized. (the only real reason i wrote this was to add [tmdbid=xyz] to the filenames)

i heavily recommend running this with the --dry-run flag first, which creates a "fake" directory structure, where the media and subtitle files are text files containing some metadata. you should probably also backup your media files before running this, (i'd recommend running zfs, and taking a snapshot)

you can use the --no-interact flag to automatically select the first result from TMDB, if there are multiple matches

then run like so:

1. python3 jellyfin-renamer.py movie --dry-run /path/to/media media_out

   IF you already have a /path/to/media_out directory, name it something else

   IF it is tv shows, use "show" instead of "movie"

2. "media_out" is now created in the parent directory of /path/to/media, and contains the new directory structure

3. verify it looks good

   if a movie/show is missing the [tmdbid=xyz] tag but exists on TMDB, means it wasnt found. the name is either misspelled, or the year is incorrect.

   if something is wrong, delete the media_out directory and make necessary adjustments to the media directory

4. delete the newly created media_out directory

5. run the script again, without the --dry-run flag

6. files are now moved. if it failed, you should restore from the backup you made earlier and either fix my shitty script or do it manually

Requires:

(probably) only works on linux

• python3
• python-requests
• python-Levenshtein
• ffprobe (optional)

TODO:

• add feature to use metadata from dry run, since library scans are quite s l o w
• rewrite the super shit path parser
• make it idempotent