JSON Metadata Agent for Plex

A metadata agent for Plex that reads from JSON files co-located with your media.

Why?

Not all metadata is easily accessed online via an HTML web page. There are some websites for which it is impossible to write a traditional Plex metadata agent — for example, those that:

• require username and password authentication, possibly even captcha
• hide information behind javascript
• detect the automated nature of the metadata agent, throttling or blocking scrapers
• perform poorly, crash and timeout
• frequently change their HTML structure

This agent doesn't collect any metadata itself — it simply loads metadata from JSON files found with your media into Plex.

It's designed to work alongside other tools and methods of collecting metadata, be it a custom scraper (it doesn't even have to be written in Python), browser plugin, command line tool or GUI. You could even edit the files by hand :)

Media Preparation

At the time of writing, JSON metadata is supported for movies only.

To define metadata for a movie, a JSON file named exactly Info.json must be present in the same directory as your movie file(s). For example:

Movies
  |- Spaceballs (1987)
      |- Spaceballs.mp4
      |- Info.json
      |- Cover.jpg
      |- Background.jpg

This means you are limited to a single movie and Info.json file per directory.

Example JSON

The structure of the Info.json file follows as closely as possible that of the Movie model defined by Plex itself (although it's basically undocumented). It should look something like this:

{
    "title": "Spaceballs",
    "original_title": "Spaceballs",
    "summary": "Planet Spaceballs' President Skroob sends Lord Dark Helmet to steal planet Druidia's abundant supply of air to replenish their own, and only Lone Starr can stop them.",
    "year": 1987,
    "originally_available_at": "1987-06-24",
    "rating": 6.6,
    "content_rating": "PG",
    "studio": "STUDIO NAME",
    "duration": 96,
    "roles": [
      {
        "name" : "Mel Brooks",
        "role": "President Skroob/Yoghurt",
        "photo": "https://image.tmdb.org/t/p/w264_and_h264_bestv2/ndFo3LOYNCUghQTK833N1Wtuynr.jpg"
      },
      {
        "name" : "Rick Moranis",
        "role": "Dark Helmet",
        "photo": "https://image.tmdb.org/t/p/w264_and_h264_bestv2/27dRb7OyRGzQP8D4tzyY6dEdmQY.jpg"
      },
      {
        "name" : "Bill Pullman",
        "role": "Lone Starr",
        "photo": "https://image.tmdb.org/t/p/w264_and_h264_bestv2/pIpTEQVbDif8m8OdjAxQKNCj0D6.jpg"
      }
    ],
    "genres": [
      "Adventure",
      "Comedy",
      "Sci-Fi"
    ],
    "collections": [
      "android",
      "lasergun",
      "swordplay",
      "temple",
      "space marine",
      "space battle",
      "space travel",
      "space mission",
      "galaxy",
      "comb",
      "altar",
      "magnet beam",
      "jam",
      "speed of light",
      "plastic surgery",
      "password",
      "shield",
      "nose",
      "spoof"
    ]
}

Example Artoo Scraper

If you're like me you hate manually building the info.json file manually. That's where Artoo comes in. By using the follwing structure, it's pretty easy to create a custom scraper to build your Info.json file automatically

artoo.scrape('SELECTOR', {
  title: {sel: 'SELECTOR'},
  original_title: {sel: 'SELECTOR'},
  summary: {sel: 'SELECTOR'},
  year: {sel: 'SELECTOR'},
  originally_available_at: {sel: 'SELECTOR'},
  rating: {sel: 'SELECTOR'},
  content_rating: {sel: 'SELECTOR'},
  studio: {sel: 'SELECTOR'},
  duration: {sel: 'SELECTOR'},
  roles: {
    scrape: {
      iterator: 'SELECTOR',
      data: {
        name: {sel: 'SELECTOR'},
        role: {sel: 'SELECTOR'},
        photo: {sel: 'SELECTOR' attr: 'href'}
      }
    }
  },
  genres: {
    scrape: {
    iterator: 'SELECTOR',
    data: 'text'
    }
  },
  collections: {
    scrape: {
    iterator: 'SELECTOR',
    data: 'text'
    }
  }
}, artoo.savePrettyJson(data, [Info.json]));