Mediadex

Index metadata about media libraries into ArangoDB and/or OpenSearch

Usage

Quick Start

mediadex --config /path/to/config.yaml

Config File

A minimal config file contains a search path and authentication for a backend.

Minimal ArangoDB Example

An example that indexes a single directory of movies into an ArangoDB database listening on localhost port 8259:

mediadex:
  paths:
    movies:
     - /path/to/movies
  backend:
    arangodb:
      auth:
        user: someUser
        pass: secretPassword

Minimal OpenSearch Example

An example indexing a single directory of music into an OpenSearch cluster listening on localhost port 9200:

mediadex:
  paths:
    music:
     - /path/to/music
  backend:
    opensearch:
      auth:
        user: someUser
        pass: secretPassword

Full Example

Both of the minimal examples assume that the backend is running on localhost, but alternate hosts and ports can be specified.

There are also settings for index and collection naming, and for tuning channels and go routines.

A complete list of configuration options:

mediadex:
  paths:
    movies:
      - /path/one
      - /path/two
    music:
      - /path/three
    series:
      - /path/four
      - /path/five
      - /path/six
  backends:
    arangodb:
      auth:
        insecure: false
        host: "arangodb.example.com"
        port: 8529
        user: service-user
        pass: service-pass
      backend:
        database_name: "cool-project"
        collection_prefix: "my-project-"
        collection_suffix: "-scope0"
        movies_collection: "Movies"
        music_collection: "Music"
        series_collection: "Series"
    opensearch:
      auth:
        insecure: false
        hosts:
          - "https://node1.example.com:9200"
          - "https://node2.example.com:9200"
          - "https://node2.example.com:9201"
        user: service-user
        pass: service-pass
      backend:
        index_prefix: "cool-project-"
        index_suffix: "-scope0"
        movie_index: "my-Movies"
        music_index: "my-Music"
        series_index: "my-Series"
        replicas: 2
        shards: 3
  threads:
    workers_per_path_type: 16
    channel_size_files: 256
    channel_size_backend: 384

TODO

• Real documentation
• Logging levels / verbosity
• Vaccuum (remove) non-existent files from backends
• Get metadata from various websites