Skip to content

Latest commit

 

History

History
160 lines (104 loc) · 5.99 KB

PLUGINS.md

File metadata and controls

160 lines (104 loc) · 5.99 KB
Raw

Plugin interface

Base plugin

All plugins should be a ES6 JS class, and are required to have a metadata object that gets passed to the super(metadata, defaultSettings). The basis for a plugin should look similar to the following:

const Plugin = require('../../plugin_handler/base');

class ShowRssSource extends Plugin {
  constructor() {
    const metadata = {
      pluginId: 'info_showrss',                 // Unique ID of plugin
      pluginName: 'ShowRss',                    // Display name of plugin
      pluginTypes: ['source', 'series'],         // 'source', 'downloader', 'player'
      sourceType: 'continuous',                 // 'adhoc', 'continuous'
      requires: ['com_flexget', 'com_transmissionbt'],  // this plugin requires the flexget plugin
      link: 'http://showrss.info/',             // Link to provider site
      description: 'Updated feed of TV shows'   // Description of plugin provider
    };
  
    const defaultSettings = {
      enabled: true
    };
  
    super(metadata, defaultSettings);
  }
}

PluginTypes

A plugin can be a combination of any pluginTypes as long as they meet all of the requirements described below, and in the Required plugin-specific functions section.

The current possible values are as follows:

  • trending - intended to return info about what is currently trending in the world
  • source - goes out to a external source and runs a media search against that source
  • downloader - a plugin that will provide a method to get data from a source
  • player - a media player used to consume the media

SourceTypes

If the pluginType is set to source, you can also provide the additional metadata about how the data should be retrieved:

  • adhoc - one time media download
  • continuous - this is a source that you can search upon, and when a download is requested, will download over time using the series endpoints

Provided helpers

There are a few things that each plugin object gets when it's created, in order to easily access these global constructs:

this.events // reference to global event emitter for the entire api. events described below
this.config // reference to global config
this.logger // usable logger with `info`, `debug`, `error`, etc methods

Events

Currently defined events in kapture:

  • continuous:added - a continuous source has had a new entry added. entry detail in first argument

Not yet added:

  • continuous:removed -
  • search:query -
  • download:started -
  • download:completed -
  • download:removed -

Required plugin-specific functions

All data methods described below should return a Promise with the resultant data

Search source plugin

Metadata

Must have:

  • pluginTypes: source must be present
  • sourceTypes: can be continuous or adhoc

Functions

  • search(query) - searches the source repository for string query and returns a collection of results that the source provides.

Output

{}

Downloader plugins

The downloader plugins are a bit unique to the system, in that they need to store data about the current state of their downloads somewhere. The pluginStateStore is provided for that purpose. See the youtube plugin for an example of how that is used.

Metadata

Must have:

  • pluginType contains downloader (either array or string)
  • downloadProviders must contain a string that this plugin can perform downloads for

Functions

  • downloadSlug(slug) - starts a new download of given slug identifier (usually provided by the search results defined by a source plugin). Returns a Promise with success or failure to add that download. Do not block, use the status function to report progress. This will be called via the downloads/method:{methodId}/{slug} approach.
  • downloadId(id) - starts a new download of a source-defined id. this will be called by the downloads/source:{sourceId}/{id} method
  • status() - returns a Promise with the data (an array) about all of the current and active downloads.
  • status(id) - provides info about a specific download ID (id defined by the plugin itself - could be either the id of the search result, or some hash as long as it's unique).
  • removeId(id, deleteFromDisk) - stops active downloads, or removes complete ones. In the case of deleteFromDisk boolean is set - this function should also remove the file from disk
  • removeSlug(slug, deleteFromDisk) -

Output

Trending plugins

Metadata

Must have the trending sourceType

Functions

The plugin must also provide the following functions:

  • trending() - provides a kapture-formatted object (see below) that will be aggregated with other providers in response to query
  • trendingInfo(id) - provides information about a specific entry (denoted by trending object field id)

Output

The standard trending json object should look something like this (written in yaml for easy reading):

series:
  - score: 68               # how popular on a scale of 1-100
    sourceId: 'trakt'       # id of source provider
    id: 1611                # id of media entry (specific to source provider)
    type: 'series'          # kapture-media type
    title: 'Supernatural'   # display name
    slug: 'supernatural'    # slug used for whatever is desired
    additionalInfoUrl: 'https://trakt.tv/shows/supernatural'
movies:
  # ...
# other kapture-media-types

Url downloader plugins

Certain sites may need their own approach to downloading content from the site (ie youtube). As a result, the com_kapture_url plugin provides a method for handling a http.* url in the q string to search. It searches other plugins for the functions below and if found, will pass to the url function of that plugin

Functions

  • urlMatches(url) - this is a function that returns true/false depending on if the url passed can handle the url given
  • url(url) - will take a url and start a download on it. this may just call the download function for ease of use