lol-js
v2.0.2
Published
Node.js bindings for the Riot API, with caching and rate limiting
Downloads
35
Maintainers
Readme
lol.js is a client for fetching data from the Riot API for League of Legends. There are many node.js packages out there which give you access to the Riot API, but none has as complete a feature set as lol.js.
Note 2.x has breaking changes. See the changelog for details.
Features
- Support for the following APIs:
- game-v1.3
- lol-static-data-v1.2 (partial)
- match-v2.2
- matchlist-v2.2
- summoner-v1.4
- team-v2.4
- it's fairly easy to add other APIs - if you need something please raise an issue and I'll add it in for you, or send a Pull Request.
- Native promise support
- Built in support for rate limiting.
- Built in support for flexible caching with the caching engine of your choice. Built in support is available for Redis and an in-memory cache, however it is easy to write support for other databases if you wish.
Installation
npm install --save lol-js
Example Usage
With callbacks:
var lol = require('lol-js');
var lolClient = lol.client({
apiKey: 'blahblahblah',
cache: lol.redisCache({host: '127.0.0.1', port: 6379})
};
lolClient.getChampionById('na', 53, {champData: ['all']}, function(err, data) {
console.log("Found ", data.name);
lolClient.destroy();
});
With promises:
var lol = require('lol-js');
var lolClient = lol.client({
apiKey: 'blahblahblah',
cache: lol.redisCache({host: '127.0.0.1', port: 6379})
});
lolClient.getChampionById('na', 53, {champData: ['all']})
.then(function (data) {
console.log("Found ", data.name);
lolClient.destroy();
});
Creating a Client
The first step to using lol-js is to create a new client by calling lol.client()
. This function
takes a configuation object with the following options:
apiKey
- the API key assigned to you by Riot.cache
- a cache object ornull
to disable caching (see below).cacheTTL
- a{long, short, flex}
object which controls how long objects are cached for. Each is a value in seconds.long
applies to match objects, which don't change very often. Most objects are cached for theshort
duration. Default is one month forlong
and 5 minutes forshort
. Ifflex
is not null (defaults to one month) then all objects will be stored in the cache for theflex
TTL, however when an object is retrieved from the cache, we will still try to fetch a new copy of the object from Riot if the object is older than the short/long TTLs. If the Riot API is unavailable for some reason (the Riot API rather frequently returns 503) then the "expired" value will be used from the cache. Note that ifflex
is null then in such a case lol-js will retry the request a few times instead.rateLimit
- a list of limit objects. Each limit object is a{time, limit}
pair wheretime
is a duration in seconds andlimit
is the maximum number of requests to make in that duration. Defaults to[{time: 10, limit: 10}, {time: 600, limit: 500}]
.
Functions
You can browse the well-documented code in the src/api
folder for a complete description of all the functions you can call. Each file in src/api
is a
mixin object which is added to the client object's prototype. Note that the source files only
contain the promise version of the code - the callback versions are generated automatically.
Caching
lol-js will automatically cache results from Riot's API for you, allowing you to focus on using the data rather than worrying about rate limits and the Riot server going down.
The easiest way to cache objects is to use a built-in cache type. The following built in cache types exist:
lol.lruCache(options)
- Caches all objects in memory. Based on isaacs's lru-cache, and can take any options that theLRU
constructor can take.lol.redisCache({host, port, keyPrefix})
- Caches objects in Redis.host
andport
are the connection details for your redis server and default to'127.0.0.1'
and6379
, respectively.keyPrefix
is a prefix to prepend to all keys stored in Redis and defaults to'loljs-'
. You may want to read about using Redis as an LRU cache if you want to use Redis, or you can risk running out of memory.
If you don't want to use one of the built in cache types, you can easily specify your own caching
functions. The cache
parameter passed to a new client is a {set, get, destroy}
object, where
set(params, value)
stores an object in the cache, and get(params, callback)
is a function which retrieves an object from the cache and calls callback(err, value)
returning
the cached value or null
if the value is not available. In both cases, params is an object
consisting of:
params.key
- A string which uniquely identifies this object.params.ttl
- The suggested length, in seconds, to cache the object for.null
if the object can be cached forever.params.api
- A{name, version}
object (e.g.{name: 'match', version: 'v2.2'}
.)params.objectType
- The type of object being cached. This is always a string consisting of only letters.params.region
- The region used to make the request.params.params
- A hash of parameters which identify the object - this is different for each object type. For example, for a "summonerByName"objectType
, this will be a{summonerName}
object.
When calling cache.set()
, lol-js may pass in "none"
in place of a value to indicate the value
does not exist. cache.get()
should return "none"
if such a params object is passed in.
lol-js will always try to retrieve an object from the cache first, and will only make a REST
request if the cached object is not available.
Finally, if the cache defines a destroy()
function, this will be called when the lol-js client
is destroyed.
Development
If you're interested in contributing, check out the development docs.