TeetoJS

TeetoJS is a lightweight Node.js Riot API wrapper, ready for multi-instances szenarios with the help of redis. It has around 300 lines of code only, offers priority queuing of API requests, an integrated rate-limiter and is easy to customize.

Installation

npm install teetojs --save

Usage

const TeetoJS = require("teetojs");
let api = TeetoJS("RGAPI-KEY-HERE", {
  redis: {
    host: "YOUR-HOSTNAME",
    port: 6379, // or whatever port you have
    db: 1,
    password: "YOUR-PASSWORD"
  }
});

api
  .get("euw", "summoner.getBySummonerName", encodeURIComponent("TheCraccer"))
  .then(res => console.log(res))
  .catch(err => console.log(err));

api
  .get("na1", "match.getMatchlist", 0, 78247)
  .then(res => console.log(res))
  .catch(err => console.log(err));

All requests are done via .get(...).

• The first argument is the region.
• The second is the endpoint path (see config.json).
• The third argument is the priority of the request. 1 for high, 0 for normal
• Then come any path arguments (usually zero or one, or two for getChampionMastery) which are for summoner/match IDs, names, etc.
• Last is an optional object for any query parameters. (Important: You should encode names when searching for summoners, because some characters need to be encoded to be found from riot)

Configuration

To customize the wrapper TeetoJS takes options as second argument which is a configuration object. config.json is used by default. The supplied option object will override any corresponding values in the config. The configuration specifies the number of retries, backOffTimings, loggings, endpoints, your redis config and everything else. You should at least give your valid redis config to make the wrapper work.

config Object

• debug [boolean]: Weather to show debug outputs or not
• namespacePrefix [string]: Add a prefix if you have multiple limits running that need to be counted seperately. NOTE: Dev and Prod environments are already seperated via environment variables
• showWarn [boolean]: Weather to console.error occuring 429 errors or not
• exceededCallback [func(err)]: Callback to be called if a 429 error appeared. Parameter is an error object containing headers and url
• applimit [array]: Your apiKey application limits, e.g. ["20:1","100:120"]
• spreadToSlowest [boolean]: All requests are spreaded according to your slowest limit. Set this value to false to only respect your smaller limits (e.g. your 10s limit). Recommendation: Use true for production
• edgeCaseFixValue [number]: Since Riot resets it's limits by their own time there is a small edge case where requests are resetted in the wrapper but not at riot. This value slows down your requests, so that this does not happen to often. A value of 1.1 means that you are running at 90% speed. WARNING: Setting this value very low might cause 429-errors.
• maxRetriesAmnt [number]: Number of retries after getting a 429, 500 or 503 error before aborting the request
• retryMS [number]: Default time before retrying a failed request. This value only applies to 429 retries if retry-after is not given in the riot response

Rate-Limiting

To rate limit all your requests accross multiple instances, this wrapper is using a modified version of the rolling-rate-limiter from classdojo.