Nuns Practicing Misanthropy

    firehoser

    1.3.6 • Public • Published

    Firehoser is no longer being maintained at this time. If someone would like to step in and take over the project please reply to Issue #11

    firehoser

    A friendly wrapper around AWS Kinesis Firehose with retry logic and custom queuing behavior. Get your data from node into S3 or Redshift in as easy a manner as possible.

    Please Note: Firehoser is written in es6 and needs node.js ≥ 6.0

    Installation

    yarn add firehoser or npm install firehoser

    Usage

    To use Firehoser, you need to create an instance of a DeliveryStream. There are several variations, but they all expose the same API: putRecord and it's pluralized counterpart putRecords.

    The basic DeliveryStream only requires a single argument, which is the name of the delivery stream:

    var AWS = require('aws-sdk');
    var firehoser = require('firehoser')
     
    AWS.config.update({
      accessKeyId: 'hardcoded-credentials',
      secretAccessKey: 'are-not-a-good-idea'
    });
     
    let firehose = new firehoser.DeliveryStream('my_delivery_stream_name');
     
    // Send a single record to Kinesis Firehose...
    firehose.putRecord('value1|value2|value3');

    putRecords behaves identically to putRecord except that the former accepts an Array of records.

    // Send multiple records...
    firehose.putRecords([
      'value1|value2|value3',
      'valueX|valueY|valueZ',
    ]);
     

    Both putRecord and putRecords return a Promise which resolve once the data has been successfully accepted by AWS. If the rate at which you call them exceeds the capacity of the delivery stream to which you are sending them, they will continue to retry the send until they succeed, and only then will the promise resolve.

    Amazon Kinesis Firehose limits the number of records you can send at a single time to 500. Firehoser automatically chunks your records into batches of 400 to stay well below this limit.

    A common use case for Firehose is to send JSON data into Redshift (or even just S3). This is easily accomplished by using a JSONDeliveryStream:

    let firehose = new firehoser.JSONDeliveryStream('my_delivery_stream_name');
     
    firehose.putRecord({
      jsonIs: 'a cool format',
      itMakes: 'pipe-separated values look so dated.'
    });
     
    firehose.putRecords([
      {a: 1},
      {a: 2},
      {a: 3}
    ]).then(()=>{
      console.log("Cool!");
    });

    Both DeliveryStream and JSONDeliveryStream will immediately send data to AWS Kinesis Firehose as soon as (and as often as) you call putRecord. Sometimes what you want instead is to queue up records as they are generated, and only send them to AWS Kinesis Firehose once a certain number of them have been gathered, or a certain time limit has passed. For these cases, we've created QueueableDeliveryStream and QueueableJSONDeliveryStream. Simply pass in a cutoff value for the number of records, or the staleness of records that you're comfortable with, and we'll wait until one of those limits is passed before sending off the data.

    let maxDelay = 15000; // milliseconds
    let maxQueued = 100;
    let firehose = new firehoser.QueueableJSONDeliveryStream(
      'my_delivery_stream_name',
      maxDelay,
      maxQueued
    );
     
    firehose.putRecord({
     id: 'rec1'
    }).then(()=>{
      console.log('rec1 sent to AWS!');
    });
     
    firehose.putRecord({
      id: 'rec2'
    }).then(()=>{
      console.log('rec2 sent to AWS!');
    });
     
    //  ... wait ~15 seconds and see both console messages fire at the same time!

    Misc

    Take off, ya hoser!

    Install

    npm i firehoser

    DownloadsWeekly Downloads

    950

    Version

    1.3.6

    License

    MIT

    Unpacked Size

    55.3 kB

    Total Files

    6

    Last publish

    Collaborators

    • alec.troemel
    • donspaulding
    • eyepulp