keen-csv.js

Takes a Keen response from keen-analysis.js and converts it to a multiline CSV string.

Installation

Install this package from npm:

$ npm install keen-csv --save

Or add it directly to your package.json

{
  "dependencies": {
    "keen-csv": "^1.0.0",
  }
}

Usage

Perform a Keen query like normal, then you can run the response directly through this library in a couple of ways:

By direct instantiation

import Keen from 'keen-analysis';
var KeenCSV = require('keen-csv');

const client = new Keen({
  projectId: 'YOUR_PROJECT_ID',
  readKey: 'YOUR_READ_KEY'
});

client
  .query('count', {
    event_collection: 'pageviews',
    timeframe: 'this_14_days',
    interval: 'daily',
    group_by: 'page_info.path'
  })
  .then(res => {
    let keenToCSV = new KeenCSV(res, {
      delimiter: '&'
    });
    let csvResults = keenToCSV.generateCSV();

    // Handle CSV results
  })
  .catch(err => {
    // Handle errors
  });

As a link in a Promise chain:

import Keen from 'keen-analysis';
var KeenCSV = require('keen-csv');

const client = new Keen({
  projectId: 'YOUR_PROJECT_ID',
  readKey: 'YOUR_READ_KEY'
});

client
  .query('count', {
    event_collection: 'pageviews',
    timeframe: 'this_14_days',
    interval: 'daily',
    group_by: 'page_info.path'
  })
  .then(KeenCSV.convert)
  .then(res => {
    // Handle CSV results
  })
  .catch(err => {
    // Handle errors
  });

Or, if you'd like to include options in the Promise Chain method:

import Keen from 'keen-analysis';
var KeenCSV = require('keen-csv');

const client = new Keen({
  projectId: 'YOUR_PROJECT_ID',
  readKey: 'YOUR_READ_KEY'
});

client
  .query('count', {
    event_collection: 'pageviews',
    timeframe: 'this_14_days',
    interval: 'daily',
    group_by: 'page_info.path'
  })
  .then(KeenCSV.convertWithOptions({
    delimiter: '&'
  }))
  .then(res => {
    // Handle CSV results
  })
  .catch(err => {
    // Handle errors
  });

Options

• delimiter: Use this character rather than a comma
• delimiterSub: If we encounter any delimiter characters, we'll substitute them for this
• nestedDelimiter: The nature of a Keen response sometimes entails nested objects. In these cases we'll flatten the keys using this character/string
• filteredColumns: An array of column headers to filter out of the final CSV results