v3.1

New feature:

The Local/Offline Queries - Ability to run a query in your browser (or on your node.js server) without making another request to the API.

Use extraction query to get the data, store the data in the cache or in a CSV file. Then you can run as many queries as you want, without worrying about the costs or the response times.

Note: this is an experimental feature. In general, you should rely on the Keen's API, especially cached queries and cached datasets.

PS: it's 10x faster on the Node.js env

v3.0.2

New

• Requests in Node.js server environment are now returning a Promise. Callbacks are still working, so it's up to you which coding style you choose.

Fixed

• Error handling for Promises on the browser-side clients.

v3.0.0

New

• Browser-side Caching

client
  .query({
    analysis_type: 'count',
    event_collection: 'pageviews',
    timeframe: 'this_31_days',
    cache: {
      maxAge: 10 * 1000 // cache the result for 10 seconds
    }
  })
  .then(res => {
    // Handle results
  })
  .catch(err => {
    // Handle errors
  });

• Shorter syntax. As you may have noticed we've changed our syntax to make it more clear. We're still supporting older versions of methods, so there is a full backward compatibility.

Fetch API

NEW:

• Fetch API is now the default method of making requests.
• Lighter Promise polyfill instead of Bluebird.
• Ability to .abort() queries

Move from Gulp to Webpack / UMD

FIXED:

• Keen-Analysis.js is now a valid ES6 module (UMD)
• Root.location.protocol bug is now fixed (ios/android emulators) (#22)

v1.3.0

NEW:

Support for Cached Datasets

• This patch installs keen-core@0.2.0 and updates the client.query() method to support fetching results from Cached Datasets (#21).

FIXED:

• Syntax errors caused by a malformed JSON response are now caught properly (#22).
• Successful delete responses (204s) are no longer handled as errors (#14).