Sunday 20th July, 2014
2:30pm to 3:00pm
*Session recording at VoiceRepublic*
When Elasticsearch decided to create an array of clients in multiple languages, one of the top-most priorities was having consistent features and experience across environments. In order to achieve that we faced several challenges, such as how to document the Elasticsearch API to avoid duplication of efforts and how to make sure that the end result actually matches the expectation.
We decided to solve this via documenting, in machine-readable format (JSON), the entire Elasticsearch API on the level required for the clients and creating a test suite which runs inside each clients' own tests.
In my talk I would like to show the reasoning behind these decisions as well as lessons learned - what was gained, what proved to be the most painful, and what were the surprises along the way. Everything discussed is open-source and available for review.
Django developer and enthusiast. Python developer for Elasticsearch. bio from Twitter
Sign in to add slides, notes or videos to this session