RudderStack Transformations let you transform any data that you send through RudderStack. You can use them across Event Stream, Warehouse Actions, and Cloud Extract pipelines. They are easy to build, debug, and manage; and they run in real-time, after data collection and before it is delivered to your destination. Simply put, they're data transformations that work like you think they should.
Although Transformations is one of the most popular RudderStack features, its API wasn't accessible to users -- until now. We are pleased to release the RudderStack Transformations API. With it, you can programmatically add and remove transformations to the platform, create an organization-wide sandbox where your team can store transformations before publishing them to the live environment, define libraries, and version control your transformations.
The Transformations API allows you to perform various operations on your transformations and libraries via HTTP API calls. The API is integrated with the RudderStack UI so that you can easily access and manage your transformations and libraries. You can even publish multiple transformations in a single operation, functionality that is not yet available on the RudderStack UI.
The Transformations API is a RESTful API. It lets you create and use a transformation and get a JSON object as a response. You can leverage the standard HTTP response codes to determine the success or failure of your API requests.
Some other key features of the Transformations API include:
- Lets you version control all your transformations so that every change in a transformation creates a new version.
- Lets you store your transformations in an organization-wide sandbox before publishing them.
- Authenticates via Basic HTTP Authentication mechanism.
- Checks for successful compilation and execution of new transformations.
The Transformations API allows you to perform the standard CRUD (Create, Retrieve, Update, Delete) operations on your transformations.
Typically, you can create a transformation and get a JSON object as a response. Updating the transformation with any changes creates a new version or revision. The API also offers an optional parameter (
publish) for these operations, which allows you to publish the latest version of the transformation and make the code live for the incoming event traffic.
The API assigns an ID to every published transformation. You can leverage this ID to retrieve, update or delete a transformation. Also, you can get all the revisions made to that transformation.
The API creates a new version of the transformation or the library with every change, allowing you to track the whole lineage of that transformation/library.
The way the Transformations API implements version control is quite interesting. Let's say you create a transformation. Any update or change in it causes RudderStack to save the older version of the transformation as a revision. You can make multiple such changes to the transformation, and the API will record every such revision, while the current version is always available for you to publish and use. The API also allows you to roll back and use any specific version of the transformation you want to publish via its Publish API.
The Transformation API exposes libraries for a better development workflow for your transformations. They give you the flexibility to reuse and maintain different versions of your transformation code while sparing you manual, repetitive coding work.
Note: You can apply all the CRUD operations mentioned in the above section to the libraries as well.
For more information and technical details on using the RudderStack Transformations API, check out the docs. We've also built some handy transformation templates that you can use to create your transformations. Find them in our GitHub repository.
Test out our event stream, ELT, and reverse-ETL pipelines. Use our HTTP source to send data in less than 5 minutes, or install one of our 12 SDKs in your website or app. Get started.