Configure the Python SDK event dispatcher
The Optimizely Feature Experimentation SDKs make HTTP requests for every decision event or conversion event that gets triggered. Each SDK has a built-in event dispatcher for handling these events, but you should overriding it based on the specifics of your environment.
The Python SDK has an out-of-the-box synchronous dispatcher. You should customizing the event dispatcher you use in production to ensure that you queue and send events in a manner that scales to the volumes handled by your application. Customizing the event dispatcher lets you take advantage of features like batching, which makes it easier to handle large event volumes efficiently or to implement retry logic when a request fails. You can build your dispatcher from scratch or start with the provided dispatcher.
Important Performance risks with synchronous dispatchers
It is important to customize your event dispatcher when using an SDK for with a synchronous built-in event dispatcher (PHP, Ruby, and Python) to ensure that you can retrieve variations without waiting for the corresponding network request to return.
The examples show that to customize the event dispatcher, initialize the Optimizely client (or manager) with an event dispatcher instance.
from optimizely.event_dispatcher import EventDispatcher as event_dispatcher
from optimizely import optimizely
# Create an Optimizely client with the default event dispatcher
optimizely_client = optimizely.Optimizely(datafile,
event_dispatcher=event_dispatcher)The event dispatcher should implement a dispatchEvent function, which takes in three arguments: httpVerb, url, and params, all of which are created by the internal EventBuilder class. In this function, you should send a POST request to the given url using the params as the body of the request (be sure to stringify it to JSON) and {content-type: 'application/json'} in the headers.
Important If you are using a custom event dispatcher, do not modify the event payload returned from Optimizely. Modifying this payload will alter your results.
The Python SDK includes a basic synchronous event dispatcher. Examples of alternative approaches are available here: