This documentation gives clear idea of how the nLearn application is keeping track of all the events.

It mainly has followings:

• AnalyticsModule
• AnalyticsTracker
• AnalyticsHashmap
• (XYZ)Analytics

This module takes in all the basic component from the App Component and then eventually provides the instance of AnalyticsTracker which will further be used by (XYZ)Analytics to call the methods available in AnalyticsTracker itself. The methods in AnalyticsTracker that will be used by (XYZ)Analytics will further use AnalyticsHashmap, which will provide a HashMap with basic OS-level attributes already added to it as a key-value pair. Inside AnalyticsTracker, methods will further add more attribute to the available HashMap to make it ready to lauch as an event using AnalyticsClient.

For Analytics events, we're using Amazon Web Service SDK

Now, let's discuss about the individual player of the ecosystem.

AnalyticsModule is like any other module which is used to provide instance to be used by Dagger @inject annotation. It uses CognitoPoolID, Region, CognitoCachingCredentialsProvider and PinpointManager to eventually provide the instance of AnalyticsTracker at runtime.

(XYZ)Analytics is just a common name of any custom-analytics class for any Activity/Fragment or any other View, XYZ will be replaced by the specific name of the View. Ex:- PracticeAnalytics etc.

(XYZ)Analytics uses Singleton design pattern to create only one instance throughout.

AnalyticsHashMap class does nothing but creates a HashMap and adds certain OS-level attributes like PLATFORM, APP_VERSION, NETWORK_TYPE, OS_VERSION, DEVICE_ID(which is a unique session ID from Settings.Secure) and TIMESTAMP.

AnalyticsTracker takes in all the attributes from AnalyticsHashMap and (XYZ)Analytics and then adds some more preference-level attribute(like STUDENT_ID, ADMISSION_NO, UUID, EVENT_GROUP) to the resultant HashMap to submit the event instance that was created as shown using PinpointManager.