SB-29870: Lag on User Enrollment List sorting by lastAccessdate on the continue learning section

Overview:

Currently Lag on sorting by lastAccessdate on the continue learning section appears in the portal. The reason behind it is that the enrollment list API uses default cache. This cache only updates with the same API when enrollment data is empty. Hence after a few mins the cache gets cleared and reloads while the API is consumed

Pr-requisites:

User is a logged in user and has enrolled into at-least one traceable collection

Linked Issue:

https://project-sunbird.atlassian.net/browse/SB-29083

Solutions:

Consuming the user enrollment list API with disabling cache only when after consuming content state API Or Disabling cache from course service

2. On updating user content state API update the existing cache

Option 1: Disabling cache for User’s enrollment :

Disabling cache needs to be done from course service by setting below statement in CourseEnrolmentActor as false

if (isCacheEnabled && request.getContext.get("cache").asInstanceOf[Boolean])

Steps to follow:

Set user_enrolments_response_cache_enable to false
portal to send request.getContext.get("cache").asInstanceOf[Boolean] as false

Pros:

Faster implementation with only configuration changes

Cons:

The API performance can go down

Option 2: On updating user content state API update the existing cache

Current cache implementation reloads the cache whenever the cache is cleared. Now the cache will be required to update/reloaded when the update content state API is triggered..

Steps to follow :

Update cache

Prepare the map to add in cache:
1. Add enrolment record all fields  from user_enrolments table 
                by userId, courseId, batchId, contentId from the request
2. Add batch details all fields  by CourseId, batchId from CourseBatch index 
               as batch object 
3. Add course details fields from response of  composite search with courseId
       Details : COURSE_NAME,DESCRIPTION,LEAF_NODE_COUNT,COURSE_LOGO_URL,
       CONTENT_ID,COLLECTION_ID,CONTENT
4. Update completionPercentage and status     
Update this map to cache for key request UserId first index

Reload Cache

Pros:

Permanent solution, no portal changes
No performance degradation, as the operation can be performed asynchronously

Cons:

Complexity in implementation