beingayush / fampay-project

Readme

Tech Stack

• Fast API - Python web framework
• AWS DynamoDB - NoSQL Database
• AWS Elastic search - for searching
• Docker

Setup Instructions

• Create a dynamo db table called "Youtube"
• Create an elastic search domain on AWS ElasticSearch
• Create a lambda trigger from the dynamo db table to stream data to elastic search
• The blueprint for the lambda trigger is in the file app/lambda.py
• All the environment variables like region, elastic search domain need to be set for the lambda
• The lambda should be given all access(read/write/list) to dynamo db and to the elastic search domain
• Create a google developer api key with access to Youtube V3 api and insert it in the API_KEY variable in the file recurring_task.py
• The AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY need to be entered in the app/.env file
• The elastic search domain without https:// (starting with search-XXXXX.ap-south-1.es.amazonaws.com) should be entered in main.py in the variable named host
• After everything is setup, run the application from the root directory using the following command

docker-compose up -d

• Swagger UI can be now accessed at https://localhost:80/docs
• The list/search api can be tested through swagger UI

API

• There is a unified api to do both the get and search operations
• The search implemented is very efficient (done through elastic search) and paginated
• The service is highly scalable and optimised as dynamo db has a very low latency for reads and writes, elastic search is highly optimised for searching
• The service has relevant validations for various request parameters