bhanuprasadcherukuvada / long-tail-assgn

Steps for Execution

• You need have docker installed on your machine to run the project

• clone the project using the given command

  git  clone https://github.com/bhanuprasadcherukuvada/long-tail-assgn.git

•  docker compose up --build -d 

• To build and run the services

• The docker compose file runs postgres, hasura-engine, nextjs frontend in detached mode

• frontend application would be deployed at http://localhost:3000

• You can check the hasura console at http://localhost:8080

- the application is seeded with initial data which is done with the help of migrations. - I have used ```hasura-cli``` to generate migrations. - i used this [resource](https://hasura.io/blog/moving-from-local-development-staging-production-with-hasura/) for seed migrations

Adding remote Schema

• I have used 2nd use case in the given blog.
• Run hasura metadata apply to update the metadata which includes config regarding remote schema after adding remote schema using hasura console.

How it works ?

• we have a long_tails table in the postgres database
• long_tails have tail and json_id in the table
• json_id is used as a join key with a remote schema
• In backend we run a apollo server which has a post_info type which is added to long_tails table with json_id as join key
• With this we can fetch title and description with remote join in hasura.
• We can fetch both json_id and title and description in single fetch corresponding tail
• The advantage here is that we are avoiding mutiple calls and reducing network time
• We have json file in backend so that we can make updates to the json file and they will be reflected in the frontend on page reload

Link to the deployement

• https://long-tail-assgn.vercel.app/

• For the purpose of deployement, I used vercel platform for deploying nextjs

• For hasura graphql engine,I used hasura cloud

• For remote Schema I deployed it on glitch

Alternative deployement

• we can use ECS service for deploying the docker compose services directly