g-kartik / pratilipi-demo

Main repo for pratilipi demo

Pratilipi Demo

Main repo for pratilipi demo for the Microservices assignment

Instructions

• This repo contains three submodules namely, content, user and user interaction. Use the command git clone --recurse-submodules <GIT_URL> to clone the repo including its submodules.
• The env folder contains .env files for both development and production environment following the 12-factor app principle.
• The docker-compose files for production and development evironment are docker-compose.yml and docker-compose_dev.yml respectively.
• Run the commnad docker-compose up -d to build and run all the containers in production environment. Dockerfile at the root directory of the repo will be used to build the base image and the Dockerfile present in the each of the service directory will inherit from this base image.
• In your localhost, the content, user and the user_interaction services are available at port 3000, 8000, and 5000 respectively
• The swagger and redoc pages for each of the services are available at /swagger and /redoc respectively. You can use them to test the api or use the following Postman workspace for the same.
  
• Three test users are already seeded into the database, with user_id, 1, 2, and 3.
• For the data ingestion of books content use the , you should upload a csv file having title, story, date_published and user_id fields. The user_id entered into the file must be registered in the database, else the operation would fail.
• The content repo has a test data file at path content/tests/test_content.csv. You can use it as well.
• The data ingestion will be queued to the redis queue and will return job_id. You can check its status via books/status using the job_id.
• The content list api will be sorted by the user interaction which includes sorting by number of likes followed by number of reads. The books which don't have any interaction will be sorted by their primary key.

Development and production environment

In development environment the service container uses host network mode to communicate with each other. Which means there is no distinction between container and localhost ports. Whereas in production environment, the container talk to each other via the default bridge network created by the docker with their DNS being resolved by their service names. ALLOWED_HOSTS settings of Django is set according to network mode being used. If it is host mode, then localhost and 127.0.0.1 are listed else the name of the services are listed which will be resolved by Docker at runtime. Same thing is followed for POSTGRES_HOST settings.