IGroceryStore

Loosely coupled monolith app where users will track personal basket price.

Projects Structure:

Main features:

• Watching product prices over time
• Comparing basket price across stores
• Searching for allergens free products
• Rating products and stores
• Searching for similar products

Modules:

Basket

• Event Store for storing historical data about products prices and promotions
• MongoDb for storing projections and users baskets

Available actions:

• Create Basket
• Consumers for product creation events and price changes events

Users

• Auth with JWT
• Refresh tokens handling (in future I'll move storing tokens to Redis or integrate project with OpenId)
• BCrypt for hashing passwords
• Postgres for storing users (I'm considering changing it to noSql - mongoDb, casandra or dynamodb)

Available actions:

• Register
• Login
• Refresh Token
• GetAll/GetUser

Shops

• Storing shops location and current price of product
• calculating best basket price in shops near me (coming soon)

Available actions:

• Report new product price

Products

• Postgres for relational data
• Module publishes events for other modules about new products
• I'm considering using apache lucene to enable searching for similar products

Available actions:

• Create Product
• Mark Product As Obsolete
• Add Allergens To Product
• CRUD for Categories
• CRUD for Allergens

Stack

.Net7.0 C#11 with Minimal API
Docker - containerization for tests aln local development
Terraform - for infrastructure as code (IaC)
Event Store - storing historical data of products prices
MongoDb - storing projections from historical data
DynamoDb - storing data from shop module
AWS Systems Manager - runtime configuration
JWT.Net - authorisation and authentication
BCrypt.net - password hashing
GitHubAction - CI/CD
RabbitMQ - asynchronous messaging
MassTransit - asynchronous messaging library
PostgreSQL - database with json support
Entity Framework Core - ORM used for product module
OpenTelemetry - Traces, Metrics
Jaeger - Exporter/UI for OpenTelemetry

For Contributors

Git Branching Strategy

I'm using Feature Branch Workflow for simplicity and
I encourage you to use it also in your fork.
You can read more about flow here

Requirements

• docker
• terraform (optional)
• aws account (optional)

How to run

• From the tools/docker directory run the command:
  docker compose --profiles webapi,jaeger up -d (look at profiles section)
• if you want to run shop module you can create all required resources with terraform from tools/terraform/dev directory
  terraform init
terraform apply
• Run ASP and/or Worker Project (add compound configuration in Rider)
• API endpoints: localhost:5000/swagger
• All ports for infrastructure available in docker-compose file
• To connect to pgadmin from docker use:
  docker run -p 5050:80 -e "PGADMIN_DEFAULT_EMAIL=name@example.com" -e "PGADMIN_DEFAULT_PASSWORD=admin" -d dpage/pgadmin4

docker compose profiles (optional):

• all - all containers
• webapi - all except observability
• shops - dynamodb
• users - postgres
• baskets - eventstore, mongodb
• products - postgres
• logs - elasticsearch, kibana
• kibana
• jaeger
• elastic

rabbitmq and redis will run by default