Table of Contents

• 1. Overview Architect
• 2. Local setup
• 3. CURL command for REST API
• 4. Run Jest test
• 5. Source structure
  • 5.1. Run eslint in every service folder
  • 5.2. Structure
  • 5.3. Library Usage
  • 5.4. Database
• 6. Sequence diagram in some use cases

1. Overview Architect

• orchestrator-service: It gathers all APIs from others product-service, search-service, checkout-service. It will be mainly used as endpoints for users or frontend services. This service is also a producer to receive mesage and send checkout data message to Message MQ. The checkout-service as consumer, it will read messages from Rabbit MQ, they are persisted into MongoDB.
• search-service: It provides search and sort functionality for products.
• product-service: Service relates to add and delete products.
• checkout-service: It is relevant to payment and checkout.
• service-registry: It is central service and help to keep track of other services such as register or unregister once we start or stop service.

⬆️ Back to Top

2. Local setup

• Start MongoDB docker-compose

cd etc/mongo-docker/
# Create and start containers
docker-compose up -d --build

# Display services
docker-compose ps

# Stop services
docker-compose stop

# Start services
docker-compose start

# Stop and remove containers, networks, images, and volumes
docker-compose down -v

• Use Robo 3T to accesss MongoDB with defaut db name and account

db.createUser(
{
	user: "admin",
	pwd: "admin",
	roles: [
		{
			role: "readWrite",
			db: "my-mongo-db"
		}
	]
}
)

• Start RabbitMQ docker

docker run -it --rm --name rabbitmq -p 5672:5672 -p 15672:15672 rabbitmq:3-management

• Start service-registry

cd service-registry
npm run start

• Start search-service

cd search-service
npm run start

• Start product-service

cd product-service
npm run start

• Start checkout-service

cd checkout-service
npm run start

• Start orchestrator-service

cd orchestrator-service
npm run start

⬆️ Back to Top

3. CURL command for REST API

• Insert products into Product table

curl -X POST --header "Content-Type: application/json" --header "Accept: application/json" -d @etc/product_insert.json "http://localhost:3080/products"

• List all products

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/products' | jq .

• Search products by name

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/products?name=name1&sort_by=name' | jq .

• Search products by name with operator

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/products?name=eq:name1' | jq .

[
  {
    "_id": "609de68be9d8da4a94453107",
    "name": "name1",
    "price": 2,
    "branch": "Branch Test",
    "color": "Testing",
    "__v": 0
  }
]

• Search products by name with operator and sort by price on ascending order

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/products?price=gt:1&sort_by=-price' | jq .

• Delete products

curl -X DELETE --header "Content-Type: application/json" --header "Accept: application/json" -d @etc/product_delete.json "http://localhost:3080/products"  | jq .

• Send data checkout to Message MQ, data will be populated and upserted into Payments table

curl -X POST --header "Content-Type: application/json" --header "Accept: application/json" -d @etc/checkout_send.json "http://localhost:3080/checkout" | jq .

• Search all payments

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/payments' | jq .

• Search payments by product name

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/payments?productName=eq:product%203' | jq .

[
  {
    "_id": "609de7b6f3255553dc410746",
    "userName": "dat",
    "orderId": 3,
    "productName": "product 3",
    "quality": 1,
    "price": 500,
    "__v": 0
  }
]

• Search payments by product name and sort price on ascending order

curl -X GET --header 'Accept: application/json' 'http://localhost:3080/payments?productName=5&sort_by=-price,orderId' | jq .

• Delete payments

curl -X DELETE --header "Content-Type: application/json" --header "Accept: application/json" -d @etc/checkout_delete.json "http://localhost:3080/payments" | jq .

⬆️ Back to Top

4. Run Jest test

• Make sure we start services service-registry, search-service, product-service, checkout-service, we can stop service orchestrator-service in this case. Also start MongoDB and Message MQ.
• We have 2 test suits in tests/payments.test.js and tests/product.test.js

# start all services before run jest test
cd orchestrator-service
run npm test

⬆️ Back to Top

5. Source structure

5.1. Run eslint in every service folder

• In every service, type the following command line

npm run lint
npm run lint:fix

5.2. Structure

• Most of service have a similar structure, it includes src and tests folder.
• Currently we only focus testing on orchestrator-service, a full flow from orchestrator-service to backend and database.

5.3. Library Usage

• axios: A promise based HTTP clien
• bunyan: A JSON logging library for node.js services
• mongoose: A MongoDB ODM
• amqplib: Library for message broker, RabbitMQ
• http-errors: Create HTTP error objects
• semver: The semantic version parser.

5.4. Database

• To simplify the persistent we just have 2 tables (collections) in MongoDB, Product and Payment.
• These collections dynamiclly created by using ORM mongoose once run services.
• Currently, most of backend services are using ORM persistence including search-service, checkout-service, product-service
• Here are schemes

const productSchema = new mongoose.Schema({
	name: String,
	price: Number,
	branch: String,
	color: String
});

const paymentSchema = new mongoose.Schema({
	userName: String,
	orderId: Number,
	productName: String,
	quality: Number,
	price: Number
});

⬆️ Back to Top

6. Sequence diagram in some use cases

• Search and sort products with some condition

• Post checkout data into Message RabbitMQ

⬆️ Back to Top