데이터 계약

# data_contracts/orders_contract.yaml
# 주문 데이터 계약 정의

apiVersion: datacontract/v1.0
kind: DataContract
metadata:
  name: orders-events
  version: 2.1.0
  owner: order-service-team
  contact: order-team@company.com

info:
  title: 주문 이벤트 데이터 계약
  description: 주문 서비스에서 발행하는 주문 이벤트 스트림

schema:
  type: object
  properties:
    order_id:
      type: string
      format: uuid
      description: "주문 고유 식별자"
    customer_id:
      type: string
      pattern: "^CUS-[0-9]{10}$"
    order_date:
      type: string
      format: date-time
    status:
      type: string
      enum: [pending, confirmed, shipped, delivered, cancelled]
    total_amount:
      type: number
      minimum: 0
  required:
    - order_id
    - customer_id
    - order_date
    - status
    - total_amount

quality:
  - name: order_id_unique
    type: unique
    column: order_id
  - name: freshness
    type: freshness
    maxDelay: 5 minutes

sla:
  availability: 99.9%
  latency: p95 < 100ms

---
# Great Expectations 검증 (Python)
import great_expectations as gx

context = gx.get_context()
validator = context.get_validator(
    datasource_name="orders_stream",
    data_asset_name="orders"
)

# 데이터 계약 검증 규칙
validator.expect_column_values_to_not_be_null("order_id")
validator.expect_column_values_to_be_unique("order_id")
validator.expect_column_values_to_be_in_set(
    "status",
    ["pending", "confirmed", "shipped", "delivered", "cancelled"]
)
validator.expect_column_values_to_be_between(
    "total_amount", min_value=0
)

# 검증 결과 저장
validator.save_expectation_suite()
print("Data contract validation complete!")