Phase 8: Platform Service Simplification

Agent Assignment

Primary Agent: platform-agent Duration: 35-45 minutes

Prerequisites

Phase 1 (Docker Compose) must be complete

Objectives

Remove MSSQL database dependency
Remove ETL container and pipeline
Update to use mvp-postgres and mvp-redis
Load VPIC data at startup (not weekly ETL)
Simplify to single container

Step-by-Step Instructions

Step 1: Update Platform Service Config

Modify mvp-platform-services/vehicles/config.py:

# Update database connection
DATABASE_URL = os.getenv(
    'DATABASE_URL',
    'postgresql://postgres:password@mvp-postgres:5432/motovaultpro'
)

# Update Redis connection
REDIS_URL = os.getenv(
    'REDIS_URL',
    'redis://mvp-redis:6379'
)

# Use vehicles_platform schema for isolation
SCHEMA = 'vehicles_platform'

Step 2: Update requirements.txt

Remove MSSQL dependencies:

# REMOVE:
# pymssql
# pyodbc

# KEEP:
psycopg2-binary
redis
fastapi
uvicorn

Step 3: Remove ETL Code

cd mvp-platform-services/vehicles/

# Remove ETL scripts
rm -rf etl/

# Remove MSSQL migration scripts
rm -rf migrations/mssql/

Step 4: Create Startup Data Loader

Create mvp-platform-services/vehicles/startup_loader.py:

"""Load VPIC data at service startup"""
import asyncio
from database import get_db
from vpic_client import VPICClient

async def load_initial_data():
    """Load vehicle data from VPIC API at startup"""
    db = await get_db()
    vpic = VPICClient()

    # Check if data already loaded
    result = await db.fetch_one("SELECT COUNT(*) FROM vehicles_platform.makes")
    if result[0] > 0:
        print("Data already loaded, skipping...")
        return

    # Load makes, models, etc. from VPIC
    print("Loading initial vehicle data...")
    # Implementation here...

Step 5: Update main.py

from startup_loader import load_initial_data

@app.on_event("startup")
async def startup_event():
    """Run on service startup"""
    await load_initial_data()

Step 6: Update Dockerfile

FROM python:3.11-slim

WORKDIR /app

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .

# No ETL dependencies needed
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

Step 7: Rebuild and Test

# Rebuild platform service
docker compose build mvp-platform

# Restart
docker compose up -d mvp-platform

# Check logs
docker compose logs mvp-platform

# Test API
curl http://localhost:8000/health
curl http://localhost:8000/vehicles/makes?year=2024

Validation Criteria

No MSSQL dependencies in requirements.txt
No ETL code remains
Uses mvp-postgres for database
Uses mvp-redis for cache
Service starts successfully
API endpoints work
VIN decode functional

Validation Commands:

grep -r "mssql\|pymssql\|pyodbc" mvp-platform-services/vehicles/
# Expected: 0 results

docker compose exec mvp-platform python -c "import psycopg2; print('OK')"
# Expected: OK

curl http://localhost:8000/vehicles/makes?year=2024
# Expected: JSON response with makes

Update EXECUTION-STATE.json

{
  "phases": {"8": {"status": "completed", "validation_passed": true}}
}

3.2 KiB Raw Blame History