Skip to content

Latest commit

 

History

History
95 lines (76 loc) · 2.62 KB

LEARNING docs.md

File metadata and controls

95 lines (76 loc) · 2.62 KB

Learning Guide: Local API Mock Server

Core Concepts

1. HTTP Server Basics

  • http.server module: Python's built-in HTTP server framework
  • BaseHTTPRequestHandler: Base class for handling HTTP requests
  • HTTPServer: Creates a socket server that listens for HTTP requests

2. Threading and Concurrency

  • Thread Safety: Using locks to protect shared data (config, logs)
  • Concurrent Requests: Server handles multiple requests using threading
  • Race Conditions: Why we need threading.Lock() for shared state

3. Resilience Testing

  • Latency Simulation: time.sleep() to simulate slow networks
  • Failure Injection: Random failures to test error handling
  • Flaky Services: Testing retry logic and circuit breakers

4. API Design Patterns

  • Configuration as Code: JSON-based endpoint definitions
  • Hot Reload: Update config without restarting server
  • Special Endpoints: __reload, __logs for server management

Exercises

Beginner

  1. Add a new endpoint to config.json
  2. Create a template variable for {{date}} (just the date, not time)
  3. Add request body logging

Intermediate

  1. Implement request body validation
  2. Add support for custom response headers per endpoint
  3. Create a CLI command to generate config from OpenAPI spec

Advanced

  1. Implement the full record-and-replay proxy
  2. Add WebSocket support
  3. Create a web UI for managing endpoints
  4. Add authentication simulation (JWT tokens)

Key Files

  • mock_server.py: Main server implementation
  • config.json: Endpoint configuration
  • recorder.py: Traffic recording (stretch goal)
  • test_server.py: Integration tests

Common Patterns

Adding a New Template Variable

# In TemplateEngine._render_string()
template = re.sub(
    r'\{\{your_variable\}\}',
    'your_value',
    template
)

Adding a New HTTP Method

# In MockRequestHandler
def do_PATCH(self):
    self._handle_request('PATCH')

Configuring Endpoint Behavior

{
  "path": "/api/endpoint",
  "method": "GET",
  "response": {"data": "value"},
  "status": 200,
  "latency_ms": 100,
  "failure_rate": 0.1
}

Limitations and Trade-offs

  1. Single Process: Not suitable for high-load testing
  2. No State: Each request is independent (no session management)
  3. Simple Templates: Not a full template engine like Jinja2
  4. No Request Matching: Only exact path matches (no wildcards)

Extension Ideas

  • Add request/response middleware
  • Support for GraphQL endpoints
  • Rate limiting simulation
  • Response caching
  • Metrics and monitoring dashboard
  • Docker container for easy deployment