Wikipedia API - Lightning-Fast Search

High-performance Wikipedia search API serving 4.8M articles with millisecond response times.

Features

• Lightning-fast search: Optimized FTS queries with sub-millisecond exact match
• Dual search strategy: Exact title matching and Full-text search with FTS5
• Result caching: LRU cache for ultra-fast repeated queries
• Async logging: Non-blocking performance monitoring
• Production-ready: Gunicorn WSGI server with 8 workers

Architecture

• Framework: Flask
• Database: SQLite with FTS5 full-text search
• WSGI Server: Gunicorn with multi-worker configuration
• Caching: Python functools.lru_cache for query results

API Endpoints

Search Articles

GET /api/search?q=query&limit=10

Parameters:

• q (required): Search query string
• limit (optional): Number of results (default: 10, max: 20)

Installation

1. Install dependencies: pip install -r requirements.txt
2. Ensure database is accessible
3. Run with Gunicorn: gunicorn --config gunicorn.conf.py wsgi:application

Configuration

• Workers: 8
• Bind: 0.0.0.0:80
• Database: /var/www/wikipedia-api/wikipedia_production.db (7GB, 4.8M articles)

Performance

• Exact match: <1ms
• FTS search: 20-50ms
• Cached queries: <1ms
• Memory usage: ~70MB per worker

Files

• wsgi.py - WSGI entry point
• wikipedia_api_lightning.py - Main application
• async_worker.py - Async logging worker
• gunicorn.conf.py - Gunicorn configuration
• requirements.txt - Python dependencies