api-monitor-sdk 0.1.0

Automatic HTTP request monitoring for Python applications

API Monitor SDK

Automatically monitor all HTTP requests in your Python application with zero code changes.

Installation

pip install api-monitor-sdk

Quick Start

from apimonitor import Monitor
import requests

# Initialize monitor
monitor = Monitor(api_key="your-api-key-from-dashboard")
monitor.start()

# All HTTP requests now automatically tracked
response = requests.get("https://api.github.com/users/octocat")
# Metrics sent in background - your app continues normally

# Optional: track custom business events
monitor.track_custom("payment_processed", amount=100, user_id="123")

Features

• ✅ Zero-code monitoring - Just call monitor.start()
• ✅ Supports multiple libraries - requests, httpx, aiohttp
• ✅ Background sending - Non-blocking, won't slow your app
• ✅ Batching - Sends 100 requests or every 10 seconds
• ✅ Offline queue - Buffers metrics when network fails
• ✅ Sampling - Monitor 10% of traffic to reduce overhead
• ✅ Thread-safe - Works with multi-threaded applications

Configuration

monitor = Monitor(
    api_key="your-api-key",
    endpoint="https://api.yourmonitor.com/ingest",  # Custom endpoint
    batch_size=100,              # Send after N requests
    batch_timeout=10,            # Or after N seconds
    sampling_rate=1.0,           # 1.0 = 100%, 0.1 = 10%
    offline_mode=True,           # Buffer when network fails
    debug=True                   # Print debug logs
)
monitor.start()

Framework Examples

Flask

from flask import Flask
from apimonitor import Monitor

app = Flask(__name__)

monitor = Monitor(api_key="your-key")
monitor.start()

@app.route("/")
def home():
    # Any requests made here are auto-tracked
    return "Hello World"

FastAPI

from fastapi import FastAPI
from apimonitor import Monitor

app = FastAPI()

monitor = Monitor(api_key="your-key")
monitor.start()

@app.get("/")
def home():
    return {"message": "Hello World"}

Django

# In your settings.py or apps.py

from apimonitor import Monitor

monitor = Monitor(api_key="your-key")
monitor.start()

Advanced Usage

Custom Events

Track business-specific events:

# Track payment
monitor.track_custom("payment_processed", 
                     amount=99.99, 
                     currency="USD",
                     user_id="user_123")

# Track signup
monitor.track_custom("user_signup",
                     email="[e-mail]",
                     plan="premium")

Stopping the Monitor

# Gracefully stop and flush remaining metrics
monitor.stop()

How It Works

1. Interception - Monkey-patches HTTP libraries (requests, httpx, aiohttp)
2. Collection - Captures timing, status, URL, errors
3. Batching - Queues metrics until batch size or timeout reached
4. Sending - Background thread sends to API endpoint
5. Offline Queue - If network fails, saves to local SQLite database
6. Auto-flush - On program exit, sends remaining metrics

Supported Libraries

• ✅ requests - Most popular HTTP library
• ✅ httpx - Modern async HTTP library
• ✅ aiohttp - Async HTTP client/server
• 🚧 urllib3 - Coming soon
• 🚧 http.client - Coming soon

Requirements

• Python 3.8+
• requests (automatically installed)

License

MIT

Support

• Documentation: https://docs.yourmonitor.com
• Issues: https://github.com/yourusername/api-monitor-sdk/issues
• Dashboard: https://app.yourmonitor.com