cloud-logging-handler 0.2.3

A Python logging handler for Google Cloud Logging with request tracing support

Cloud Logging Handler

A Python logging handler for Google Cloud Logging with request tracing support.

Features

• Structured JSON Logging: Outputs logs in Google Cloud Logging's structured format
• Request Tracing: Automatic trace context propagation via X-Cloud-Trace-Context header
• Log Aggregation: Aggregates all logs within a single request into one log entry
• Severity Tracking: Automatically tracks the highest severity level per request
• Multi-Framework Support: FastAPI, Flask, Django, aiohttp, Sanic
• Auto Framework Detection: Automatically detects framework from app instance
• Custom JSON Encoder: Support for high-performance JSON libraries (e.g., ujson)
• Zero Dependencies: Core handler has no external dependencies

Installation

# Using uv (recommended)
uv add cloud-logging-handler

# Using pip
pip install cloud-logging-handler

Quick Start

FastAPI / Starlette

import logging
from fastapi import FastAPI, Request
from cloud_logging_handler import CloudLoggingHandler, RequestLogs

app = FastAPI()

# Initialize handler with app for auto framework detection
handler = CloudLoggingHandler(
    app=app,
    project="your-gcp-project-id"
)

logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)

@app.middleware("http")
async def logging_middleware(request: Request, call_next):
    handler.set_request(RequestLogs(request))
    response = await call_next(request)
    handler.flush()
    return response

@app.get("/")
async def root():
    logging.info("Processing request")
    return {"message": "Hello World"}

Flask

import logging
from flask import Flask, g, request
from cloud_logging_handler import CloudLoggingHandler, RequestLogs

app = Flask(__name__)

handler = CloudLoggingHandler(
    app=app,
    project="your-gcp-project-id"
)

logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)

@app.before_request
def before_request():
    handler.set_request(RequestLogs(request))

@app.after_request
def after_request(response):
    handler.flush()
    return response

@app.route("/")
def hello():
    logging.info("Processing request")
    return {"message": "Hello World"}

Django

# settings.py
LOGGING = {
    'version': 1,
    'handlers': {
        'cloud': {
            'class': 'cloud_logging_handler.CloudLoggingHandler',
            'framework': 'django',
            'project': 'your-gcp-project-id',
        },
    },
    'root': {
        'handlers': ['cloud'],
        'level': 'DEBUG',
    },
}

# middleware.py
from cloud_logging_handler import CloudLoggingHandler, RequestLogs
import logging

class CloudLoggingMiddleware:
    def __init__(self, get_response):
        self.get_response = get_response
        self.handler = None
        for h in logging.getLogger().handlers:
            if isinstance(h, CloudLoggingHandler):
                self.handler = h
                break

    def __call__(self, request):
        if self.handler:
            self.handler.set_request(RequestLogs(request))
        response = self.get_response(request)
        if self.handler:
            self.handler.flush()
        return response

aiohttp

import logging
from aiohttp import web
from cloud_logging_handler import CloudLoggingHandler, RequestLogs

app = web.Application()

handler = CloudLoggingHandler(
    app=app,
    project="your-gcp-project-id"
)

logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)

@web.middleware
async def logging_middleware(request, handler_func):
    handler.set_request(RequestLogs(request))
    response = await handler_func(request)
    handler.flush()
    return response

app.middlewares.append(logging_middleware)

async def hello(request):
    logging.info("Processing request")
    return web.json_response({"message": "Hello World"})

app.router.add_get("/", hello)

Sanic

import logging
from sanic import Sanic, json
from cloud_logging_handler import CloudLoggingHandler, RequestLogs

app = Sanic("MyApp")

handler = CloudLoggingHandler(
    app=app,
    project="your-gcp-project-id"
)

logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)

@app.middleware("request")
async def before_request(request):
    handler.set_request(RequestLogs(request))

@app.middleware("response")
async def after_request(request, response):
    handler.flush()

@app.get("/")
async def hello(request):
    logging.info("Processing request")
    return json({"message": "Hello World"})

Using with ujson

For better JSON serialization performance:

import ujson
from cloud_logging_handler import CloudLoggingHandler

handler = CloudLoggingHandler(
    app=app,
    json_impl=ujson,
    project="your-gcp-project-id"
)

Configuration

CloudLoggingHandler Parameters

Parameter	Type	Description
app	object	Web application instance for auto framework detection
framework	str	Explicit framework name (starlette, flask, django, aiohttp, sanic)
trace_header_name	str	HTTP header name for trace context (default: X-Cloud-Trace-Context)
json_impl	module	Custom JSON encoder module (must have dumps method)
project	str	GCP project ID for trace URL construction

Framework Detection

The handler automatically detects the framework from the app instance:

App Module	Detected Framework
starlette.*, fastapi.*	starlette
flask.*	flask
django.*	django
aiohttp.*	aiohttp
sanic.*	sanic

You can also explicitly specify the framework:

handler = CloudLoggingHandler(
    framework="flask",
    project="your-gcp-project-id"
)

Log Output Format

With Request Context

When logging within a request context, logs are aggregated and output as structured JSON:

{
  "severity": "INFO",
  "name": "root",
  "process": 12345,
  "url": "https://example.com/api/endpoint",
  "logging.googleapis.com/trace": "projects/your-project/traces/abc123",
  "logging.googleapis.com/spanId": "def456",
  "message": "\n2025-12-01T12:00:00.000000+00:00\tINFO\tProcessing request\n2025-12-01T12:00:00.001000+00:00\tINFO\tRequest completed"
}

Without Request Context

When logging outside a request context, logs are output as plain text:

Processing request

How It Works

1. Handler Initialization: Framework-specific wrapper is selected based on app or framework parameter
2. Request Start: Middleware creates a RequestLogs context
3. Log Accumulation: All log calls within the request are accumulated in message field
4. Severity Tracking: The highest severity level is tracked
5. Trace Extraction: Trace context is extracted from request headers using framework-specific methods
6. Request End: flush() emits all accumulated logs as a single structured entry

This approach provides several benefits:

• Correlate all logs from a single request
• View logs grouped by trace in Cloud Console
• Reduce log volume while maintaining detail
• Optimized header/URL extraction per framework

Development

Setup

# Clone the repository
git clone https://github.com/loplat/gcp-cloud-logging-handler.git
cd cloud-logging-handler

# Install with dev dependencies using uv
uv sync --all-extras

# Run tests
uv run pytest

# Run linting
uv run ruff check .
uv run ruff format .

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Inspired by Google Cloud Logging documentation