Skip to content

A high-performance Python logging library designed for multiprocessing environments. Features colored output, multiprocessing support, and flexible configuration. Benchmarks show up to 29K msgs/sec with minimal memory footprint. Zero external dependencies.

License

Notifications You must be signed in to change notification settings

vertok/prismalog

Repository files navigation

prismalog

A robust, multi-process safe logging system for Python applications that integrates perfectly with the psy-supabase package.

PyPI version PyPI - Python Version Coverage Python 3.8 Pylint Python 3.8 Coverage Python 3.10 Pylint Python 3.10 Coverage Python 3.11 Pylint Python 3.11

Features

  • ⚛️ Zero-dependency core
  • 🚀 High performance
  • 🎨 Colored console output
  • 📁 Automatic log file rotation
  • 🔄 Multi-process safe logging
  • 🧵 Multithreading-safe logging
  • ⚙️ YAML-based configuration
  • 🔇 Control for verbose third-party libraries
  • 🧪 Testing support

Performance Characteristics

prismalog was designed for high-performance applications. My testing shows:

  • Overhead per log call: ~0.15ms (typical)
  • Memory impact: Minimal (~0.3MB for 10,000 messages)
  • Throughput: Capable of handling 20,000+ messages per second
  • Multi-process/Multi-threading safety: No measurable performance penalty compared to single-process

Performance Notes

While prismalog achieves excellent performance characteristics, it's important to note that the primary bottleneck is filesystem I/O when writing to log files. This limitation is inherent to disk-based logging systems:

  • File locking mechanisms required for multi-process safety introduce some overhead
  • Synchronous writes to ensure log integrity can impact throughput during high-volume logging events
  • Storage device speed directly impacts maximum sustainable throughput

For applications with extreme logging requirements, consider:

  • Using an asynchronous logging configuration
  • Implementing log batching for high-volume events
  • Configuring separate log files for different components to distribute I/O load

The current performance metrics were achieved with standard SSD hardware. With specialized I/O optimization or enterprise-grade storage systems, significantly higher throughput is achievable.

Quick Start

from prismalog import get_logger, LoggingConfig

# Initialize with configuration file
LoggingConfig.initialize(config_file="config.yaml")

# Get a logger
logger = get_logger("my_module")
logger.info("Application started")

Command-Line Integration

Any application using prismalog automatically supports these command-line arguments:

--log-config PATH          # Path to logging configuration file
--log-level {DEBUG,INFO,WARNING,ERROR,CRITICAL}
                           # Set the default logging level
--log-dir PATH             # Directory where log files will be stored

Usage

from prismalog.log import get_logger, LoggingConfig

# Initialize logging (with command-line support)
LoggingConfig.initialize(use_cli_args=True)

# Get logger and use it
logger = get_logger("my_app")
logger.info("Application started")

Dependencies

prismalog is designed to work with zero external dependencies for core functionality. It relies solely on the Python standard library, making it lightweight and easy to integrate into any project.

Optional Dependencies

  • YAML Configuration: If YAML config files needed, install with pip install prismalog[yaml]
  • Development: For running tests and examples, install with pip install prismalog[dev]

Installation Options

# Prepare before installation (recommended)
python -m venv .venv
source source .venv/bin/activate

# Basic installation - no external dependencies
pip install -e .

# With documentation support
pip install prismalog[doc]

# For development and testing
pip install prismalog[dev]

# With all optional features
pip install prismalog[all]

About

A high-performance Python logging library designed for multiprocessing environments. Features colored output, multiprocessing support, and flexible configuration. Benchmarks show up to 29K msgs/sec with minimal memory footprint. Zero external dependencies.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages