Skip to content

omnipkg v2.2.0 - Per-Process Python Isolation for Safe, Concurrent, CI-Ready Workflows

Choose a tag to compare

View all tags
@1minds3t 1minds3t released this 10 Feb 11:57
v2.2.0
8198bba
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Overview

Version 2.2.0 represents a comprehensive architectural redesign focused on performance optimization, automation enhancement, and code quality improvements. This release introduces a persistent daemon architecture, optional hardware-accelerated atomic operations, and fully automated continuous delivery pipelines.

Major Features

High-Performance Worker Daemon

Replaced the on-demand worker system with WorkerPoolDaemon, a persistent service that maintains pre-forked Python processes. This architecture delivers near-instantaneous response times for package execution by eliminating cold-start overhead.

Key Benefits:

  • Sub-millisecond latency for subsequent executions
  • Reduced system resource overhead
  • Improved process isolation and lifecycle management

Hardware-Accelerated Atomic Operations

Introduced an optional C extension (omnipkg_atomic) implementing compare-and-swap (CAS) operations using native CPU atomic instructions. The system gracefully degrades to a pure-Python implementation when compilation is unavailable.

Technical Details:

  • HFT-grade concurrency control
  • Lock-free data structures
  • Zero-overhead abstraction with automatic fallback

Automated Conda-Forge Release Pipeline

Implemented end-to-end automation for conda-forge distribution:

  1. Monitors PyPI publication
  2. Computes package SHA256 checksums
  3. Generates conda-forge feedstock pull requests
  4. Validates CI/CD test results
  5. Auto-merges upon successful validation
  6. Manages multi-platform build orchestration

Enhancements

Internationalization Expansion

Expanded translation coverage across CLI interfaces and logging systems, improving accessibility for international users.

Intelligent Resource Monitoring

Enhanced 8pkg daemon monitor with daemon API integration for real-time worker pool status and accurate process tracking.

Dependency-Aware Verification

Improved installation verification logic to include transitive dependencies in bubble environments (e.g., providing TensorFlow when verifying Keras imports).

Integrated Historical Package Support

Consolidated time-machine functionality directly into core installation logic, eliminating the need for a separate module.

Code Quality Improvements

Architectural Simplification

  • Removed deprecated components: worker_controller.py, deptimemachine.py, lockmanager.py
  • Consolidated daemon logic into centralized modules
  • Reduced codebase complexity by ~1,500 lines

Test Suite Modernization

  • Refactored test suite to align with daemon architecture
  • Removed obsolete test files
  • Focused on production-grade performance benchmarks

Breaking Changes

None. This release maintains backward compatibility with v2.1.x configurations.

Upgrade Instructions

pip install --upgrade omnipkg
# or
conda install -c conda-forge omnipkg

For users running the daemon, restart is recommended:

8pkg daemon stop && 8pkg daemon start
Assets 64
Loading
0 Join discussion