All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Runtime Image Update: Updated to
ghcr.io/opentechil/offline-python-runtime-docker:v1.1.0-release.0 - Documentation Enhancement: Added clear guidance on offline package management process
- Repository Links: Added direct links to runtime image repository for package requests
- Offline Package Process: Documented process for adding packages to offline environments via runtime image
- Repository Integration: Enhanced documentation linking to runtime image repository
- Runtime Image Integration: Now uses
ghcr.io/opentechil/offline-python-runtime-docker:v1.0.0-release.0as base image - Runtime Package Management: Wheelhouse and package management moved to runtime for better compatibility
- Oracle Driver Support: Leverages Oracle Instant Client from runtime image
- Runtime Wheelhouse: Package installation handled at runtime through pre-configured wheelhouse system
- Simplified Architecture: Removed local
requirements.txtdependencies, now managed by runtime image
- Package Management: Core packages (pandas, oracledb, cryptography, etc.) now provided by runtime image
- Build Process: Streamlined build process leveraging runtime image foundation
- Compatibility: Enhanced offline compatibility through runtime wheelhouse system
- Runtime Repository: Integration with offline-python-runtime-docker repository
- Core Python packages now sourced from runtime image
- Oracle Instant Client 19.29 included via runtime image
- Data science stack (pandas, numpy, scikit-learn) available from runtime
- Multi-stage build: Runtime image as foundation with devcontainer customization
- Optimized layers: Better Docker layer caching and reduced build times
- Runtime flexibility: Package installation available both at build time and runtime
- Updated README.md to reflect runtime image usage
- Added runtime architecture explanations
- Updated configuration examples for runtime package management
- Package management complexity with local wheelhouse
- Build image size optimization needed
- Oracle driver compatibility issues in some environments
Breaking Changes:
- Local
requirements.txtpackages are no longer built into the image - Package management now happens at runtime
Recommended Actions:
-
Update your development workflow to install packages at runtime:
# Inside devcontainer pip install your-package -
For persistent packages, create project-specific
requirements.txtand install when container starts -
Update build scripts to use the new runtime image base
Benefits:
- Faster build times
- Better package compatibility
- Simplified maintenance
- Access to runtime wheelhouse with pre-compiled packages
| Version | Status | Support End |
|---|---|---|
| 1.1.x | ✅ Current | TBD |
| 1.0.x | 2026-03-07 |
For detailed release notes and migration guides, check the GitHub Releases page.