v0.10.0

Changelog

[Version 0.10.0] - 2024-12-29

This release of Git-GPT introduces significant enhancements and new functionalities, expanding its capabilities to support a wider range of AI providers and improving configuration management for user convenience.

Added

• Introduced support for multiple AI providers, including OpenAI, Azure OpenAI, Ollama, Claude (Anthropic), and Google Generative AI, allowing users to select the most suitable model for their needs.
• Added a new module ai_client.py to handle API requests to different AI providers.
• Implemented a more interactive git-gpt config command with prompt-based configuration setup, allowing users to specify or update model configurations easily.
• Added new commands:
  • git-gpt set-default: Set a default model configuration for simplified usage.
  • git-gpt delete-model: Easily delete a specific model configuration.
  • git-gpt show-models: Display all configured models along with their providers.
• Integrated the prompt_toolkit library to enhance user interaction in command-line prompts.

Changed

• Refactored the internal architecture to remove the request_module.py and replace it with ai_client.py for better maintainability and expanded functionality.
• Updated configuration handling to store multiple model configurations, allowing users to switch between models as needed.
• Revised error messaging to offer more specific guidance based on the new configuration model.

Removed

• Removed request_module.py to unify and improve the management of API requests through ai_client.py.

Fixed

• Improved error handling and messaging for configuration-related operations, including guidance for setting or modifying API keys and model configurations.

Security

• Enhanced security by masking API keys in configuration displays to protect sensitive information during configuration management.

v0.9.0 ollama

Changelog

[Version 0.9.0] - 2024-09-29

This release introduces significant enhancements to the Git-GPT CLI tool, expanding its capabilities to support Ollama alongside OpenAI's API. Users can now choose between two APIs for generating commit messages, issues, and performing quality checks.

Added

• Support for Ollama, an open-source, locally hosted language model.
• New request_module.py for handling API requests to both OpenAI and Ollama.
• Configuration options for specifying the API type and Ollama base URL.
• New command options in configuration and usage commands to accommodate Ollama integration.

Changed

• Updated README.md to reflect the support for Ollama.
• Enhanced the configuration command to allow setting the API type and Ollama base URL.
• Modified existing commands to use the new request module for sending requests to the configured API.
• The default model setting is now retrieved from the configuration instead of being hardcoded.

Deprecated

• The previous hardcoded API interactions with OpenAI are now deprecated in favor of the new request handling module.

Fixed

• Improved error handling in API requests, providing clearer messages when issues occur.
• Ensured consistency in the handling of default values for model and language across various commands.

Security

• Addressed potential vulnerabilities related to API key management by masking the key in configuration outputs.

v0.8.0

Changelog

[Version 0.8.0] - 2023-10-10

This release introduces significant organizational changes to the git-gpt project, enhances its functionality, and improves usability through a modular command structure. We've also updated several command options for clarity and specificity.

Added

• Introduced a new command ask_command.py to enable users to ask custom questions about code diffs with specified models and commit ranges.
• Added changelog_command.py for generating changelogs based on recent commits using the OpenAI API.
• Integrated commit_command.py to facilitate crafting of commit messages based on diff analysis.
• Included config_command.py to assist users in configuring the settings like API keys and token limits.
• Added issue_command.py for generating GitHub-style development issues from code diffs.
• The quality_command.py was introduced for executing quality checks on code changes, focusing on adherence to coding standards, potential bugs, and security vulnerabilities.

Changed

• Revised the README to include a detailed project structure, improving understanding of the codebase organization.
• Refactored the command-line interface in main.py to modularize individual commands, enhancing maintainability and scalability.
• Adjusted token option names for clarity, such as --max-tokens-quality, --max-tokens-issue, and --max-tokens-changelog to specify module-specific token limits.

Removed

• Outdated and redundant amalgamated command implementations in main.py, which have now been separated into dedicated command files for better organization.

Fixed

• Corrected and improved several command flags for consistency in usage, including --max-tokens override options for specific prompts.

Security

• Enhanced the security of the codebase by examining diff outputs for vulnerabilities in the quality_command.py.

The development team has revised several core functionalities to adhere to our strategic goal of enhancing automation and configurability for developer workflows. This version marks a leap towards more organized and user-friendly interfaces.

v0.7.0

Changelog

[Version 0.7.0] - 2024-09-28

This release introduces several updates to improve usability and streamline the development process.

Added

• Added a new pyproject.toml file for enhanced project metadata and build configuration.
• Detailed instructions for setting up the development environment have been added to the README.

Changed

• Updated the installation command in the README to remove unnecessary parameters.
• Increased the version number from 0.6.0 to 0.7.0 in __init__.py.

Removed

• The setup.py file has been removed as part of the transition to using pyproject.toml for package management.

Fixed

• No documented bug fixes in this release.

Security

• No security vulnerabilities addressed in this release.

v0.6.0

Version 0.6.0 - 2024-09-16

Breaking Changes

Revised API client base URLs to improve the call to OpenAI services without appending /v1.
You'll need to configure your base URL again.

Install this version

pip install --upgrade git+https://github.com/ShinChven/[email protected]#egg=git-gpt

Added

• Included .pytest_cache in the .gitignore file for ignoring pytest cache files.
• Added a new 'Development' section in README.md with instructions for setting up a virtual environment and installing the package in editable mode.
• Added support for generating a changelog and querying specific questions using the ask command in git_gpt.
• Introduced several new CLI command options with shorthand arguments for improved usability.
• Added options for --changelog-max-tokens and --quality-check-max-tokens in the configuration step for more precise control over the token limit in CLI commands.

Changed

• Breaking changes: Revised API client base URLs to improve the call to OpenAI services without appending /v1.
• Enhanced the README.md to provide more detailed instructions on configurations and usage.
• Updated command-line options in git_gpt to include shorthand variations for easier access.
• Refined command options in git_gpt for consistency and better handling of user input.
• Increased default max tokens for issue, quality, and changelog generations to 2000.

Fixed

• Corrected the handling of variable names in git_gpt main methods to ensure better functionality and avoid conflicts.

No deprecated, removed, or security updates were included in this release.

v0.3.1 add quality check with GPT

feat: Update user prompt message for quality check

- Update user prompt message to include instructions on quality, security, and performance checks
- Refactor the prompt message to request a quality check report in the specified language

v0.2.1 Updated prompts to generate commit message

Merge branch 'main' of https://github.com/ShinChven/git-gpt

v0.2.0

Use latest OpenAI SDK.