Bot releases are hidden (Show)
Published by DjangoPeng about 1 month ago
Published by DjangoPeng about 2 months ago
Published by DjangoPeng about 2 months ago
Unit Testing Enhancements: Introduced comprehensive unit testing across the project using the unittest
module. The unit tests now cover critical components, including subscription management, report generation, and interaction with large language models (LLMs). Detailed explanations of the unittest
tools, such as @patch
and MagicMock
, are included in the newly added unit_test.md documentation.
Docker Integration for CI/CD: Implemented a Docker-based continuous integration and deployment (CI/CD) system to ensure code reliability before deployment. This includes:
Dockerfile
for building the project into a Docker image, running unit tests during the build process.build_image.sh
script that automatically tags the Docker image with the current Git branch name.validate_tests.sh
script that executes all unit tests and aborts the Docker build if any test fails, ensuring only passing code is deployed.Automated Testing with Shell Scripts: Added new shell scripts for automated testing and Docker image validation:
build_image.sh
: Automates the Docker image build process, tagging images based on the current Git branch.validate_tests.sh
: Ensures that all unit tests are executed during the Docker image build, halting the process if any test fails.Enhanced Unit Test Coverage: Improved unit test coverage across the project. The testing framework now thoroughly checks the functionality of key components, including GitHub updates retrieval, email notifications, and report generation through both OpenAI
and Ollama
models.
Documentation Updates: Updated the main README.md
to include new sections on unit testing and Docker integration. Additionally, detailed descriptions and usage examples for build_image.sh
and validate_tests.sh
have been provided.
unittest
framework.Dockerfile
with integrated unit test validation during the build process to ensure reliable Docker images.build_image.sh
to streamline Docker image creation and tagging based on Git branches.validate_tests.sh
to automate the execution of unit tests during Docker image builds, ensuring only passing code is deployed.README.md
and added unit_test.md
to document new features, testing methodologies, and Docker integration.This release enhances the reliability and maintainability of GitHub Sentinel by integrating automated testing and Docker-based CI/CD practices. The new unit tests and validation scripts ensure that only thoroughly tested code reaches production, significantly reducing the risk of deployment issues.
Published by DjangoPeng about 2 months ago
Extended LLM Class
Support: Enhanced the LLM Class
to support calling privatized large model services such as Ollama
. This allows users to utilize private AI models in addition to public offerings like OpenAI GPT
models, providing flexibility in sensitive or specific operational environments.
Configuration Management Enhancements: Added new configuration options in the llm
configuration file to support both OpenAI GPT
models and Ollama
privatized deployment model services. This update ensures seamless integration and switching between different AI models based on user requirements.
Updated Documentation: Published new installation and service deployment documentation for Ollama
, providing users with detailed guidelines on setting up and deploying Ollama
services effectively.
Ollama
.LLM Class
to include support for Ollama
, enabling the use of privatized large model services.llm
configuration management to accommodate both OpenAI GPT
and Ollama
models, enhancing flexibility.Ollama
to assist users in setting up privatized AI services.Published by DjangoPeng about 2 months ago
Scheduled Updates: Integrated the Schedule library for simpler and more reliable scheduling of tasks, replacing the original DaemonContext and Threading methods used for daemon processes. This enhancement allows for easier management and setup of periodic updates.
Email Notification System: Implemented a basic email notification feature that maintains the format and title of the report when sending updates. This allows users to receive timely updates directly in their email with preserved content aesthetics.
Daemon Process Management: Added a daemon_control.sh
script for managing daemon processes, facilitating support for containerized deployments. This script provides straightforward commands for starting, stopping, and monitoring the daemon processes.
Hacker News Trend Report: Introduced hacker_news_client.ipynb
, a new Jupyter Notebook demonstration on how to use ChatGPT and GPT-4o-mini to generate trend reports for technology topics on Hacker News. This feature leverages the latest AI to provide insights into the most discussed technology news.
daemon_control.sh
script for efficient management of daemon processes, enhancing support for containerized environments.hacker_news_client.ipynb
for demonstrating the generation of technology trend reports on Hacker News using ChatGPT and GPT-4o-mini.Published by DjangoPeng about 2 months ago
Decoupled Prompt Management: Extracted prompt definitions from llm.py
to enable more flexible and maintainable prompt management through a system role, enhancing the customizability of report generation tasks in terms of content and format.
Model Upgrade: Upgraded the default large model from GPT-3.5-Turbo to GPT-4o-mini, improving the quality of text generation and increasing the accuracy and relevance of generated reports.
llm.py
and introduced a system role for dynamic content and format injection in report generation tasks.Published by DjangoPeng 2 months ago
Added github_client
Jupyter Notebook:
GitHubClient
module.Added report_generator
Jupyter Notebook:
ReportGenerator
and LLM
modules to automatically generate progress reports.GitHubClient
, ReportGenerator
, and LLM
modules.GitHubClient
and ReportGenerator
to demonstrate module functionality.Published by DjangoPeng 2 months ago
Added github_client
Jupyter Notebook:
GitHubClient
module.Added report_generator
Jupyter Notebook:
ReportGenerator
and LLM
modules to automatically generate progress reports.GitHubClient
, ReportGenerator
, and LLM
modules.GitHubClient
and ReportGenerator
to demonstrate module functionality.Published by DjangoPeng 2 months ago
Multi-Mode Operation:
nohup
and managed logging.Enhanced Documentation: The README has been extensively updated to include detailed setup instructions for all three operating modes, ensuring users can configure and run GitHub Sentinel in the way that best suits their needs.
Published by DjangoPeng 2 months ago
export_progress_by_date_range
function for clearer filename formatting and more intuitive time range reporting.Published by DjangoPeng 3 months ago
get_subcriptions()
as list_subscriptions
of SubscriptionManager
Published by DjangoPeng 3 months ago
Published by DjangoPeng 4 months ago
Published by DjangoPeng 4 months ago