DailyAzureUpdatesGenerator

Azure Updates Automatic Summary

Japanese version

Japanese Version is here

Overview

Automatic summaries of Azure Updates are stored in the updates_en directory. The automatic summary feature runs daily at noon Japan time via GitHub Actions, collecting Azure Updates information from the past day and generating English summaries.

Disclaimer

The automatic summary reports are generated based on information obtained from RSS feeds and do not guarantee accuracy or completeness. Users should use the provided information at their own risk. If you find content of interest, we recommend checking the original information by referring to the links included in the reports.

Azure Updates RSS Feed Processing Application

This application reads Azure Updates RSS feeds, summarizes Azure Update information that has been updated within a specified time period using Azure OpenAI’s GPT-4.1-mini model in English, and outputs the results in Markdown format.

Features

• Robust RSS feed retrieval: Tries multiple URLs to ensure reliable feed acquisition
• Details mode: Retrieve detailed information from Azure Updates API (--details option)
• Filtering of new updates within specified time periods
• English summary generation using Azure OpenAI
• Detailed report output in Markdown format
• Comprehensive error handling and retry functionality
• Detailed logging and monitoring
• Test mode: Feed retrieval testing functionality

Required Environment Variables

Before running the application, set the following environment variables:

• AOAI_ENDPOINT: Azure OpenAI endpoint URL
• AOAI_KEY: Azure OpenAI API key
• DEPLOYMENT: Azure OpenAI deployment name
• API_VER: Azure OpenAI API version (default: 2024-02-15-preview)
• CHECK_HOURS: Target check time (in hours, default: 24)

Installation

1. Install required packages:

pip install -r requirements.txt

1. Test RSS feed:

test_feed.bat

or

python getlatestupdate.py --test-feed --verbose

Usage

Test Execution (Recommended: Run First)

# Test RSS feed retrieval
python getlatestupdate.py --test-feed --verbose

# Test execution with batch file
test_feed.bat

Basic Usage

python getlatestupdate.py

Execution with Options

# Check updates from the past 12 hours
python getlatestupdate_en.py --hours 12

# Details mode (retrieve detailed information from Azure Updates API)
python getlatestupdate_en.py --details

# Specify output directory
python getlatestupdate_en.py --output-dir reports

# Output detailed logs
python getlatestupdate_en.py --verbose

# Run RSS feed test only
python getlatestupdate_en.py --test-feed

# Combine multiple options (details mode + detailed logs + 72 hours)
python getlatestupdate_en.py --details --verbose --hours 72

Using Details Mode

Details mode (--details) retrieves detailed information from the Azure Updates API in addition to RSS feed information:

# Run in details mode
python getlatestupdate_en.py --details

# Batch file example with details mode
run_details_example.bat

Details mode features:

• Retrieve detailed information for each article from Azure Updates API
• Use more accurate and complete article titles and content
• Display information retrieved from API in reports
• Detailed information source tracking and log output

Environment Variable Setup Example (Windows)

set AOAI_ENDPOINT=https://your-openai-resource.openai.azure.com
set AOAI_KEY=your-api-key-here
set DEPLOYMENT=gpt-4-mini
set API_VER=2024-02-15-preview
set CHECK_HOURS=24

Troubleshooting

RSS Feed Retrieval Errors

1. Check with test mode:

   python getlatestupdate_en.py --test-feed --verbose
   

2. Check log file: azure_updates.log

3. Verify network connection: Internet connection and proxy settings

4. Multiple URL attempts: The application automatically tries multiple RSS URLs

Details Mode Related Troubleshooting

1. Test details mode:

   python test_details.py
   

2. API access errors: If Azure Updates API is temporarily unavailable, processing continues in standard mode

3. Update ID extraction failure: When ID cannot be extracted for some articles, RSS information is used

Common Errors and Solutions

• “No entries in feed”: RSS URL may have been changed
• “not well-formed (invalid token)”: XML parse error. Trying alternative URLs
• HTTP connection errors: Check network settings and firewall

Output

The application generates:

1. Markdown report: updates/azure-updates-YYYY-MM-DD.md
2. Log file: azure_updates.log

Report Contents

Standard Mode

• Generation date/time and target period
• Summary of update count
• Details of each update:
  • Title (from RSS feed)
  • Publication date/time
  • Link
  • Category
  • AI-generated English summary
  • Original detailed description (from RSS feed)

Details Mode (--details)

In addition to the above:

• Update ID: Individual article identifier
• Data source: Clear indication of Azure Updates API usage
• Enhanced title: More accurate title retrieved from API
• Detailed content: Complete article content retrieved from API
• Improved summary: Summary based on more detailed and accurate information

Error Handling

• Multiple RSS URLs: Automatic fallback when primary URL fails
• Network errors: Automatic retry and timeout settings
• XML parse errors: Minor errors are ignored to continue processing
• API rate limiting: Appropriate wait time settings
• Detailed logs: Comprehensive logging for troubleshooting

Security Considerations

• API keys retrieved from environment variables
• Use of HTTPS communication
• Appropriate timeout settings
• Session management and resource cleanup
• Proper User-Agent configuration

Limitations

• Dependent on Azure OpenAI rate limits
• Dependent on RSS feed availability
• Date parsing accuracy depends on feed format

This site is open source. Improve this page.