madihowa / lab_manager

Hutzler Lab Manager

This project serves as a lab manager for the Hutzler Lab at Caltech. It automates reminders for maintenance, snacks, and manages lab meeting schedules through emails, Slack, and Google Calendar.

Project Structure

• .github/: Contains GitHub workflows for automation.
• calendar_manager.py: Manages Google Calendar integration.
• config_loader.py: Loads configuration from JSON files.
• email_notifier.py: Handles email notifications.
• main.py: The main script for managing notifications.
• slack_notifier.py: Manages Slack notifications.
• duty_tracker.json: Tracks the rotation of lab duties.
• trigger.sh: Script for running main.py in a scheduled manner.
• check_and_trigger.sh: Checks for missed executions and triggers main.py if needed.
• markers/: Directory where the marker file emissions are stored.

Setup and Operation

Local Setup

1. Install dependencies from requirements.txt.
2. Set up environment variables for Gmail, Slack, and Google Calendar credentials.

   export GMAIL_USER=<email>
   export GMAIL_PASSWORD=<password>
   export SLACK_TOKEN=<token>
   export LAB_MEMBERS_INFO=<json file | base64>
   export GOOGLE_CALENDAR_SERVICE_KEY=<calendar service key | base64>

3. Move/Generate the token.pickle for Google Calendar API on the local machine.

PythonAnywhere Setup

1. Upload the script files to PythonAnywhere.
2. Set up a scheduled task to run main.py daily at 7 AM using cron.
   • Cron file:

     0 7 * * * /home/<username>/lfl-lab-manager/venv/bin/python /home/<username>/lfl-lab-manager/main.py

3. Ensure client_secret.json and token.pickle are safely uploaded and handled.

Handling Authentication

• The script checks the validity of token.pickle.
• If re-authentication is required, it sends an email notification.
• Manually update token.pickle on PythonAnywhere after re-authenticating locally.

Mac Setup for Scheduled Execution

For Mac users, a method is provided to ensure trigger.sh runs even if the Mac is asleep at the scheduled time:

1. Marker System: The trigger.sh script creates a daily marker file upon successful execution, indicating the script has run for that day.

2. Missed Execution Check: check_and_trigger.sh checks for the presence of this marker file. If it's missing (indicating a missed execution), it runs trigger.sh.

3. launchd Daemon: A launchd service is set up to run check_and_trigger.sh every time the Mac wakes up, ensuring missed executions are caught.

   • Create com.user.checkandtrigger.plist in ~/Library/LaunchAgents/ with the following content:

     <?xml version="1.0" encoding="UTF-8"?>
     <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
     <plist version="1.0">
     <dict>
         <key>Label</key>
         <string>com.user.checkandtrigger</string>
         <key>ProgramArguments</key>
         <array>
             <string>/path/to/check_and_trigger.sh</string>
         </array>
         <key>WatchPaths</key>
         <array>
             <string>/var/log/powermanagement</string>
         </array>
         <key>RunAtLoad</key>
         <true/>
     </dict>
     </plist>

   • Load the launchd job:

     launchctl load ~/Library/LaunchAgents/com.user.checkandtrigger.plist

Note: For security, never store sensitive information like lab members' details and service keys in the repository.

GitHub Actions

The project can be configured to use the GitHub Action defined in .github/workflows/main.yml to automate reminders.

Security

Sensitive information is handled securely, and environment variables are used to store credentials.

---

Remember to keep the token.pickle and client_secret.json files secure and handle them carefully during deployment and updates.