Tube Archivist import

gistfile1.txt

#!/usr/bin/env python3
"""
Script to bulk import YouTube channels into TubeArchivist from a text file.
"""

import sys
import time
import subprocess

def subscribe_to_channel(channel_identifier):
    """Subscribe to a channel in TubeArchivist using the identifier (URL, handle, or ID)."""
    try:
        # Clean up the identifier
        channel = channel_identifier.strip()
        
        # Format the identifier based on the debug info we observed
        
        # If it's a URL with a handle, extract just the handle
        if '@' in channel and ('youtube.com' in channel or 'youtu.be' in channel):
            parts = channel.split('@')
            channel = '@' + parts[1].split('/')[0]
        
        # If it's a standard channel ID URL, extract the ID
        elif '/channel/' in channel:
            channel = channel.split('/channel/')[1].split('/')[0].strip()
        
        # Use the subscribe_to function with the identifier
        cmd = ['docker', 'exec', 'tube-archivist', 'bash', '-c', 
               f'cd /app && python manage.py shell -c "from task.tasks import subscribe_to; subscribe_to.delay(\\"{channel}\\")"']
        
        result = subprocess.run(cmd, capture_output=True, text=True)
        
        if result.returncode == 0:
            return True
        else:
            print(f"Error output: {result.stderr}")
            return False
    except Exception as e:
        print(f"Error subscribing to channel {channel_identifier}: {e}")
        return False

def main():
    if len(sys.argv) != 2:
        print(f"Usage: {sys.argv[0]} channels_file.txt")
        sys.exit(1)
    
    channels_file = sys.argv[1]
    
    try:
        with open(channels_file, 'r') as f:
            channels = f.readlines()
    except Exception as e:
        print(f"Error reading file: {e}")
        sys.exit(1)
    
    # Clean up channel identifiers
    channels = [channel.strip() for channel in channels if channel.strip()]
    
    successful = 0
    failed = 0
    
    # Process each channel
    for i, channel in enumerate(channels):
        print(f"Processing {i+1}/{len(channels)}: {channel}")
        
        if subscribe_to_channel(channel):
            successful += 1
            print(f"  Success!")
        else:
            failed += 1
            print(f"  Failed to subscribe.")
        
        # Sleep briefly to avoid overwhelming the system
        time.sleep(1)
    
    print(f"Finished processing {len(channels)} channels")
    print(f"Successfully subscribed to {successful} channels")
    print(f"Failed to subscribe to {failed} channels")

if __name__ == "__main__":
    main()