Python script to recursively download missing subtitles for video files

download_subtitles.py

#!/usr/bin/env python3

import os
import sys
from babelfish import Language
from subliminal import download_best_subtitles, region, save_subtitles, scan_video

# Configure subliminal cache
region.configure('dogpile.cache.dbm', arguments={'filename': 'cachefile.dbm'})

# Supported video extensions
VIDEO_EXTENSIONS = ('.mp4', '.mkv', '.avi', '.mov')

def download_subtitles_for_directory(directory, languages=['eng']):
    """
    Recursively scan a directory for video files and download subtitles if missing.
    :param directory: Path to the directory to scan
    :param languages: List of languages for subtitles (default: ['eng'])
    """
    for root, _, files in os.walk(directory):
        for file in files:
            # Skip files starting with dot (e.g., ".file" or "._file")
            if file.startswith('.') or file.startswith('._'):
                print(f"Skipping hidden or system file: {file}")
                continue
                
            if file.endswith(VIDEO_EXTENSIONS):
                video_path = os.path.join(root, file)
                subtitle_path = os.path.splitext(video_path)[0] + '.srt'
                
                # Skip if subtitles already exist
                if os.path.exists(subtitle_path):
                    print(f"Subtitle already exists for: {video_path}")
                    continue

                print(f"Processing video: {video_path}")
                
                try:
                    # Scan the video
                    video = scan_video(video_path)
                    if not video:
                        print(f"Could not process video: {video_path}")
                        continue

                    # Download subtitles
                    subtitles = download_best_subtitles([video], {Language(lang) for lang in languages})
                    if video in subtitles:
                        save_subtitles(video, subtitles[video])
                        print(f"Subtitle downloaded and saved for: {video_path}")
                    else:
                        print(f"No subtitles found for: {video_path}")
                except Exception as e:
                    print(f"Error processing {video_path}: {e}")

if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: python download_subtitles.py <directory> [language]")
        sys.exit(1)

    directory = sys.argv[1]
    language = sys.argv[2] if len(sys.argv) > 2 else 'eng'
    
    if not os.path.isdir(directory):
        print(f"Invalid directory: {directory}")
        sys.exit(1)

    print(f"Scanning directory: {directory} for language: {language}")
    download_subtitles_for_directory(directory, languages=[language])