YouTube_predownloader/main.py

import os
import time
import datetime
import pickle
import threading
from googleapiclient.discovery import build
import subprocess
from google.auth.transport.requests import Request
import sys

# Constants
OUTPUT_DIR = "/media/media/YouTube"  # Change this to your Emby folder
DAYS_BACK = 5  # Videos published in the last 5 days
DAYS_TO_KEEP = 10  # Delete videos older than 10 days
TOKEN_FILE = "youtube_token.pickle"  # Token file from the OAuth script
EXIT_TIME = datetime.time(6, 0)  # Exit the script at 06:00
POPULAR_VIDEOS_MAX_RESULTS = 10  # Number of popular videos to fetch (you can change this)
REGION_CODES = ["US", "GB", "LU"]  # Regions to fetch popular videos from

def update_yt_dlp():
    """Update yt-dlp to the latest version."""
    print("Updating yt-dlp to the latest version...")
    subprocess.run(["pip", "install", "--upgrade", "yt-dlp"], check=True)
    print("yt-dlp updated successfully!")

def authenticate_with_token():
    """Authenticate using the saved token."""
    if not os.path.exists(TOKEN_FILE):
        raise FileNotFoundError(f"Token file '{TOKEN_FILE}' not found. Please run the token retrieval script first.")

    with open(TOKEN_FILE, "rb") as token:
        creds = pickle.load(token)

    # Refresh the token if needed
    if creds and creds.expired and creds.refresh_token:
        creds.refresh(Request())

    youtube = build("youtube", "v3", credentials=creds)
    return youtube

def get_subscribed_channels(youtube):
    """Fetch the list of subscribed channels."""
    channels = []
    request = youtube.subscriptions().list(
        part="snippet",
        mine=True,
        maxResults=50
    )
    while request:
        response = request.execute()
        for item in response["items"]:
            channels.append(item["snippet"]["resourceId"]["channelId"])
        request = youtube.subscriptions().list_next(request, response)
    return channels

def get_recent_videos(youtube, channel_id):
    """Fetch videos from a channel published in the last `DAYS_BACK` days."""
    recent_videos = []
    published_after = (datetime.datetime.now() - datetime.timedelta(days=DAYS_BACK)).isoformat("T") + "Z"
    
    request = youtube.search().list(
        part="snippet",
        channelId=channel_id,
        publishedAfter=published_after,
        maxResults=50,
        type="video",
        order="date"
    )
    response = request.execute()
    for item in response["items"]:
        recent_videos.append({
            "video_id": item["id"]["videoId"],
            "title": item["snippet"]["title"],
            "description": item["snippet"]["description"],
            "published_at": item["snippet"]["publishedAt"],
            "channel_title": item["snippet"]["channelTitle"]
        })
    return recent_videos

def get_popular_videos(youtube, region_code):
    """Fetch the most popular videos for a given region."""
    popular_videos = []
    
    request = youtube.videos().list(
        part="snippet,statistics",
        chart="mostPopular",
        regionCode=region_code,  # Region code passed here
        maxResults=POPULAR_VIDEOS_MAX_RESULTS
    )
    response = request.execute()
    for item in response["items"]:
        popular_videos.append({
            "video_id": item["id"],
            "title": item["snippet"]["title"],
            "description": item["snippet"]["description"],
            "published_at": item["snippet"]["publishedAt"],
            "channel_title": item["snippet"]["channelTitle"]
        })
    return popular_videos

def download_video(video_id, output_path):
    """Download a YouTube video in the best quality, handling errors gracefully."""
    try:
        command = [
            "yt-dlp",
            "-f", "bestvideo+bestaudio/best",  # Best available format
            f"https://www.youtube.com/watch?v={video_id}",
            "-o", os.path.join(output_path, "%(title)s.%(ext)s")
        ]
        subprocess.run(command, check=True)
    except subprocess.CalledProcessError as e:
        print(f"Error downloading video {video_id}: {e}")
        return False  # Return False to indicate failure
    return True  # Return True if download is successful

# Temporarily commented out video conversion functionality
# def convert_to_mp4(input_file):
#     """Convert a video to mp4 and delete the original file."""
#     mp4_file = os.path.splitext(input_file)[0] + ".mp4"
#     print(f"Converting {input_file} to {mp4_file}...")
#     command = ["ffmpeg", "-i", input_file, "-c:v", "libx264", "-preset", "fast", "-c:a", "aac", mp4_file]
#     subprocess.run(command, check=True)
#     print(f"Conversion completed: {mp4_file}")
#     if os.path.exists(mp4_file):
#         os.remove(input_file)
#         print(f"Original file deleted: {input_file}")

def create_nfo_file(video_data, file_path):
    """Create an Emby-compatible NFO file."""
    nfo_content = f"""<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<movie>
  <title>{video_data['title']}</title>
  <plot>{video_data['description']}</plot>
  <year>{video_data['published_at'][:4]}</year>
  <premiered>{video_data['published_at']}</premiered>
  <studio>{video_data['channel_title']}</studio>
</movie>
"""
    with open(file_path, "w", encoding="utf-8") as nfo_file:
        nfo_file.write(nfo_content)

def delete_old_videos(folder, days=10):
    """Delete video files and corresponding NFO files older than `days`."""
    now = time.time()
    cutoff = now - (days * 86400)  # Convert days to seconds

    for root, dirs, files in os.walk(folder):
        for file in files:
            file_path = os.path.join(root, file)
            if os.path.isfile(file_path):
                # Check the file's last modification time
                if os.path.getmtime(file_path) < cutoff:
                    print(f"Deleting: {file_path}")
                    os.remove(file_path)

def video_exists(output_path, video_id):
    """Check if a video already exists in the directory."""
    return os.path.exists(os.path.join(output_path, video_id))

def check_exit_time():
    """Check if the current time is past the exit time."""
    now = datetime.datetime.now().time()
    if now >= EXIT_TIME:
        print("Current time exceeds exit time. Terminating script.")
        sys.exit(0)

def main():
    # Update yt-dlp
    update_yt_dlp()

    # Authenticate YouTube API
    print("Authenticating with token...")
    youtube = authenticate_with_token()

    if not os.path.exists(OUTPUT_DIR):
        os.makedirs(OUTPUT_DIR)

    # Clean up old videos
    print(f"Deleting videos older than {DAYS_TO_KEEP} days in {OUTPUT_DIR}...")
    delete_old_videos(OUTPUT_DIR, days=DAYS_TO_KEEP)
    
    # Fetch subscribed channels
    print("Fetching subscribed channels...")
    channels = get_subscribed_channels(youtube)
    
    # Fetch and process recent videos from subscribed channels
    for channel_id in channels:
        print(f"Fetching recent videos for channel: {channel_id}")
        videos = get_recent_videos(youtube, channel_id)
        
        for video in videos:
            # Check the exit time before processing each video
            check_exit_time()
            
            print(f"Processing video: {video['title']}")
            video_dir = os.path.join(OUTPUT_DIR, video["video_id"])
            
            # Check if the video already exists
            if os.path.exists(video_dir):
                print(f"Skipping download, already exists: {video['title']} ({video['video_id']})")
                continue
            
            os.makedirs(video_dir, exist_ok=True)
            
            # Download the video
            if not download_video(video["video_id"], video_dir):
                print(f"Skipping video {video['video_id']} due to download error.")
                continue  # Skip this video if download failed
            
            # Temporarily comment out conversion, but keep it in the code for future use
            # video_file = next(
            #     (f for f in os.listdir(video_dir) if not f.endswith(".mp4")),
            #     None
            # )
            # if video_file:
            #     input_path = os.path.join(video_dir, video_file)
            #     threading.Thread(target=convert_to_mp4, args=(input_path,)).start()
            
            # Create the NFO file
            nfo_file_path = os.path.join(video_dir, "movie.nfo")
            create_nfo_file(video, nfo_file_path)

    # Fetch and process most popular videos worldwide (from multiple regions)
    for region_code in REGION_CODES:
        print(f"Fetching the most popular videos from {region_code}...")
        popular_videos = get_popular_videos(youtube, region_code)

        for video in popular_videos:
            # Check the exit time before processing each video
            check_exit_time()
            
            print(f"Processing popular video: {video['title']} ({region_code})")
            video_dir = os.path.join(OUTPUT_DIR, video["video_id"])
            
            # Check if the video already exists
            if os.path.exists(video_dir):
                print(f"Skipping download, already exists: {video['title']} ({video['video_id']})")
                continue
            
            os.makedirs(video_dir, exist_ok=True)
            
            # Download the popular video
            if not download_video(video["video_id"], video_dir):
                print(f"Skipping popular video {video['video_id']} due to download error.")
                continue  # Skip this video if download failed
            
            # Temporarily comment out conversion, but keep it in the code for future use
            # video_file = next(
            #     (f for f in os.listdir(video_dir) if not f.endswith(".mp4")),
            #     None
            # )
            # if video_file:
            #     input_path = os.path.join(video_dir, video_file)
            #     threading.Thread(target=convert_to_mp4, args=(input_path,)).start()
            
            # Create the NFO file
            nfo_file_path = os.path.join(video_dir, "movie.nfo")
            create_nfo_file(video, nfo_file_path)
    
    print("All videos processed successfully!")

if __name__ == "__main__":
    main()