yt-dlp_e/test.py

96 lines
3.3 KiB
Python

import subprocess
import os
from datetime import datetime
import concurrent.futures
formats = "18"
cookies_file = "cookies.txt"
output_template = "%(id)s/%(id)s.f%(format_id)s.%(ext)s"
output_infojson = "infojson:%(id)s/%(id)s.t%(duration_string)s.%(ext)s"
paths = "~/Downloads/staging"
paths_temp = "temp:~/Downloads/temp"
cache_dir = "~/Downloads/cache"
ffmpeg_location = "~/"
num_threads = 16
url_file = "video_urls.txt"
log_file = "download_log.txt"
successful_downloads_file = "successful_downloads.txt"
failed_downloads_file = "failed_downloads.txt"
def load_downloaded_urls(file):
if os.path.exists(file):
with open(file, "r") as f:
return set(line.strip() for line in f if line.strip())
return set()
def save_downloaded_url(file, url):
with open(file, "a") as f:
f.write(f"{url}\n")
def download_video(url):
if url in successful_downloads:
print(f"Video {url} already downloaded. Skipping.")
return
if url in failed_downloads:
print(f"Video {url} previously failed. Retrying.")
start_time = datetime.now()
try:
command = [
"yt-dlp",
"--paths", paths,
"--paths", paths_temp,
"--cache-dir", cache_dir,
"--ffmpeg-location", ffmpeg_location,
"--format", formats,
"--cookies", cookies_file,
"--output", output_template,
"--output", output_infojson,
"--ignore-config",
"--no-progress",
"--write-info-json",
url
]
print(f"Start: {' '.join(command)}")
result = subprocess.run(command, check=True, capture_output=True, text=True)
end_time = datetime.now()
log_download(url, start_time, end_time)
save_downloaded_url(successful_downloads_file, url)
except subprocess.CalledProcessError as e:
error_message = e.stderr if e.stderr else str(e)
print(f"ERROR VIDEO {url}: {error_message}")
if "This video has been removed" in error_message or "Video unavailable" in error_message:
save_downloaded_url(failed_downloads_file, url)
if "Sign in to confirm youre not a bot" in error_message:
print("Critical error detected: Sign-in required. Stopping execution.")
exit(1)
def log_download(url, start_time, end_time):
with open(log_file, "a") as log:
log.write(f"{url} Download.\nStart: {start_time}\nEnd: {end_time}\n\n")
if not os.path.exists(url_file):
print(f"File {url_file} Empty.")
exit(1)
successful_downloads = load_downloaded_urls(successful_downloads_file)
failed_downloads = load_downloaded_urls(failed_downloads_file)
with open(url_file, "r") as file:
video_urls = [line.strip() for line in file if line.strip()]
video_urls = [url for url in video_urls if url not in successful_downloads]
os.makedirs(os.path.expanduser("~/Downloads/staging"), exist_ok=True)
os.makedirs(os.path.expanduser("~/Downloads/temp"), exist_ok=True)
os.makedirs(os.path.expanduser("~/Downloads/cache"), exist_ok=True)
with concurrent.futures.ThreadPoolExecutor(max_workers=num_threads) as executor:
futures = [executor.submit(download_video, url) for url in video_urls]
for future in concurrent.futures.as_completed(futures):
try:
future.result()
except Exception as e:
print(f"ERROR: {e}")