139 lines
5.1 KiB
Python
139 lines
5.1 KiB
Python
import subprocess
|
|
from pathlib import Path
|
|
from typing import Union
|
|
|
|
import yt_dlp
|
|
from mergedeep import merge
|
|
|
|
|
|
class YDL:
|
|
def __init__(self, ydl_opts: dict = None, extra_ydlp_opts: dict = None):
|
|
self.ydl_opts = ydl_opts if ydl_opts else {}
|
|
extra_ydlp_opts = extra_ydlp_opts if extra_ydlp_opts else {}
|
|
self.ydl_opts = merge(ydl_opts, extra_ydlp_opts)
|
|
self.ydl_opts['logger'] = self.ydl_opts.get('logger')
|
|
self.yt_dlp = yt_dlp.YoutubeDL(ydl_opts)
|
|
|
|
def get_formats(self, url: Union[str, Path]) -> tuple:
|
|
"""
|
|
Not used since we're letting youtube-dl manage filesize filters for us.
|
|
"""
|
|
sizes = []
|
|
with self.yt_dlp as ydl:
|
|
for video in ydl.extract_info(url, download=False)['formats']:
|
|
d = {
|
|
'format_id': video['format_id'],
|
|
'format_note': video['format_note'],
|
|
}
|
|
if video.get('filesize'):
|
|
d['filesize'] = round(video['filesize'] / 1e+6, 1) # MB
|
|
else:
|
|
d['filesize'] = -1
|
|
sizes.append(d)
|
|
return tuple(sizes)
|
|
|
|
def playlist_contents(self, url: str) -> Union[dict, bool]:
|
|
ydl_opts = {
|
|
'extract_flat': True,
|
|
'skip_download': True,
|
|
'ignoreerrors': True,
|
|
'logger': self.ydl_opts['logger'],
|
|
}
|
|
with yt_dlp.YoutubeDL(ydl_opts) as ydl:
|
|
info = self.get_info(url)
|
|
if not info:
|
|
return False
|
|
entries = []
|
|
if info['_type'] == 'playlist':
|
|
if 'entries' in info.keys():
|
|
# When downloading a channel youtube-dl returns a playlist for videos and another for shorts.
|
|
# We need to combine all the videos into one list.
|
|
for item in info['entries']:
|
|
if item['_type'] in ('video', 'url'):
|
|
entries.append(item)
|
|
elif item['_type'] == 'playlist':
|
|
for video in self.get_info(item['webpage_url'])['entries']:
|
|
entries.append(video)
|
|
else:
|
|
raise ValueError(f"Unknown sub-media type: {item['_type']}")
|
|
elif info['_type'] == 'video':
|
|
# `info` doesn't seem to contain the `url` key so we'll add it manually.
|
|
# If any issues arise in the future make sure to double check there isn't any weirdness going on here.
|
|
entries.append(info)
|
|
entries[0]['url'] = f"https://www.youtube.com/watch?v={info['id']}"
|
|
else:
|
|
raise ValueError(f"Unknown media type: {info['_type']}")
|
|
return {
|
|
'title': info['title'],
|
|
'id': info['id'],
|
|
'entries': entries,
|
|
}
|
|
|
|
# def filter_filesize(self, info, *, incomplete):
|
|
# duration = info.get('duration')
|
|
# if duration and duration < 60:
|
|
# return 'The video is too short'
|
|
|
|
def extract_info(self, *args, **kwargs):
|
|
return self.yt_dlp.extract_info(*args, **kwargs)
|
|
|
|
def prepare_filename(self, *args, **kwargs):
|
|
return self.yt_dlp.prepare_filename(*args, **kwargs)
|
|
|
|
def process_info(self, *args, **kwargs):
|
|
return self.yt_dlp.process_info(*args, **kwargs)
|
|
|
|
def get_info(self, url):
|
|
ydl_opts = {
|
|
'extract_flat': True,
|
|
'skip_download': True,
|
|
'ignoreerrors': True,
|
|
'logger': self.ydl_opts['logger'],
|
|
}
|
|
ydl = yt_dlp.YoutubeDL(ydl_opts)
|
|
return ydl.sanitize_info(ydl.extract_info(url, download=False))
|
|
|
|
def __call__(self, *args, **kwargs):
|
|
return self.yt_dlp.download(*args, **kwargs)
|
|
|
|
|
|
def update_ytdlp():
|
|
package_name = 'yt-dlp'
|
|
try:
|
|
result = subprocess.run(
|
|
["pip", "install", "--disable-pip-version-check", "--upgrade", package_name],
|
|
capture_output=True,
|
|
text=True,
|
|
check=True
|
|
)
|
|
|
|
if f"Successfully installed {package_name}" in result.stdout:
|
|
# print(f"{package_name} was updated.")
|
|
return True
|
|
else:
|
|
# print(f"{package_name} was not updated.")
|
|
return False
|
|
|
|
except subprocess.CalledProcessError as e:
|
|
print(f"An error occurred while updating {package_name}:")
|
|
print(e.output)
|
|
return False
|
|
|
|
|
|
class ytdl_no_logger(object):
|
|
def debug(self, msg):
|
|
return
|
|
|
|
def info(self, msg):
|
|
return
|
|
|
|
def warning(self, msg):
|
|
return
|
|
|
|
def error(self, msg):
|
|
return
|
|
|
|
|
|
def get_output_templ(video_id: str = None, title: str = None, uploader: str = None, uploader_id: str = None, include_ext: bool = True):
|
|
return f'[{video_id if video_id else "%(id)s"}] [{title if title else "%(title)s"}] [{uploader if uploader else "%(uploader)s"}] [{uploader_id if uploader_id else "%(uploader_id)s"}]{".%(ext)s" if include_ext else ""}' # leading dash can cause issues due to bash args so we surround the variables in brackets
|