yt-dlp plugin to bypass youtube age-gate

Youtube_AgeGateBypass.py

"""
SPDX-License-Identifier: MIT https://opensource.org/licenses/MIT
Copyright © 2022 [email protected]

yt-dlp plugin extractor to bypass youtube age-gate
Needs yt-dlp 2022.04.08 or above

See https://github.com/yt-dlp/yt-dlp#plugins for how to use

Uses account proxy https://youtube-proxy.zerody.one by https://github.com/zerodytrash
to fetch video formats and proxy service https://www.4everproxy.com for download (if required)

Set "VIDEO_PROXY = None" to disable the proxy service

Note: The account proxy has limited resources. Please do not abuse
"""


import base64
from yt_dlp.utils import get_first, parse_qs
from yt_dlp.extractor.youtube import YoutubeIE


ACCOUNT_PROXY = 'https://youtube-proxy.zerody.one'
VIDEO_PROXY = 'https://phx.4everproxy.com'


class Youtube_AgeGateBypassIE(YoutubeIE):
    IE_NAME = 'youtube:AGB'
    _VALID_URL = YoutubeIE._VALID_URL

    def _download_player_responses(self, url, smuggled_data, video_id, *args, **kwargs):
        AGB_CLIENT = 'web'

        ret = super()._download_player_responses(url, smuggled_data, video_id, *args, **kwargs)
        _, ytcfg, player_responses, player_url, *_ = ret

        #if all(map(self._is_agegated, player_responses)):
        if True:
            pr = self._download_json(
                f'{ACCOUNT_PROXY}/getPlayer', video_id,
                'Downloading Zerody API JSON', fatal=False, query={
                    'videoId': video_id,
                    'clientName': self._extract_client_name(ytcfg, AGB_CLIENT),
                    'clientVersion': self._extract_client_version(ytcfg, AGB_CLIENT),
                    'signatureTimestamp': self._extract_signature_timestamp(video_id, player_url, ytcfg),
                    'reason': 'LOGIN_REQUIRED',
                    'startTimeSecs': 0,
                    'client': 'yt-dlp',
                })
            if pr:
                player_responses.append(pr)
                streaming_data = get_first(player_responses, 'streamingData') or {}
                if VIDEO_PROXY:
                    streaming_data['_use_proxy'] = True
        return ret

    def _extract_formats(self, streaming_data, *args, **kwargs):
        has_gcr = False
        for f in super()._extract_formats(streaming_data, *args, **kwargs):
            if get_first(streaming_data, '_use_proxy') and parse_qs(f.get('url')).get('gcr'):
                has_gcr = True
                f['url'] = f'{VIDEO_PROXY}/direct/' + base64.b64encode(f['url'].encode()).decode()
            yield f
        if has_gcr:
            self.to_screen('Video formats have GCR. Using proxy for download')

    # Pretend to be YoutubeIE
    def _real_extract(self, url):
        res = super()._real_extract(url)
        res['extractor_key'] = YoutubeIE.ie_key()
        return res