def extract_info(video_id, use_invidious, playlist_id=None, index=None):
# bpctr=9999999999 will bypass are-you-sure dialogs for controversial
# videos
url = 'https://m.youtube.com/watch?v=' + video_id + '&pbj=1&bpctr=9999999999'
if playlist_id:
url += '&list=' + playlist_id
if index:
url += '&index=' + index
polymer_json = util.fetch_url(url, headers=headers, debug_name='watch')
polymer_json = polymer_json.decode('utf-8')
# TODO: Decide whether this should be done in yt_data_extract.extract_watch_info
try:
polymer_json = json.loads(polymer_json)
except json.decoder.JSONDecodeError:
traceback.print_exc()
return {'error': 'Failed to parse json response'}
info = yt_data_extract.extract_watch_info(polymer_json)
# age restriction bypass
if info['age_restricted']:
print('Fetching age restriction bypass page')
data = {
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
}
url = 'https://www.youtube.com/get_video_info?' + urllib.parse.urlencode(
data)
video_info_page = util.fetch_url(
url,
debug_name='get_video_info',
report_text='Fetched age restriction bypass page').decode('utf-8')
yt_data_extract.update_with_age_restricted_info(info, video_info_page)
# signature decryption
decryption_error = decrypt_signatures(info)
if decryption_error:
decryption_error = 'Error decrypting url signatures: ' + decryption_error
info['playability_error'] = decryption_error
# check if urls ready (non-live format) in former livestream
# urls not ready if all of them have no filesize
if info['was_live']:
info['urls_ready'] = False
for fmt in info['formats']:
if fmt['file_size'] is not None:
info['urls_ready'] = True
else:
info['urls_ready'] = True
# livestream urls
# sometimes only the livestream urls work soon after the livestream is over
if (info['hls_manifest_url'] and
(info['live'] or not info['formats'] or not info['urls_ready'])):
manifest = util.fetch_url(
info['hls_manifest_url'],
debug_name='hls_manifest.m3u8',
report_text='Fetched hls manifest').decode('utf-8')
info['hls_formats'], err = yt_data_extract.extract_hls_formats(
manifest)
if not err:
info['playability_error'] = None
for fmt in info['hls_formats']:
fmt['video_quality'] = video_quality_string(fmt)
else:
info['hls_formats'] = []
# check for 403
info['invidious_used'] = False
info['invidious_reload_button'] = False
if settings.route_tor and info['formats'] and info['formats'][0]['url']:
try:
response = util.head(info['formats'][0]['url'],
report_text='Checked for URL access')
except urllib3.exceptions.HTTPError:
print('Error while checking for URL access:\n')
traceback.print_exc()
return info
if response.status == 403:
print('Access denied (403) for video urls.')
if use_invidious:
print(' Retrieving urls from Invidious...')
info['invidious_used'] = True
try:
video_info = util.fetch_url(
'https://invidio.us/api/v1/videos/' + video_id +
'?fields=adaptiveFormats,formatStreams',
report_text='Retrieved urls from Invidious',
debug_name='invidious_urls')
except (util.FetchError, urllib3.exceptions.HTTPError) as e:
traceback.print_exc()
playability_error = (
'Access denied (403) for video urls.' +
' Failed to use Invidious to get the urls: ' + str(e))
if info['playability_error']:
info['playability_error'] += '\n' + playability_error
else:
info['playability_error'] = playability_error
# include button to reload without invidious
info['invidious_reload_button'] = True
return info
video_info = json.loads(video_info.decode('utf-8'))
# collect invidious urls for each itag
itag_to_url = {}
for invidious_fmt in (video_info['adaptiveFormats'] +
video_info['formatStreams']):
itag_to_url[invidious_fmt['itag']] = invidious_fmt['url']
# replace urls with urls from invidious
for fmt in info['formats']:
itag = str(fmt['itag'])
if itag not in itag_to_url:
print(('Warning: itag ' + itag +
' not found in invidious urls'))
continue
fmt['url'] = itag_to_url[itag]
else:
info['playability_error'] = ('Access denied (403) for video '
'urls')
elif 300 <= response.status < 400:
print('Error: exceeded max redirects while checking video URL')
return info
def extract_info(video_id, use_invidious, playlist_id=None, index=None):
# bpctr=9999999999 will bypass are-you-sure dialogs for controversial
# videos
url = 'https://m.youtube.com/embed/' + video_id + '?bpctr=9999999999'
if playlist_id:
url += '&list=' + playlist_id
if index:
url += '&index=' + index
watch_page = util.fetch_url(url, headers=watch_headers, debug_name='watch')
watch_page = watch_page.decode('utf-8')
info = yt_data_extract.extract_watch_info_from_html(watch_page)
context = {
'client': {
'clientName': 'ANDROID',
'clientVersion': '16.20',
'gl': 'US',
'hl': 'en',
},
# https://github.com/yt-dlp/yt-dlp/pull/575#issuecomment-887739287
'thirdParty': {
'embedUrl': 'https://google.com', # Can be any valid URL
}
}
if info['age_restricted'] or info['player_urls_missing']:
if info['age_restricted']:
print('Age restricted video. Fetching /youtubei/v1/player page')
else:
print('Missing player. Fetching /youtubei/v1/player page')
context['client']['clientScreen'] = 'EMBED'
else:
print('Fetching /youtubei/v1/player page')
# https://github.com/yt-dlp/yt-dlp/issues/574#issuecomment-887171136
# ANDROID is used instead because its urls don't require decryption
# The URLs returned with WEB for videos requiring decryption
# couldn't be decrypted with the base.js from the web page for some
# reason
url = 'https://youtubei.googleapis.com/youtubei/v1/player'
url += '?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'
data = {
'videoId': video_id,
'context': context,
}
data = json.dumps(data)
content_header = (('Content-Type', 'application/json'), )
player_response = util.fetch_url(
url,
data=data,
headers=util.mobile_ua + content_header,
debug_name='youtubei_player',
report_text='Fetched youtubei player page').decode('utf-8')
yt_data_extract.update_with_age_restricted_info(info, player_response)
# signature decryption
decryption_error = decrypt_signatures(info, video_id)
if decryption_error:
decryption_error = 'Error decrypting url signatures: ' + decryption_error
info['playability_error'] = decryption_error
# check if urls ready (non-live format) in former livestream
# urls not ready if all of them have no filesize
if info['was_live']:
info['urls_ready'] = False
for fmt in info['formats']:
if fmt['file_size'] is not None:
info['urls_ready'] = True
else:
info['urls_ready'] = True
# livestream urls
# sometimes only the livestream urls work soon after the livestream is over
if (info['hls_manifest_url'] and
(info['live'] or not info['formats'] or not info['urls_ready'])):
manifest = util.fetch_url(
info['hls_manifest_url'],
debug_name='hls_manifest.m3u8',
report_text='Fetched hls manifest').decode('utf-8')
info['hls_formats'], err = yt_data_extract.extract_hls_formats(
manifest)
if not err:
info['playability_error'] = None
for fmt in info['hls_formats']:
fmt['video_quality'] = video_quality_string(fmt)
else:
info['hls_formats'] = []
# check for 403. Unnecessary for tor video routing b/c ip address is same
info['invidious_used'] = False
info['invidious_reload_button'] = False
info['tor_bypass_used'] = False
if (settings.route_tor == 1 and info['formats']
and info['formats'][0]['url']):
try:
response = util.head(info['formats'][0]['url'],
report_text='Checked for URL access')
except urllib3.exceptions.HTTPError:
print('Error while checking for URL access:\n')
traceback.print_exc()
return info
if response.status == 403:
print('Access denied (403) for video urls.')
print('Routing video through Tor')
info['tor_bypass_used'] = True
for fmt in info['formats']:
fmt['url'] += '&use_tor=1'
elif 300 <= response.status < 400:
print('Error: exceeded max redirects while checking video URL')
return info
def extract_info(video_id, use_invidious, playlist_id=None, index=None):
# bpctr=9999999999 will bypass are-you-sure dialogs for controversial
# videos
url = 'https://m.youtube.com/watch?v=' + video_id + '&pbj=1&bpctr=9999999999'
if playlist_id:
url += '&list=' + playlist_id
if index:
url += '&index=' + index
polymer_json = util.fetch_url(url, headers=headers, debug_name='watch')
polymer_json = polymer_json.decode('utf-8')
# TODO: Decide whether this should be done in yt_data_extract.extract_watch_info
try:
polymer_json = json.loads(polymer_json)
except json.decoder.JSONDecodeError:
traceback.print_exc()
return {'error': 'Failed to parse json response'}
info = yt_data_extract.extract_watch_info(polymer_json)
# request player urls if it's missing
# see https://github.com/user234683/youtube-local/issues/22#issuecomment-706395160
if info['age_restricted'] or info['player_urls_missing']:
if info['age_restricted']:
print('Age restricted video. Fetching get_video_info page')
else:
print('Missing player. Fetching get_video_info page')
data = {
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
}
url = 'https://www.youtube.com/get_video_info?' + urllib.parse.urlencode(
data)
video_info_page = util.fetch_url(
url,
debug_name='get_video_info',
report_text='Fetched get_video_info page').decode('utf-8')
yt_data_extract.update_with_age_restricted_info(info, video_info_page)
# signature decryption
decryption_error = decrypt_signatures(info)
if decryption_error:
decryption_error = 'Error decrypting url signatures: ' + decryption_error
info['playability_error'] = decryption_error
# check if urls ready (non-live format) in former livestream
# urls not ready if all of them have no filesize
if info['was_live']:
info['urls_ready'] = False
for fmt in info['formats']:
if fmt['file_size'] is not None:
info['urls_ready'] = True
else:
info['urls_ready'] = True
# livestream urls
# sometimes only the livestream urls work soon after the livestream is over
if (info['hls_manifest_url'] and
(info['live'] or not info['formats'] or not info['urls_ready'])):
manifest = util.fetch_url(
info['hls_manifest_url'],
debug_name='hls_manifest.m3u8',
report_text='Fetched hls manifest').decode('utf-8')
info['hls_formats'], err = yt_data_extract.extract_hls_formats(
manifest)
if not err:
info['playability_error'] = None
for fmt in info['hls_formats']:
fmt['video_quality'] = video_quality_string(fmt)
else:
info['hls_formats'] = []
# check for 403. Unnecessary for tor video routing b/c ip address is same
info['invidious_used'] = False
info['invidious_reload_button'] = False
info['tor_bypass_used'] = False
if (settings.route_tor == 1 and info['formats']
and info['formats'][0]['url']):
try:
response = util.head(info['formats'][0]['url'],
report_text='Checked for URL access')
except urllib3.exceptions.HTTPError:
print('Error while checking for URL access:\n')
traceback.print_exc()
return info
if response.status == 403:
print('Access denied (403) for video urls.')
print('Routing video through Tor')
info['tor_bypass_used'] = True
for fmt in info['formats']:
fmt['url'] += '&use_tor=1'
elif 300 <= response.status < 400:
print('Error: exceeded max redirects while checking video URL')
return info
def extract_info(video_id):
# bpctr=9999999999 will bypass are-you-sure dialogs for controversial
# videos
polymer_json = util.fetch_url('https://m.youtube.com/watch?v=' + video_id +
'&pbj=1&bpctr=9999999999',
headers=headers,
debug_name='watch').decode('utf-8')
# TODO: Decide whether this should be done in yt_data_extract.extract_watch_info
try:
polymer_json = json.loads(polymer_json)
except json.decoder.JSONDecodeError:
traceback.print_exc()
return {'error': 'Failed to parse json response'}
info = yt_data_extract.extract_watch_info(polymer_json)
# age restriction bypass
if info['age_restricted']:
print('Fetching age restriction bypass page')
data = {
'video_id': video_id,
'eurl': 'https://youtube.googleapis.com/v/' + video_id,
}
url = 'https://www.youtube.com/get_video_info?' + urllib.parse.urlencode(
data)
video_info_page = util.fetch_url(
url,
debug_name='get_video_info',
report_text='Fetched age restriction bypass page').decode('utf-8')
yt_data_extract.update_with_age_restricted_info(info, video_info_page)
# signature decryption
decryption_error = decrypt_signatures(info)
if decryption_error:
decryption_error = 'Error decrypting url signatures: ' + decryption_error
info['playability_error'] = decryption_error
# check for 403
info['invidious_used'] = False
if settings.route_tor and info['formats'] and info['formats'][0]['url']:
try:
response = util.head(info['formats'][0]['url'],
report_text='Checked for URL access')
except urllib3.exceptions.HTTPError:
print('Error while checking for URL access:\n')
traceback.print_exc()
return info
if response.status == 403:
print(('Access denied (403) for video urls.'
' Retrieving urls from Invidious...'))
info['invidious_used'] = True
try:
video_info = util.fetch_url(
'https://invidio.us/api/v1/videos/' + video_id +
'?fields=adaptiveFormats,formatStreams',
report_text='Retrieved urls from Invidious',
debug_name='invidious_urls')
except (urllib3.exceptions.HTTPError) as e:
traceback.print_exc()
playability_error = (
'Access denied (403) for video urls.' +
' Failed to use Invidious to get the urls: ' + str(e))
if info['playability_error']:
info['playability_error'] += '\n' + playability_error
else:
info['playability_error'] = playability_error
return info
video_info = json.loads(video_info.decode('utf-8'))
# collect invidious urls for each itag
itag_to_url = {}
for invidious_fmt in (video_info['adaptiveFormats'] +
video_info['formatStreams']):
itag_to_url[invidious_fmt['itag']] = invidious_fmt['url']
# replace urls with urls from invidious
for fmt in info['formats']:
itag = str(fmt['itag'])
if itag not in itag_to_url:
print(('Warning: itag ' + itag +
' not found in invidious urls'))
continue
fmt['url'] = itag_to_url[itag]
elif 300 <= response.status < 400:
print('Error: exceeded max redirects while checking video URL')
return info
