from ..swfinterp import SWFInterpreter
from ..compat import (
compat_chr,
+ compat_HTTPError,
compat_kwargs,
compat_parse_qs,
compat_urllib_parse_unquote,
)
from ..utils import (
clean_html,
+ dict_get,
error_to_compat_str,
ExtractorError,
float_or_none,
orderedSet,
parse_codecs,
parse_duration,
+ qualities,
remove_quotes,
remove_start,
smuggle_url,
+ str_or_none,
str_to_int,
try_get,
unescapeHTML,
unified_strdate,
unsmuggle_url,
uppercase_escape,
+ url_or_none,
urlencode_postdata,
)
# If True it will raise an error if no login info is provided
_LOGIN_REQUIRED = False
- _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
+ _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
def _set_language(self):
self._set_cookie(
If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
"""
- (username, password) = self._get_login_info()
+ username, password = self._get_login_info()
# No authentication to be performed
if username is None:
- if self._LOGIN_REQUIRED:
+ if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
return True
warn('Unable to extract result entry')
return False
- tfa = try_get(res, lambda x: x[0][0], list)
- if tfa:
- tfa_str = try_get(tfa, lambda x: x[2], compat_str)
- if tfa_str == 'TWO_STEP_VERIFICATION':
+ login_challenge = try_get(res, lambda x: x[0][0], list)
+ if login_challenge:
+ challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
+ if challenge_str == 'TWO_STEP_VERIFICATION':
# SEND_SUCCESS - TFA code has been successfully sent to phone
# QUOTA_EXCEEDED - reached the limit of TFA codes
- status = try_get(tfa, lambda x: x[5], compat_str)
+ status = try_get(login_challenge, lambda x: x[5], compat_str)
if status == 'QUOTA_EXCEEDED':
warn('Exceeded the limit of TFA codes, try later')
return False
check_cookie_url = try_get(
tfa_results, lambda x: x[0][-1][2], compat_str)
+ else:
+ CHALLENGES = {
+ 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
+ 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
+ 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
+ }
+ challenge = CHALLENGES.get(
+ challenge_str,
+ '%s returned error %s.' % (self.IE_NAME, challenge_str))
+ warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
+ return False
else:
check_cookie_url = try_get(res, lambda x: x[2], compat_str)
return True
- def _download_webpage(self, *args, **kwargs):
- kwargs.setdefault('query', {})['disable_polymer'] = 'true'
- return super(YoutubeBaseInfoExtractor, self)._download_webpage(
+ def _download_webpage_handle(self, *args, **kwargs):
+ query = kwargs.get('query', {}).copy()
+ query['disable_polymer'] = 'true'
+ kwargs['query'] = query
+ return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
*args, **compat_kwargs(kwargs))
def _real_initialize(self):
if not mobj:
break
- more = self._download_json(
- 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
- 'Downloading page #%s' % page_num,
- transform_source=uppercase_escape)
+ count = 0
+ retries = 3
+ while count <= retries:
+ try:
+ # Downloading page may result in intermittent 5xx HTTP error
+ # that is usually worked around with a retry
+ more = self._download_json(
+ 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
+ 'Downloading page #%s%s'
+ % (page_num, ' (retry #%d)' % count if count else ''),
+ transform_source=uppercase_escape)
+ break
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
+ count += 1
+ if count <= retries:
+ continue
+ raise
+
content_html = more['content_html']
if not content_html.strip():
# Some webpages show a "Load more" button but they don't
(?:www\.)?hooktube\.com/|
(?:www\.)?yourepeat\.com/|
tube\.majestyc\.net/|
+ (?:(?:www|dev)\.)?invidio\.us/|
+ (?:www\.)?invidiou\.sh/|
+ (?:www\.)?invidious\.snopyta\.org/|
+ (?:www\.)?invidious\.kabi\.tk/|
+ (?:www\.)?vid\.wxzm\.sx/|
youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
'135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
'137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
- '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
+ '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
'160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
'212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
# RTMP (unnamed)
'_rtmp': {'protocol': 'rtmp'},
}
- _SUBTITLE_FORMATS = ('ttml', 'vtt')
+ _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
_GEO_BYPASS = False
'uploader': 'Philipp Hagemeister',
'uploader_id': 'phihag',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
+ 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
+ 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
'upload_date': '20121002',
- 'license': 'Standard YouTube License',
'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
'categories': ['Science & Technology'],
'tags': ['youtube-dl'],
'duration': 10,
+ 'view_count': int,
'like_count': int,
'dislike_count': int,
'start_time': 1,
'uploader': 'Icona Pop',
'uploader_id': 'IconaPop',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
- 'license': 'Standard YouTube License',
'creator': 'Icona Pop',
+ 'track': 'I Love It (feat. Charli XCX)',
+ 'artist': 'Icona Pop',
}
},
{
'id': '07FYdnEawAQ',
'ext': 'mp4',
'upload_date': '20130703',
- 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
+ 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
'alt_title': 'Tunnel Vision',
- 'description': 'md5:64249768eec3bc4276236606ea996373',
+ 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
'duration': 419,
'uploader': 'justintimberlakeVEVO',
'uploader_id': 'justintimberlakeVEVO',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
- 'license': 'Standard YouTube License',
'creator': 'Justin Timberlake',
+ 'track': 'Tunnel Vision',
+ 'artist': 'Justin Timberlake',
'age_limit': 18,
}
},
'uploader': 'SET India',
'uploader_id': 'setindia',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
- 'license': 'Standard YouTube License',
'age_limit': 18,
}
},
'uploader_id': 'phihag',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
'upload_date': '20121002',
- 'license': 'Standard YouTube License',
'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
'categories': ['Science & Technology'],
'tags': ['youtube-dl'],
'duration': 10,
+ 'view_count': int,
'like_count': int,
'dislike_count': int,
},
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
'description': '',
'uploader': '8KVIDEO',
- 'license': 'Standard YouTube License',
'title': 'UHDTV TEST 8K VIDEO.mp4'
},
'params': {
'info_dict': {
'id': 'IB3lcPjvWLA',
'ext': 'm4a',
- 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
- 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
+ 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
+ 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
'duration': 244,
'uploader': 'AfrojackVEVO',
'uploader_id': 'AfrojackVEVO',
'upload_date': '20131011',
- 'license': 'Standard YouTube License',
},
'params': {
'youtube_include_dash_manifest': True,
'id': 'nfWlot6h_JM',
'ext': 'm4a',
'title': 'Taylor Swift - Shake It Off',
- 'alt_title': 'Shake It Off',
- 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
+ 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
'duration': 242,
'uploader': 'TaylorSwiftVEVO',
'uploader_id': 'TaylorSwiftVEVO',
'upload_date': '20140818',
- 'license': 'Standard YouTube License',
'creator': 'Taylor Swift',
},
'params': {
'ext': 'mp4',
'duration': 219,
'upload_date': '20100909',
- 'uploader': 'The Amazing Atheist',
+ 'uploader': 'Amazing Atheist',
'uploader_id': 'TheAmazingAtheist',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
- 'license': 'Standard YouTube License',
'title': 'Burning Everyone\'s Koran',
'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
}
'uploader_id': 'WitcherGame',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
'upload_date': '20140605',
- 'license': 'Standard YouTube License',
'age_limit': 18,
},
},
'ext': 'mp4',
'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
- 'duration': 247,
+ 'duration': 246,
'uploader': 'LloydVEVO',
'uploader_id': 'LloydVEVO',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
'upload_date': '20110629',
- 'license': 'Standard YouTube License',
'age_limit': 18,
},
},
- # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
+ # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
# YouTube Red ad is not captured for creator
{
'url': '__2ABJjxzNo',
'creator': 'deadmau5',
'description': 'md5:12c56784b8032162bb936a5f76d55360',
'uploader': 'deadmau5',
- 'license': 'Standard YouTube License',
'title': 'Deadmau5 - Some Chords (HD)',
'alt_title': 'Some Chords',
},
'DASH manifest missing',
]
},
- # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
+ # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
{
'url': 'lqQg6PlCWgI',
'info_dict': {
'upload_date': '20150827',
'uploader_id': 'olympic',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
- 'license': 'Standard YouTube License',
'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
'uploader': 'Olympic',
'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
'uploader_id': 'AllenMeow',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
- 'uploader': '孫艾倫',
- 'license': 'Standard YouTube License',
+ 'uploader': '孫ᄋᄅ',
'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
},
},
},
'skip': 'This live event has ended.',
},
- # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
+ # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
{
'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
'info_dict': {
'id': 'FIl7x6_3R5Y',
- 'ext': 'mp4',
+ 'ext': 'webm',
'title': 'md5:7b81415841e02ecd4313668cde88737a',
'description': 'md5:116377fd2963b81ec4ce64b542173306',
'duration': 220,
'uploader_id': 'dorappi2000',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
'uploader': 'dorappi2000',
- 'license': 'Standard YouTube License',
- 'formats': 'mincount:32',
+ 'formats': 'mincount:31',
},
+ 'skip': 'not actual anymore',
},
# DASH manifest with segment_list
{
'uploader': 'Airtek',
'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
- 'license': 'Standard YouTube License',
'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
},
'params': {
'params': {
'skip_download': True,
},
+ 'skip': 'This video is not available.',
},
{
- # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
+ # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
'info_dict': {
'id': 'gVfLd0zydlo',
'only_matching': True,
},
{
- # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
# Also tests cut-off URL expansion in video description (see
- # https://github.com/rg3/youtube-dl/issues/1892,
- # https://github.com/rg3/youtube-dl/issues/8164)
+ # https://github.com/ytdl-org/youtube-dl/issues/1892,
+ # https://github.com/ytdl-org/youtube-dl/issues/8164)
'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
'info_dict': {
'id': 'lsguqyKfVQg',
'ext': 'mp4',
'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
- 'alt_title': 'Dark Walk',
+ 'alt_title': 'Dark Walk - Position Music',
'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
'duration': 133,
'upload_date': '20151119',
'uploader_id': 'IronSoulElf',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
'uploader': 'IronSoulElf',
- 'license': 'Standard YouTube License',
- 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
+ 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
+ 'track': 'Dark Walk - Position Music',
+ 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
+ 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
},
'params': {
'skip_download': True,
},
},
{
- # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
'only_matching': True,
},
'description': 'md5:dda0d780d5a6e120758d1711d062a867',
'duration': 4060,
'upload_date': '20151119',
- 'uploader': 'Bernie 2016',
+ 'uploader': 'Bernie Sanders',
'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
'license': 'Creative Commons Attribution license (reuse allowed)',
'only_matching': True,
},
{
- # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
+ # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
'only_matching': True,
},
'params': {
'skip_download': True,
},
+ 'skip': 'This video is not available.',
},
{
# YouTube Red video with episode data
'id': 'iqKdEhx-dD4',
'ext': 'mp4',
'title': 'Isolation - Mind Field (Ep 1)',
- 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
+ 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
'duration': 2085,
'upload_date': '20170118',
'uploader': 'Vsauce',
'uploader_id': 'Vsauce',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
- 'license': 'Standard YouTube License',
'series': 'Mind Field',
'season_number': 1,
'episode_number': 1,
'uploader': 'New Century Foundation',
'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
- 'license': 'Standard YouTube License',
- 'view_count': int,
},
'params': {
'skip_download': True,
'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
'only_matching': True,
},
+ {
+ 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
+ 'only_matching': True,
+ },
+ {
+ # DRM protected
+ 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
+ 'only_matching': True,
+ },
+ {
+ # Video with unsupported adaptive stream type formats
+ 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
+ 'info_dict': {
+ 'id': 'Z4Vy8R84T1U',
+ 'ext': 'mp4',
+ 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
+ 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+ 'duration': 433,
+ 'upload_date': '20130923',
+ 'uploader': 'Amelia Putri Harwita',
+ 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
+ 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
+ 'formats': 'maxcount:10',
+ },
+ 'params': {
+ 'skip_download': True,
+ 'youtube_include_dash_manifest': False,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
+ 'info_dict': {
+ 'id': 'MgNrAu2pzNs',
+ 'ext': 'mp4',
+ 'title': 'Voyeur Girl',
+ 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
+ 'upload_date': '20190312',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
+ 'artist': 'Stephen',
+ 'track': 'Voyeur Girl',
+ 'album': 'it\'s too much love to know my dear',
+ 'release_date': '20190313',
+ 'release_year': 2019,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # Retrieve 'artist' field from 'Artist:' in video description
+ # when it is present on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
+ 'info_dict': {
+ 'id': 'k0jLE7tTwjY',
+ 'ext': 'mp4',
+ 'title': 'Latch Feat. Sam Smith',
+ 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
+ 'upload_date': '20150110',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
+ 'artist': 'Disclosure',
+ 'track': 'Latch Feat. Sam Smith',
+ 'album': 'Latch Featuring Sam Smith',
+ 'release_date': '20121008',
+ 'release_year': 2012,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle multiple artists on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
+ 'info_dict': {
+ 'id': '74qn0eJSjpA',
+ 'ext': 'mp4',
+ 'title': 'Eastside',
+ 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
+ 'upload_date': '20180710',
+ 'uploader': 'Benny Blanco - Topic',
+ 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
+ 'artist': 'benny blanco, Halsey, Khalid',
+ 'track': 'Eastside',
+ 'album': 'Eastside',
+ 'release_date': '20180713',
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle youtube music video with release_year and no release_date
+ 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
+ 'info_dict': {
+ 'id': '-hcAI0g-f5M',
+ 'ext': 'mp4',
+ 'title': 'Put It On Me',
+ 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
+ 'upload_date': '20180426',
+ 'uploader': 'Matt Maeson - Topic',
+ 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
+ 'artist': 'Matt Maeson',
+ 'track': 'Put It On Me',
+ 'album': 'The Hearse',
+ 'release_date': None,
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
]
def __init__(self, *args, **kwargs):
def _extract_signature_function(self, video_id, player_url, example_sig):
id_m = re.match(
- r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
+ r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
player_url)
if not id_m:
raise ExtractorError('Cannot identify player %r' % player_url)
def _parse_sig_js(self, jscode):
funcname = self._search_regex(
(r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
- r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
+ r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
jscode, 'Initial JS player signature function name', group='sig')
jsi = JSInterpreter(jscode)
# regex won't capture the whole JSON. Yet working around by trying more
# concrete regex first keeping in mind proper quoted string handling
# to be implemented in future that will replace this workaround (see
- # https://github.com/rg3/youtube-dl/issues/7468,
- # https://github.com/rg3/youtube-dl/pull/7599)
+ # https://github.com/ytdl-org/youtube-dl/issues/7468,
+ # https://github.com/ytdl-org/youtube-dl/pull/7599)
r';ytplayer\.config\s*=\s*({.+?});ytplayer',
r';ytplayer\.config\s*=\s*({.+?});',
)
self._downloader.report_warning(err_msg)
return {}
- def _mark_watched(self, video_id, video_info):
- playback_url = video_info.get('videostats_playback_base_url', [None])[0]
+ def _mark_watched(self, video_id, video_info, player_response):
+ playback_url = url_or_none(try_get(
+ player_response,
+ lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
+ video_info, lambda x: x['videostats_playback_base_url'][0]))
if not playback_url:
return
parsed_playback_url = compat_urlparse.urlparse(playback_url)
if dash_mpd and dash_mpd[0] not in dash_mpds:
dash_mpds.append(dash_mpd[0])
+ def add_dash_mpd_pr(pl_response):
+ dash_mpd = url_or_none(try_get(
+ pl_response, lambda x: x['streamingData']['dashManifestUrl'],
+ compat_str))
+ if dash_mpd and dash_mpd not in dash_mpds:
+ dash_mpds.append(dash_mpd)
+
is_live = None
view_count = None
def extract_view_count(v_info):
return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
+ def extract_token(v_info):
+ return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
+
+ player_response = {}
+
# Get video info
embed_webpage = None
if re.search(r'player-age-gate-content">', video_webpage) is not None:
ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
if ytplayer_config:
args = ytplayer_config['args']
- if args.get('url_encoded_fmt_stream_map'):
+ if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
# Convert to the same format returned by compat_parse_qs
video_info = dict((k, [v]) for k, v in args.items())
add_dash_mpd(video_info)
# Rental video is not rented but preview is available (e.g.
# https://www.youtube.com/watch?v=yYr8q0y5Jfg,
- # https://github.com/rg3/youtube-dl/issues/10532)
+ # https://github.com/ytdl-org/youtube-dl/issues/10532)
if not video_info and args.get('ypc_vid'):
return self.url_result(
args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
if args.get('livestream') == '1' or args.get('live_playback') == 1:
is_live = True
sts = ytplayer_config.get('sts')
+ if not player_response:
+ pl_response = str_or_none(args.get('player_response'))
+ if pl_response:
+ pl_response = self._parse_json(pl_response, video_id, fatal=False)
+ if isinstance(pl_response, dict):
+ player_response = pl_response
if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
+ add_dash_mpd_pr(player_response)
# We also try looking in get_video_info since it may contain different dashmpd
# URL that points to a DASH manifest with possibly different itag set (some itags
# are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
# manifest pointed by get_video_info's dashmpd).
# The general idea is to take a union of itags of both DASH manifests (for example
- # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
+ # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
self.report_video_info_webpage_download(video_id)
- for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
+ for el in ('embedded', 'detailpage', 'vevo', ''):
query = {
'video_id': video_id,
'ps': 'default',
if not video_info_webpage:
continue
get_video_info = compat_parse_qs(video_info_webpage)
+ if not player_response:
+ pl_response = get_video_info.get('player_response', [None])[0]
+ if isinstance(pl_response, dict):
+ player_response = pl_response
+ add_dash_mpd_pr(player_response)
add_dash_mpd(get_video_info)
if view_count is None:
view_count = extract_view_count(get_video_info)
if not video_info:
video_info = get_video_info
- if 'token' in get_video_info:
+ get_token = extract_token(get_video_info)
+ if get_token:
# Different get_video_info requests may report different results, e.g.
# some may report video unavailability, but some may serve it without
- # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
+ # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
# the original webpage as well as el=info and el=embedded get_video_info
# requests report video unavailability due to geo restriction while
# el=detailpage succeeds and returns valid data). This is probably
# due to YouTube measures against IP ranges of hosting providers.
# Working around by preferring the first succeeded video_info containing
# the token if no such video_info yet was found.
- if 'token' not in video_info:
+ token = extract_token(video_info)
+ if not token:
video_info = get_video_info
break
- if 'token' not in video_info:
- if 'reason' in video_info:
- if 'The uploader has not made this video available in your country.' in video_info['reason']:
- regions_allowed = self._html_search_meta(
- 'regionsAllowed', video_webpage, default=None)
- countries = regions_allowed.split(',') if regions_allowed else None
- self.raise_geo_restricted(
- msg=video_info['reason'][0], countries=countries)
- raise ExtractorError(
- 'YouTube said: %s' % video_info['reason'][0],
- expected=True, video_id=video_id)
- else:
- raise ExtractorError(
- '"token" parameter not in video info for unknown reason',
- video_id=video_id)
+
+ def extract_unavailable_message():
+ return self._html_search_regex(
+ r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
+ video_webpage, 'unavailable message', default=None)
+
+ if not video_info:
+ unavailable_message = extract_unavailable_message()
+ if not unavailable_message:
+ unavailable_message = 'Unable to extract video data'
+ raise ExtractorError(
+ 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
+
+ video_details = try_get(
+ player_response, lambda x: x['videoDetails'], dict) or {}
# title
if 'title' in video_info:
video_title = video_info['title'][0]
+ elif 'title' in player_response:
+ video_title = video_details['title']
else:
self._downloader.report_warning('Unable to extract video title')
video_title = '_'
else:
video_description = ''
- if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
+ if not smuggled_data.get('force_singlefeed', False):
if not self._downloader.params.get('noplaylist'):
- entries = []
- feed_ids = []
- multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
- for feed in multifeed_metadata_list.split(','):
- # Unquote should take place before split on comma (,) since textual
- # fields may contain comma as well (see
- # https://github.com/rg3/youtube-dl/issues/8536)
- feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
- entries.append({
- '_type': 'url_transparent',
- 'ie_key': 'Youtube',
- 'url': smuggle_url(
- '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
- {'force_singlefeed': True}),
- 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
- })
- feed_ids.append(feed_data['id'][0])
- self.to_screen(
- 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
- % (', '.join(feed_ids), video_id))
- return self.playlist_result(entries, video_id, video_title, video_description)
- self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
+ multifeed_metadata_list = try_get(
+ player_response,
+ lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
+ compat_str) or try_get(
+ video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
+ if multifeed_metadata_list:
+ entries = []
+ feed_ids = []
+ for feed in multifeed_metadata_list.split(','):
+ # Unquote should take place before split on comma (,) since textual
+ # fields may contain comma as well (see
+ # https://github.com/ytdl-org/youtube-dl/issues/8536)
+ feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
+ entries.append({
+ '_type': 'url_transparent',
+ 'ie_key': 'Youtube',
+ 'url': smuggle_url(
+ '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
+ {'force_singlefeed': True}),
+ 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
+ })
+ feed_ids.append(feed_data['id'][0])
+ self.to_screen(
+ 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
+ % (', '.join(feed_ids), video_id))
+ return self.playlist_result(entries, video_id, video_title, video_description)
+ else:
+ self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
if view_count is None:
view_count = extract_view_count(video_info)
+ if view_count is None and video_details:
+ view_count = int_or_none(video_details.get('viewCount'))
# Check for "rental" videos
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
- raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
-
- # Start extracting information
- self.report_information_extraction(video_id)
-
- # uploader
- if 'author' not in video_info:
- raise ExtractorError('Unable to extract uploader name')
- video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
-
- # uploader_id
- video_uploader_id = None
- video_uploader_url = None
- mobj = re.search(
- r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
- video_webpage)
- if mobj is not None:
- video_uploader_id = mobj.group('uploader_id')
- video_uploader_url = mobj.group('uploader_url')
- else:
- self._downloader.report_warning('unable to extract uploader nickname')
-
- # thumbnail image
- # We try first to get a high quality image:
- m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
- video_webpage, re.DOTALL)
- if m_thumb is not None:
- video_thumbnail = m_thumb.group(1)
- elif 'thumbnail_url' not in video_info:
- self._downloader.report_warning('unable to extract video thumbnail')
- video_thumbnail = None
- else: # don't panic if we can't find it
- video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
-
- # upload date
- upload_date = self._html_search_meta(
- 'datePublished', video_webpage, 'upload date', default=None)
- if not upload_date:
- upload_date = self._search_regex(
- [r'(?s)id="eow-date.*?>(.*?)</span>',
- r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
- video_webpage, 'upload date', default=None)
- upload_date = unified_strdate(upload_date)
-
- video_license = self._html_search_regex(
- r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
- video_webpage, 'license', default=None)
-
- m_music = re.search(
- r'''(?x)
- <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
- <ul[^>]*>\s*
- <li>(?P<title>.+?)
- by (?P<creator>.+?)
- (?:
- \(.+?\)|
- <a[^>]*
- (?:
- \bhref=["\']/red[^>]*>| # drop possible
- >\s*Listen ad-free with YouTube Red # YouTube Red ad
- )
- .*?
- )?</li
- ''',
- video_webpage)
- if m_music:
- video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
- video_creator = clean_html(m_music.group('creator'))
- else:
- video_alt_title = video_creator = None
-
- m_episode = re.search(
- r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
- video_webpage)
- if m_episode:
- series = m_episode.group('series')
- season_number = int(m_episode.group('season'))
- episode_number = int(m_episode.group('episode'))
- else:
- series = season_number = episode_number = None
-
- m_cat_container = self._search_regex(
- r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
- video_webpage, 'categories', default=None)
- if m_cat_container:
- category = self._html_search_regex(
- r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
- default=None)
- video_categories = None if category is None else [category]
- else:
- video_categories = None
-
- video_tags = [
- unescapeHTML(m.group('content'))
- for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
-
- def _extract_count(count_name):
- return str_to_int(self._search_regex(
- r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
- % re.escape(count_name),
- video_webpage, count_name, default=None))
-
- like_count = _extract_count('like')
- dislike_count = _extract_count('dislike')
+ raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
- # subtitles
- video_subtitles = self.extract_subtitles(video_id, video_webpage)
- automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
-
- video_duration = try_get(
- video_info, lambda x: int_or_none(x['length_seconds'][0]))
- if not video_duration:
- video_duration = parse_duration(self._html_search_meta(
- 'duration', video_webpage, 'video duration'))
-
- # annotations
- video_annotations = None
- if self._downloader.params.get('writeannotations', False):
- video_annotations = self._extract_annotations(video_id)
-
- chapters = self._extract_chapters(description_original, video_duration)
+ def _extract_filesize(media_url):
+ return int_or_none(self._search_regex(
+ r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
self.report_rtmp_download()
'url': video_info['conn'][0],
'player_url': player_url,
}]
- elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
+ elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
if 'rtmpe%3Dyes' in encoded_url_map:
- raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
+ raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
formats_spec = {}
fmt_list = video_info.get('fmt_list', [''])[0]
if fmt_list:
'width': int_or_none(width_height[0]),
'height': int_or_none(width_height[1]),
}
+ q = qualities(['small', 'medium', 'hd720'])
+ streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
+ if streaming_formats:
+ for fmt in streaming_formats:
+ itag = str_or_none(fmt.get('itag'))
+ if not itag:
+ continue
+ quality = fmt.get('quality')
+ quality_label = fmt.get('qualityLabel') or quality
+ formats_spec[itag] = {
+ 'asr': int_or_none(fmt.get('audioSampleRate')),
+ 'filesize': int_or_none(fmt.get('contentLength')),
+ 'format_note': quality_label,
+ 'fps': int_or_none(fmt.get('fps')),
+ 'height': int_or_none(fmt.get('height')),
+ 'quality': q(quality),
+ # bitrate for itag 43 is always 2147483647
+ 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
+ 'width': int_or_none(fmt.get('width')),
+ }
formats = []
for url_data_str in encoded_url_map.split(','):
url_data = compat_parse_qs(url_data_str)
- if 'itag' not in url_data or 'url' not in url_data:
+ if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
+ continue
+ stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
+ # Unsupported FORMAT_STREAM_TYPE_OTF
+ if stream_type == 3:
continue
format_id = url_data['itag'][0]
url = url_data['url'][0]
else:
player_version = self._search_regex(
[r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
- r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
+ r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
player_url,
'html5 player', fatal=False)
player_desc = 'html5 player %s' % player_version
signature = self._decrypt_signature(
encrypted_sig, video_id, player_url, age_gate)
- url += '&signature=' + signature
+ sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
+ url += '&%s=%s' % (sp, signature)
if 'ratebypass' not in url:
url += '&ratebypass=yes'
dct.update(formats_spec[format_id])
# Some itags are not included in DASH manifest thus corresponding formats will
- # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
+ # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
# Trying to extract metadata from url_encoded_fmt_stream_map entry.
mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
+ filesize = int_or_none(url_data.get(
+ 'clen', [None])[0]) or _extract_filesize(url)
+
+ quality = url_data.get('quality', [None])[0]
+
more_fields = {
- 'filesize': int_or_none(url_data.get('clen', [None])[0]),
+ 'filesize': filesize,
'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
'width': width,
'height': height,
'fps': int_or_none(url_data.get('fps', [None])[0]),
- 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
+ 'format_note': url_data.get('quality_label', [None])[0] or quality,
+ 'quality': q(quality),
}
for key, value in more_fields.items():
if value:
break
if codecs:
dct.update(parse_codecs(codecs))
+ if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
+ dct['downloader_options'] = {
+ # Youtube throttles chunks >~10M
+ 'http_chunk_size': 10485760,
+ }
formats.append(dct)
- elif video_info.get('hlsvp'):
- manifest_url = video_info['hlsvp'][0]
- formats = []
- m3u8_formats = self._extract_m3u8_formats(
- manifest_url, video_id, 'mp4', fatal=False)
- for a_format in m3u8_formats:
- itag = self._search_regex(
- r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
- if itag:
- a_format['format_id'] = itag
- if itag in self._formats:
- dct = self._formats[itag].copy()
- dct.update(a_format)
- a_format = dct
- a_format['player_url'] = player_url
- # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
- a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
- formats.append(a_format)
else:
- unavailable_message = self._html_search_regex(
- r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
- video_webpage, 'unavailable message', default=None)
- if unavailable_message:
- raise ExtractorError(unavailable_message, expected=True)
- raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
+ manifest_url = (
+ url_or_none(try_get(
+ player_response,
+ lambda x: x['streamingData']['hlsManifestUrl'],
+ compat_str))
+ or url_or_none(try_get(
+ video_info, lambda x: x['hlsvp'][0], compat_str)))
+ if manifest_url:
+ formats = []
+ m3u8_formats = self._extract_m3u8_formats(
+ manifest_url, video_id, 'mp4', fatal=False)
+ for a_format in m3u8_formats:
+ itag = self._search_regex(
+ r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
+ if itag:
+ a_format['format_id'] = itag
+ if itag in self._formats:
+ dct = self._formats[itag].copy()
+ dct.update(a_format)
+ a_format = dct
+ a_format['player_url'] = player_url
+ # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
+ a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
+ formats.append(a_format)
+ else:
+ error_message = clean_html(video_info.get('reason', [None])[0])
+ if not error_message:
+ error_message = extract_unavailable_message()
+ if error_message:
+ raise ExtractorError(error_message, expected=True)
+ raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
+
+ # uploader
+ video_uploader = try_get(
+ video_info, lambda x: x['author'][0],
+ compat_str) or str_or_none(video_details.get('author'))
+ if video_uploader:
+ video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
+ else:
+ self._downloader.report_warning('unable to extract uploader name')
+
+ # uploader_id
+ video_uploader_id = None
+ video_uploader_url = None
+ mobj = re.search(
+ r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
+ video_webpage)
+ if mobj is not None:
+ video_uploader_id = mobj.group('uploader_id')
+ video_uploader_url = mobj.group('uploader_url')
+ else:
+ self._downloader.report_warning('unable to extract uploader nickname')
+
+ channel_id = (
+ str_or_none(video_details.get('channelId'))
+ or self._html_search_meta(
+ 'channelId', video_webpage, 'channel id', default=None)
+ or self._search_regex(
+ r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
+ video_webpage, 'channel id', default=None, group='id'))
+ channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
+
+ # thumbnail image
+ # We try first to get a high quality image:
+ m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
+ video_webpage, re.DOTALL)
+ if m_thumb is not None:
+ video_thumbnail = m_thumb.group(1)
+ elif 'thumbnail_url' not in video_info:
+ self._downloader.report_warning('unable to extract video thumbnail')
+ video_thumbnail = None
+ else: # don't panic if we can't find it
+ video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
+
+ # upload date
+ upload_date = self._html_search_meta(
+ 'datePublished', video_webpage, 'upload date', default=None)
+ if not upload_date:
+ upload_date = self._search_regex(
+ [r'(?s)id="eow-date.*?>(.*?)</span>',
+ r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
+ video_webpage, 'upload date', default=None)
+ upload_date = unified_strdate(upload_date)
+
+ video_license = self._html_search_regex(
+ r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
+ video_webpage, 'license', default=None)
+
+ m_music = re.search(
+ r'''(?x)
+ <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
+ <ul[^>]*>\s*
+ <li>(?P<title>.+?)
+ by (?P<creator>.+?)
+ (?:
+ \(.+?\)|
+ <a[^>]*
+ (?:
+ \bhref=["\']/red[^>]*>| # drop possible
+ >\s*Listen ad-free with YouTube Red # YouTube Red ad
+ )
+ .*?
+ )?</li
+ ''',
+ video_webpage)
+ if m_music:
+ video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
+ video_creator = clean_html(m_music.group('creator'))
+ else:
+ video_alt_title = video_creator = None
+
+ def extract_meta(field):
+ return self._html_search_regex(
+ r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
+ video_webpage, field, default=None)
+
+ track = extract_meta('Song')
+ artist = extract_meta('Artist')
+ album = extract_meta('Album')
+
+ # Youtube Music Auto-generated description
+ release_date = release_year = None
+ if video_description:
+ mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^·]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?', video_description)
+ if mobj:
+ if not track:
+ track = mobj.group('track').strip()
+ if not artist:
+ artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
+ if not album:
+ album = mobj.group('album'.strip())
+ release_year = mobj.group('release_year')
+ release_date = mobj.group('release_date')
+ if release_date:
+ release_date = release_date.replace('-', '')
+ if not release_year:
+ release_year = int(release_date[:4])
+ if release_year:
+ release_year = int(release_year)
+
+ m_episode = re.search(
+ r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
+ video_webpage)
+ if m_episode:
+ series = unescapeHTML(m_episode.group('series'))
+ season_number = int(m_episode.group('season'))
+ episode_number = int(m_episode.group('episode'))
+ else:
+ series = season_number = episode_number = None
+
+ m_cat_container = self._search_regex(
+ r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
+ video_webpage, 'categories', default=None)
+ if m_cat_container:
+ category = self._html_search_regex(
+ r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
+ default=None)
+ video_categories = None if category is None else [category]
+ else:
+ video_categories = None
+
+ video_tags = [
+ unescapeHTML(m.group('content'))
+ for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
+
+ def _extract_count(count_name):
+ return str_to_int(self._search_regex(
+ r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
+ % re.escape(count_name),
+ video_webpage, count_name, default=None))
+
+ like_count = _extract_count('like')
+ dislike_count = _extract_count('dislike')
+
+ if view_count is None:
+ view_count = str_to_int(self._search_regex(
+ r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
+ 'view count', default=None))
+
+ average_rating = (
+ float_or_none(video_details.get('averageRating'))
+ or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
+
+ # subtitles
+ video_subtitles = self.extract_subtitles(video_id, video_webpage)
+ automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
+
+ video_duration = try_get(
+ video_info, lambda x: int_or_none(x['length_seconds'][0]))
+ if not video_duration:
+ video_duration = int_or_none(video_details.get('lengthSeconds'))
+ if not video_duration:
+ video_duration = parse_duration(self._html_search_meta(
+ 'duration', video_webpage, 'video duration'))
+
+ # annotations
+ video_annotations = None
+ if self._downloader.params.get('writeannotations', False):
+ video_annotations = self._extract_annotations(video_id)
+
+ chapters = self._extract_chapters(description_original, video_duration)
# Look for the DASH manifest
if self._downloader.params.get('youtube_include_dash_manifest', True):
for df in self._extract_mpd_formats(
mpd_url, video_id, fatal=dash_mpd_fatal,
formats_dict=self._formats):
+ if not df.get('filesize'):
+ df['filesize'] = _extract_filesize(df['url'])
# Do not overwrite DASH format found in some previous DASH manifest
if df['format_id'] not in dash_formats:
dash_formats[df['format_id']] = df
# Remove the formats we found through non-DASH, they
# contain less info and it can be wrong, because we use
# fixed values (for example the resolution). See
- # https://github.com/rg3/youtube-dl/issues/5774 for an
+ # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
# example.
formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
formats.extend(dash_formats.values())
if f.get('vcodec') != 'none':
f['stretched_ratio'] = ratio
+ if not formats:
+ token = extract_token(video_info)
+ if not token:
+ if 'reason' in video_info:
+ if 'The uploader has not made this video available in your country.' in video_info['reason']:
+ regions_allowed = self._html_search_meta(
+ 'regionsAllowed', video_webpage, default=None)
+ countries = regions_allowed.split(',') if regions_allowed else None
+ self.raise_geo_restricted(
+ msg=video_info['reason'][0], countries=countries)
+ reason = video_info['reason'][0]
+ if 'Invalid parameters' in reason:
+ unavailable_message = extract_unavailable_message()
+ if unavailable_message:
+ reason = unavailable_message
+ raise ExtractorError(
+ 'YouTube said: %s' % reason,
+ expected=True, video_id=video_id)
+ else:
+ raise ExtractorError(
+ '"token" parameter not in video info for unknown reason',
+ video_id=video_id)
+
+ if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
+ raise ExtractorError('This video is DRM protected.', expected=True)
+
self._sort_formats(formats)
- self.mark_watched(video_id, video_info)
+ self.mark_watched(video_id, video_info, player_response)
return {
'id': video_id,
'uploader': video_uploader,
'uploader_id': video_uploader_id,
'uploader_url': video_uploader_url,
+ 'channel_id': channel_id,
+ 'channel_url': channel_url,
'upload_date': upload_date,
'license': video_license,
- 'creator': video_creator,
+ 'creator': video_creator or artist,
'title': video_title,
- 'alt_title': video_alt_title,
+ 'alt_title': video_alt_title or track,
'thumbnail': video_thumbnail,
'description': video_description,
'categories': video_categories,
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
- 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
+ 'average_rating': average_rating,
'formats': formats,
'is_live': is_live,
'start_time': start_time,
'series': series,
'season_number': season_number,
'episode_number': episode_number,
+ 'track': track,
+ 'artist': artist,
+ 'album': album,
+ 'release_date': release_date,
+ 'release_year': release_year,
}
(?:https?://)?
(?:\w+\.)?
(?:
- youtube\.com/
+ (?:
+ youtube\.com|
+ invidio\.us
+ )
+ /
(?:
(?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
\? (?:.*?[&;])*? (?:p|a|list)=
youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
)
(
- (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
+ (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
# Top tracks, they can also include dots
|(?:MC)[\w\.]*
)
'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
'categories': ['People & Blogs'],
'tags': list,
+ 'view_count': int,
'like_count': int,
'dislike_count': int,
},
}, {
'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
'only_matching': True,
+ }, {
+ # music album playlist
+ 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
+ 'only_matching': True,
}]
def _real_initialize(self):
search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
title_span = (
- search_title('playlist-title') or
- search_title('title long-title') or
- search_title('title'))
+ search_title('playlist-title')
+ or search_title('title long-title')
+ or search_title('title'))
title = clean_html(title_span)
return self.playlist_result(url_results, playlist_id, title)
url = self._TEMPLATE_URL % playlist_id
page = self._download_webpage(url, playlist_id)
- # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
+ # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
match = match.strip()
# Check if the playlist exists or is private
r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
page, 'title', default=None)
+ _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
+ uploader = self._search_regex(
+ r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
+ page, 'uploader', default=None)
+ mobj = re.search(
+ r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
+ page)
+ if mobj:
+ uploader_id = mobj.group('uploader_id')
+ uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
+ else:
+ uploader_id = uploader_url = None
+
has_videos = True
if not playlist_title:
except StopIteration:
has_videos = False
- return has_videos, self.playlist_result(
+ playlist = self.playlist_result(
self._entries(page, playlist_id), playlist_id, playlist_title)
+ playlist.update({
+ 'uploader': uploader,
+ 'uploader_id': uploader_id,
+ 'uploader_url': uploader_url,
+ })
+
+ return has_videos, playlist
def _check_download_just_video(self, url, playlist_id):
# Check if it's a video-specific URL
return playlist
# Some playlist URLs don't actually serve a playlist (see
- # https://github.com/rg3/youtube-dl/issues/10537).
+ # https://github.com/ytdl-org/youtube-dl/issues/10537).
# Fallback to plain video extraction if there is a video id
# along with playlist id.
return self.url_result(video_id, 'Youtube', video_id=video_id)
class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
IE_DESC = 'YouTube.com channels'
- _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
+ _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
_TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
_VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
IE_NAME = 'youtube:channel'
'id': 'UUs0ifCMCm1icqRbqhUINa0w',
'title': 'Uploads from Deus Ex',
},
+ }, {
+ 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
+ 'only_matching': True,
}]
@classmethod
class YoutubeUserIE(YoutubeChannelIE):
IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
- _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
+ _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
_TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
IE_NAME = 'youtube:user'
webpage = self._download_webpage(url, channel_id, fatal=False)
if webpage:
page_type = self._og_search_property(
- 'type', webpage, 'page type', default=None)
+ 'type', webpage, 'page type', default='')
video_id = self._html_search_meta(
'videoId', webpage, 'video id', default=None)
- if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
+ if page_type.startswith('video') and video_id and re.match(
+ r'^[0-9A-Za-z_-]{11}$', video_id):
return self.url_result(video_id, YoutubeIE.ie_key())
return self.url_result(base_url)
}]
-class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
+class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
+ _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
+
+
+class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
IE_DESC = 'YouTube.com searches'
# there doesn't appear to be a real limit, for example if you search for
# 'python' you get more than 8.000.000 results
raise ExtractorError(
'[youtube] No video results', expected=True)
- new_videos = self._ids_to_results(orderedSet(re.findall(
- r'href="/watch\?v=(.{11})', html_content)))
+ new_videos = list(self._process_page(html_content))
videos += new_videos
if not new_videos or len(videos) > limit:
break
_EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
-class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
+class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
IE_DESC = 'YouTube.com search URLs'
IE_NAME = 'youtube:search_url'
_VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
- _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
_TESTS = [{
'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
'playlist_mincount': 5,
def _real_initialize(self):
self._login()
- def _real_extract(self, url):
- page = self._download_webpage(
- 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
-
+ def _entries(self, page):
# The extraction process is the same as for playlists, but the regex
# for the video ids doesn't contain an index
ids = []
# 'recommended' feed has infinite 'load more' and each new portion spins
# the same videos in (sometimes) slightly different order, so we'll check
# for unicity and break when portion has no new videos
- new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
+ new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
if not new_ids:
break
ids.extend(new_ids)
+ for entry in self._ids_to_results(new_ids):
+ yield entry
+
mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
if not mobj:
break
content_html = more['content_html']
more_widget_html = more['load_more_widget_html']
+ def _real_extract(self, url):
+ page = self._download_webpage(
+ 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
+ self._PLAYLIST_TITLE)
return self.playlist_result(
- self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
+ self._entries(page), playlist_title=self._PLAYLIST_TITLE)
class YoutubeWatchLaterIE(YoutubePlaylistIE):
Rapsys Git / youtubedl / blobdiff