_ Git - youtube-dl/blob - youtube_dl/extractor/youtube.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5
   6 import itertools
   7 import json
   8 import os.path
   9 import random
  10 import re
  11 import time
  12 import traceback
  13
  14 from .common import InfoExtractor, SearchInfoExtractor
  15 from ..jsinterp import JSInterpreter
  16 from ..swfinterp import SWFInterpreter
  17 from ..compat import (
  18     compat_chr,
  19     compat_HTTPError,
  20     compat_kwargs,
  21     compat_parse_qs,
  22     compat_urllib_parse_unquote,
  23     compat_urllib_parse_unquote_plus,
  24     compat_urllib_parse_urlencode,
  25     compat_urllib_parse_urlparse,
  26     compat_urlparse,
  27     compat_str,
  28 )
  29 from ..utils import (
  30     clean_html,
  31     dict_get,
  32     error_to_compat_str,
  33     ExtractorError,
  34     float_or_none,
  35     get_element_by_attribute,
  36     get_element_by_id,
  37     int_or_none,
  38     mimetype2ext,
  39     orderedSet,
  40     parse_codecs,
  41     parse_duration,
  42     qualities,
  43     remove_quotes,
  44     remove_start,
  45     smuggle_url,
  46     str_or_none,
  47     str_to_int,
  48     try_get,
  49     unescapeHTML,
  50     unified_strdate,
  51     unsmuggle_url,
  52     uppercase_escape,
  53     url_or_none,
  54     urlencode_postdata,
  55 )
  56
  57
  58 class YoutubeBaseInfoExtractor(InfoExtractor):
  59     """Provide base functions for Youtube extractors"""
  60     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
  61     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
  62
  63     _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
  64     _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
  65     _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
  66
  67     _NETRC_MACHINE = 'youtube'
  68     # If True it will raise an error if no login info is provided
  69     _LOGIN_REQUIRED = False
  70
  71     _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
  72
  73     def _set_language(self):
  74         self._set_cookie(
  75             '.youtube.com', 'PREF', 'f1=50000000&hl=en',
  76             # YouTube sets the expire time to about two months
  77             expire_time=time.time() + 2 * 30 * 24 * 3600)
  78
  79     def _ids_to_results(self, ids):
  80         return [
  81             self.url_result(vid_id, 'Youtube', video_id=vid_id)
  82             for vid_id in ids]
  83
  84     def _login(self):
  85         """
  86         Attempt to log in to YouTube.
  87         True is returned if successful or skipped.
  88         False is returned if login failed.
  89
  90         If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
  91         """
  92         username, password = self._get_login_info()
  93         # No authentication to be performed
  94         if username is None:
  95             if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
  96                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  97             return True
  98
  99         login_page = self._download_webpage(
 100             self._LOGIN_URL, None,
 101             note='Downloading login page',
 102             errnote='unable to fetch login page', fatal=False)
 103         if login_page is False:
 104             return
 105
 106         login_form = self._hidden_inputs(login_page)
 107
 108         def req(url, f_req, note, errnote):
 109             data = login_form.copy()
 110             data.update({
 111                 'pstMsg': 1,
 112                 'checkConnection': 'youtube',
 113                 'checkedDomains': 'youtube',
 114                 'hl': 'en',
 115                 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
 116                 'f.req': json.dumps(f_req),
 117                 'flowName': 'GlifWebSignIn',
 118                 'flowEntry': 'ServiceLogin',
 119             })
 120             return self._download_json(
 121                 url, None, note=note, errnote=errnote,
 122                 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
 123                 fatal=False,
 124                 data=urlencode_postdata(data), headers={
 125                     'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
 126                     'Google-Accounts-XSRF': 1,
 127                 })
 128
 129         def warn(message):
 130             self._downloader.report_warning(message)
 131
 132         lookup_req = [
 133             username,
 134             None, [], None, 'US', None, None, 2, False, True,
 135             [
 136                 None, None,
 137                 [2, 1, None, 1,
 138                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
 139                  None, [], 4],
 140                 1, [None, None, []], None, None, None, True
 141             ],
 142             username,
 143         ]
 144
 145         lookup_results = req(
 146             self._LOOKUP_URL, lookup_req,
 147             'Looking up account info', 'Unable to look up account info')
 148
 149         if lookup_results is False:
 150             return False
 151
 152         user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
 153         if not user_hash:
 154             warn('Unable to extract user hash')
 155             return False
 156
 157         challenge_req = [
 158             user_hash,
 159             None, 1, None, [1, None, None, None, [password, None, True]],
 160             [
 161                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
 162                 1, [None, None, []], None, None, None, True
 163             ]]
 164
 165         challenge_results = req(
 166             self._CHALLENGE_URL, challenge_req,
 167             'Logging in', 'Unable to log in')
 168
 169         if challenge_results is False:
 170             return
 171
 172         login_res = try_get(challenge_results, lambda x: x[0][5], list)
 173         if login_res:
 174             login_msg = try_get(login_res, lambda x: x[5], compat_str)
 175             warn(
 176                 'Unable to login: %s' % 'Invalid password'
 177                 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
 178             return False
 179
 180         res = try_get(challenge_results, lambda x: x[0][-1], list)
 181         if not res:
 182             warn('Unable to extract result entry')
 183             return False
 184
 185         login_challenge = try_get(res, lambda x: x[0][0], list)
 186         if login_challenge:
 187             challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
 188             if challenge_str == 'TWO_STEP_VERIFICATION':
 189                 # SEND_SUCCESS - TFA code has been successfully sent to phone
 190                 # QUOTA_EXCEEDED - reached the limit of TFA codes
 191                 status = try_get(login_challenge, lambda x: x[5], compat_str)
 192                 if status == 'QUOTA_EXCEEDED':
 193                     warn('Exceeded the limit of TFA codes, try later')
 194                     return False
 195
 196                 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
 197                 if not tl:
 198                     warn('Unable to extract TL')
 199                     return False
 200
 201                 tfa_code = self._get_tfa_info('2-step verification code')
 202
 203                 if not tfa_code:
 204                     warn(
 205                         'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
 206                         '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
 207                     return False
 208
 209                 tfa_code = remove_start(tfa_code, 'G-')
 210
 211                 tfa_req = [
 212                     user_hash, None, 2, None,
 213                     [
 214                         9, None, None, None, None, None, None, None,
 215                         [None, tfa_code, True, 2]
 216                     ]]
 217
 218                 tfa_results = req(
 219                     self._TFA_URL.format(tl), tfa_req,
 220                     'Submitting TFA code', 'Unable to submit TFA code')
 221
 222                 if tfa_results is False:
 223                     return False
 224
 225                 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
 226                 if tfa_res:
 227                     tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
 228                     warn(
 229                         'Unable to finish TFA: %s' % 'Invalid TFA code'
 230                         if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
 231                     return False
 232
 233                 check_cookie_url = try_get(
 234                     tfa_results, lambda x: x[0][-1][2], compat_str)
 235             else:
 236                 CHALLENGES = {
 237                     'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
 238                     'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
 239                     'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
 240                 }
 241                 challenge = CHALLENGES.get(
 242                     challenge_str,
 243                     '%s returned error %s.' % (self.IE_NAME, challenge_str))
 244                 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
 245                 return False
 246         else:
 247             check_cookie_url = try_get(res, lambda x: x[2], compat_str)
 248
 249         if not check_cookie_url:
 250             warn('Unable to extract CheckCookie URL')
 251             return False
 252
 253         check_cookie_results = self._download_webpage(
 254             check_cookie_url, None, 'Checking cookie', fatal=False)
 255
 256         if check_cookie_results is False:
 257             return False
 258
 259         if 'https://myaccount.google.com/' not in check_cookie_results:
 260             warn('Unable to log in')
 261             return False
 262
 263         return True
 264
 265     def _download_webpage_handle(self, *args, **kwargs):
 266         query = kwargs.get('query', {}).copy()
 267         query['disable_polymer'] = 'true'
 268         kwargs['query'] = query
 269         return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
 270             *args, **compat_kwargs(kwargs))
 271
 272     def _real_initialize(self):
 273         if self._downloader is None:
 274             return
 275         self._set_language()
 276         if not self._login():
 277             return
 278
 279
 280 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
 281     # Extract entries from page with "Load more" button
 282     def _entries(self, page, playlist_id):
 283         more_widget_html = content_html = page
 284         for page_num in itertools.count(1):
 285             for entry in self._process_page(content_html):
 286                 yield entry
 287
 288             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
 289             if not mobj:
 290                 break
 291
 292             count = 0
 293             retries = 3
 294             while count <= retries:
 295                 try:
 296                     # Downloading page may result in intermittent 5xx HTTP error
 297                     # that is usually worked around with a retry
 298                     more = self._download_json(
 299                         'https://youtube.com/%s' % mobj.group('more'), playlist_id,
 300                         'Downloading page #%s%s'
 301                         % (page_num, ' (retry #%d)' % count if count else ''),
 302                         transform_source=uppercase_escape)
 303                     break
 304                 except ExtractorError as e:
 305                     if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
 306                         count += 1
 307                         if count <= retries:
 308                             continue
 309                     raise
 310
 311             content_html = more['content_html']
 312             if not content_html.strip():
 313                 # Some webpages show a "Load more" button but they don't
 314                 # have more videos
 315                 break
 316             more_widget_html = more['load_more_widget_html']
 317
 318
 319 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
 320     def _process_page(self, content):
 321         for video_id, video_title in self.extract_videos_from_page(content):
 322             yield self.url_result(video_id, 'Youtube', video_id, video_title)
 323
 324     def extract_videos_from_page(self, page):
 325         ids_in_page = []
 326         titles_in_page = []
 327         for mobj in re.finditer(self._VIDEO_RE, page):
 328             # The link with index 0 is not the first video of the playlist (not sure if still actual)
 329             if 'index' in mobj.groupdict() and mobj.group('id') == '0':
 330                 continue
 331             video_id = mobj.group('id')
 332             video_title = unescapeHTML(mobj.group('title'))
 333             if video_title:
 334                 video_title = video_title.strip()
 335             try:
 336                 idx = ids_in_page.index(video_id)
 337                 if video_title and not titles_in_page[idx]:
 338                     titles_in_page[idx] = video_title
 339             except ValueError:
 340                 ids_in_page.append(video_id)
 341                 titles_in_page.append(video_title)
 342         return zip(ids_in_page, titles_in_page)
 343
 344
 345 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
 346     def _process_page(self, content):
 347         for playlist_id in orderedSet(re.findall(
 348                 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
 349                 content)):
 350             yield self.url_result(
 351                 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
 352
 353     def _real_extract(self, url):
 354         playlist_id = self._match_id(url)
 355         webpage = self._download_webpage(url, playlist_id)
 356         title = self._og_search_title(webpage, fatal=False)
 357         return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
 358
 359
 360 class YoutubeIE(YoutubeBaseInfoExtractor):
 361     IE_DESC = 'YouTube.com'
 362     _VALID_URL = r"""(?x)^
 363                      (
 364                          (?:https?://|//)                                    # http(s):// or protocol-independent URL
 365                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
 366                             (?:www\.)?deturl\.com/www\.youtube\.com/|
 367                             (?:www\.)?pwnyoutube\.com/|
 368                             (?:www\.)?hooktube\.com/|
 369                             (?:www\.)?yourepeat\.com/|
 370                             tube\.majestyc\.net/|
 371                             (?:(?:www|dev)\.)?invidio\.us/|
 372                             (?:www\.)?invidiou\.sh/|
 373                             (?:www\.)?invidious\.snopyta\.org/|
 374                             (?:www\.)?invidious\.kabi\.tk/|
 375                             (?:www\.)?vid\.wxzm\.sx/|
 376                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
 377                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
 378                          (?:                                                  # the various things that can precede the ID:
 379                              (?:(?:v|embed|e)/(?!videoseries))                # v/ or embed/ or e/
 380                              |(?:                                             # or the v= param in all its forms
 381                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)?  # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
 382                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
 383                                  (?:.*?[&;])??                                # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
 384                                  v=
 385                              )
 386                          ))
 387                          |(?:
 388                             youtu\.be|                                        # just youtu.be/xxxx
 389                             vid\.plus|                                        # or vid.plus/xxxx
 390                             zwearz\.com/watch|                                # or zwearz.com/watch/xxxx
 391                          )/
 392                          |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
 393                          )
 394                      )?                                                       # all until now is optional -> you can pass the naked ID
 395                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
 396                      (?!.*?\blist=
 397                         (?:
 398                             %(playlist_id)s|                                  # combined list/video URLs are handled by the playlist IE
 399                             WL                                                # WL are handled by the watch later IE
 400                         )
 401                      )
 402                      (?(1).+)?                                                # if we found the ID, everything can follow
 403                      $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
 404     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
 405     _formats = {
 406         '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
 407         '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
 408         '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
 409         '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
 410         '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
 411         '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 412         '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 413         '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 414         # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
 415         '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
 416         '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 417         '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 418         '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
 419         '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
 420         '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
 421         '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
 422         '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 423         '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 424
 425
 426         # 3D videos
 427         '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
 428         '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
 429         '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
 430         '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
 431         '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
 432         '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
 433         '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
 434
 435         # Apple HTTP Live Streaming
 436         '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 437         '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 438         '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
 439         '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
 440         '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
 441         '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
 442         '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 443         '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
 444
 445         # DASH mp4 video
 446         '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
 447         '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
 448         '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
 449         '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
 450         '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
 451         '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'},  # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
 452         '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
 453         '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
 454         '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
 455         '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
 456         '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
 457         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
 458
 459         # Dash mp4 audio
 460         '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
 461         '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
 462         '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
 463         '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
 464         '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
 465         '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
 466         '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
 467
 468         # Dash webm
 469         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 470         '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 471         '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 472         '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 473         '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 474         '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 475         '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
 476         '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 477         '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 478         '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 479         '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 480         '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 481         '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 482         '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 483         '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 484         # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
 485         '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 486         '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 487         '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 488         '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 489         '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 490         '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 491
 492         # Dash webm audio
 493         '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
 494         '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
 495
 496         # Dash webm audio with opus inside
 497         '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
 498         '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
 499         '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
 500
 501         # RTMP (unnamed)
 502         '_rtmp': {'protocol': 'rtmp'},
 503
 504         # av01 video only formats sometimes served with "unknown" codecs
 505         '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 506         '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 507         '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 508         '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 509     }
 510     _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
 511
 512     _GEO_BYPASS = False
 513
 514     IE_NAME = 'youtube'
 515     _TESTS = [
 516         {
 517             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
 518             'info_dict': {
 519                 'id': 'BaW_jenozKc',
 520                 'ext': 'mp4',
 521                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
 522                 'uploader': 'Philipp Hagemeister',
 523                 'uploader_id': 'phihag',
 524                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
 525                 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
 526                 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
 527                 'upload_date': '20121002',
 528                 '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 .',
 529                 'categories': ['Science & Technology'],
 530                 'tags': ['youtube-dl'],
 531                 'duration': 10,
 532                 'view_count': int,
 533                 'like_count': int,
 534                 'dislike_count': int,
 535                 'start_time': 1,
 536                 'end_time': 9,
 537             }
 538         },
 539         {
 540             'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
 541             'note': 'Test generic use_cipher_signature video (#897)',
 542             'info_dict': {
 543                 'id': 'UxxajLWwzqY',
 544                 'ext': 'mp4',
 545                 'upload_date': '20120506',
 546                 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
 547                 'alt_title': 'I Love It (feat. Charli XCX)',
 548                 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
 549                 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
 550                          'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
 551                          'iconic ep', 'iconic', 'love', 'it'],
 552                 'duration': 180,
 553                 'uploader': 'Icona Pop',
 554                 'uploader_id': 'IconaPop',
 555                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
 556                 'creator': 'Icona Pop',
 557                 'track': 'I Love It (feat. Charli XCX)',
 558                 'artist': 'Icona Pop',
 559             }
 560         },
 561         {
 562             'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
 563             'note': 'Test VEVO video with age protection (#956)',
 564             'info_dict': {
 565                 'id': '07FYdnEawAQ',
 566                 'ext': 'mp4',
 567                 'upload_date': '20130703',
 568                 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
 569                 'alt_title': 'Tunnel Vision',
 570                 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
 571                 'duration': 419,
 572                 'uploader': 'justintimberlakeVEVO',
 573                 'uploader_id': 'justintimberlakeVEVO',
 574                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
 575                 'creator': 'Justin Timberlake',
 576                 'track': 'Tunnel Vision',
 577                 'artist': 'Justin Timberlake',
 578                 'age_limit': 18,
 579             }
 580         },
 581         {
 582             'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
 583             'note': 'Embed-only video (#1746)',
 584             'info_dict': {
 585                 'id': 'yZIXLfi8CZQ',
 586                 'ext': 'mp4',
 587                 'upload_date': '20120608',
 588                 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
 589                 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
 590                 'uploader': 'SET India',
 591                 'uploader_id': 'setindia',
 592                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
 593                 'age_limit': 18,
 594             }
 595         },
 596         {
 597             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
 598             'note': 'Use the first video ID in the URL',
 599             'info_dict': {
 600                 'id': 'BaW_jenozKc',
 601                 'ext': 'mp4',
 602                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
 603                 'uploader': 'Philipp Hagemeister',
 604                 'uploader_id': 'phihag',
 605                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
 606                 'upload_date': '20121002',
 607                 '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 .',
 608                 'categories': ['Science & Technology'],
 609                 'tags': ['youtube-dl'],
 610                 'duration': 10,
 611                 'view_count': int,
 612                 'like_count': int,
 613                 'dislike_count': int,
 614             },
 615             'params': {
 616                 'skip_download': True,
 617             },
 618         },
 619         {
 620             'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
 621             'note': '256k DASH audio (format 141) via DASH manifest',
 622             'info_dict': {
 623                 'id': 'a9LDPn-MO4I',
 624                 'ext': 'm4a',
 625                 'upload_date': '20121002',
 626                 'uploader_id': '8KVIDEO',
 627                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
 628                 'description': '',
 629                 'uploader': '8KVIDEO',
 630                 'title': 'UHDTV TEST 8K VIDEO.mp4'
 631             },
 632             'params': {
 633                 'youtube_include_dash_manifest': True,
 634                 'format': '141',
 635             },
 636             'skip': 'format 141 not served anymore',
 637         },
 638         # DASH manifest with encrypted signature
 639         {
 640             'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
 641             'info_dict': {
 642                 'id': 'IB3lcPjvWLA',
 643                 'ext': 'm4a',
 644                 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
 645                 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
 646                 'duration': 244,
 647                 'uploader': 'AfrojackVEVO',
 648                 'uploader_id': 'AfrojackVEVO',
 649                 'upload_date': '20131011',
 650             },
 651             'params': {
 652                 'youtube_include_dash_manifest': True,
 653                 'format': '141/bestaudio[ext=m4a]',
 654             },
 655         },
 656         # JS player signature function name containing $
 657         {
 658             'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
 659             'info_dict': {
 660                 'id': 'nfWlot6h_JM',
 661                 'ext': 'm4a',
 662                 'title': 'Taylor Swift - Shake It Off',
 663                 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
 664                 'duration': 242,
 665                 'uploader': 'TaylorSwiftVEVO',
 666                 'uploader_id': 'TaylorSwiftVEVO',
 667                 'upload_date': '20140818',
 668                 'creator': 'Taylor Swift',
 669             },
 670             'params': {
 671                 'youtube_include_dash_manifest': True,
 672                 'format': '141/bestaudio[ext=m4a]',
 673             },
 674         },
 675         # Controversy video
 676         {
 677             'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
 678             'info_dict': {
 679                 'id': 'T4XJQO3qol8',
 680                 'ext': 'mp4',
 681                 'duration': 219,
 682                 'upload_date': '20100909',
 683                 'uploader': 'Amazing Atheist',
 684                 'uploader_id': 'TheAmazingAtheist',
 685                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
 686                 'title': 'Burning Everyone\'s Koran',
 687                 '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',
 688             }
 689         },
 690         # Normal age-gate video (No vevo, embed allowed)
 691         {
 692             'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
 693             'info_dict': {
 694                 'id': 'HtVdAasjOgU',
 695                 'ext': 'mp4',
 696                 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
 697                 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
 698                 'duration': 142,
 699                 'uploader': 'The Witcher',
 700                 'uploader_id': 'WitcherGame',
 701                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
 702                 'upload_date': '20140605',
 703                 'age_limit': 18,
 704             },
 705         },
 706         # Age-gate video with encrypted signature
 707         {
 708             'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
 709             'info_dict': {
 710                 'id': '6kLq3WMV1nU',
 711                 'ext': 'mp4',
 712                 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
 713                 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
 714                 'duration': 246,
 715                 'uploader': 'LloydVEVO',
 716                 'uploader_id': 'LloydVEVO',
 717                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
 718                 'upload_date': '20110629',
 719                 'age_limit': 18,
 720             },
 721         },
 722         # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
 723         # YouTube Red ad is not captured for creator
 724         {
 725             'url': '__2ABJjxzNo',
 726             'info_dict': {
 727                 'id': '__2ABJjxzNo',
 728                 'ext': 'mp4',
 729                 'duration': 266,
 730                 'upload_date': '20100430',
 731                 'uploader_id': 'deadmau5',
 732                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
 733                 'creator': 'deadmau5',
 734                 'description': 'md5:12c56784b8032162bb936a5f76d55360',
 735                 'uploader': 'deadmau5',
 736                 'title': 'Deadmau5 - Some Chords (HD)',
 737                 'alt_title': 'Some Chords',
 738             },
 739             'expected_warnings': [
 740                 'DASH manifest missing',
 741             ]
 742         },
 743         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
 744         {
 745             'url': 'lqQg6PlCWgI',
 746             'info_dict': {
 747                 'id': 'lqQg6PlCWgI',
 748                 'ext': 'mp4',
 749                 'duration': 6085,
 750                 'upload_date': '20150827',
 751                 'uploader_id': 'olympic',
 752                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
 753                 'description': 'HO09  - Women -  GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
 754                 'uploader': 'Olympic',
 755                 'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',
 756             },
 757             'params': {
 758                 'skip_download': 'requires avconv',
 759             }
 760         },
 761         # Non-square pixels
 762         {
 763             'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
 764             'info_dict': {
 765                 'id': '_b-2C3KPAM0',
 766                 'ext': 'mp4',
 767                 'stretched_ratio': 16 / 9.,
 768                 'duration': 85,
 769                 'upload_date': '20110310',
 770                 'uploader_id': 'AllenMeow',
 771                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
 772                 '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',
 773                 'uploader': '孫ᄋᄅ',
 774                 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
 775             },
 776         },
 777         # url_encoded_fmt_stream_map is empty string
 778         {
 779             'url': 'qEJwOuvDf7I',
 780             'info_dict': {
 781                 'id': 'qEJwOuvDf7I',
 782                 'ext': 'webm',
 783                 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
 784                 'description': '',
 785                 'upload_date': '20150404',
 786                 'uploader_id': 'spbelect',
 787                 'uploader': 'Наблюдатели Петербурга',
 788             },
 789             'params': {
 790                 'skip_download': 'requires avconv',
 791             },
 792             'skip': 'This live event has ended.',
 793         },
 794         # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
 795         {
 796             'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
 797             'info_dict': {
 798                 'id': 'FIl7x6_3R5Y',
 799                 'ext': 'webm',
 800                 'title': 'md5:7b81415841e02ecd4313668cde88737a',
 801                 'description': 'md5:116377fd2963b81ec4ce64b542173306',
 802                 'duration': 220,
 803                 'upload_date': '20150625',
 804                 'uploader_id': 'dorappi2000',
 805                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
 806                 'uploader': 'dorappi2000',
 807                 'formats': 'mincount:31',
 808             },
 809             'skip': 'not actual anymore',
 810         },
 811         # DASH manifest with segment_list
 812         {
 813             'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
 814             'md5': '8ce563a1d667b599d21064e982ab9e31',
 815             'info_dict': {
 816                 'id': 'CsmdDsKjzN8',
 817                 'ext': 'mp4',
 818                 'upload_date': '20150501',  # According to '<meta itemprop="datePublished"', but in other places it's 20150510
 819                 'uploader': 'Airtek',
 820                 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
 821                 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
 822                 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
 823             },
 824             'params': {
 825                 'youtube_include_dash_manifest': True,
 826                 'format': '135',  # bestvideo
 827             },
 828             'skip': 'This live event has ended.',
 829         },
 830         {
 831             # Multifeed videos (multiple cameras), URL is for Main Camera
 832             'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
 833             'info_dict': {
 834                 'id': 'jqWvoWXjCVs',
 835                 'title': 'teamPGP: Rocket League Noob Stream',
 836                 'description': 'md5:dc7872fb300e143831327f1bae3af010',
 837             },
 838             'playlist': [{
 839                 'info_dict': {
 840                     'id': 'jqWvoWXjCVs',
 841                     'ext': 'mp4',
 842                     'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
 843                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 844                     'duration': 7335,
 845                     'upload_date': '20150721',
 846                     'uploader': 'Beer Games Beer',
 847                     'uploader_id': 'beergamesbeer',
 848                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 849                     'license': 'Standard YouTube License',
 850                 },
 851             }, {
 852                 'info_dict': {
 853                     'id': '6h8e8xoXJzg',
 854                     'ext': 'mp4',
 855                     'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
 856                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 857                     'duration': 7337,
 858                     'upload_date': '20150721',
 859                     'uploader': 'Beer Games Beer',
 860                     'uploader_id': 'beergamesbeer',
 861                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 862                     'license': 'Standard YouTube License',
 863                 },
 864             }, {
 865                 'info_dict': {
 866                     'id': 'PUOgX5z9xZw',
 867                     'ext': 'mp4',
 868                     'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
 869                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 870                     'duration': 7337,
 871                     'upload_date': '20150721',
 872                     'uploader': 'Beer Games Beer',
 873                     'uploader_id': 'beergamesbeer',
 874                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 875                     'license': 'Standard YouTube License',
 876                 },
 877             }, {
 878                 'info_dict': {
 879                     'id': 'teuwxikvS5k',
 880                     'ext': 'mp4',
 881                     'title': 'teamPGP: Rocket League Noob Stream (zim)',
 882                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 883                     'duration': 7334,
 884                     'upload_date': '20150721',
 885                     'uploader': 'Beer Games Beer',
 886                     'uploader_id': 'beergamesbeer',
 887                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 888                     'license': 'Standard YouTube License',
 889                 },
 890             }],
 891             'params': {
 892                 'skip_download': True,
 893             },
 894             'skip': 'This video is not available.',
 895         },
 896         {
 897             # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
 898             'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
 899             'info_dict': {
 900                 'id': 'gVfLd0zydlo',
 901                 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
 902             },
 903             'playlist_count': 2,
 904             'skip': 'Not multifeed anymore',
 905         },
 906         {
 907             'url': 'https://vid.plus/FlRa-iH7PGw',
 908             'only_matching': True,
 909         },
 910         {
 911             'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
 912             'only_matching': True,
 913         },
 914         {
 915             # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
 916             # Also tests cut-off URL expansion in video description (see
 917             # https://github.com/ytdl-org/youtube-dl/issues/1892,
 918             # https://github.com/ytdl-org/youtube-dl/issues/8164)
 919             'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
 920             'info_dict': {
 921                 'id': 'lsguqyKfVQg',
 922                 'ext': 'mp4',
 923                 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
 924                 'alt_title': 'Dark Walk - Position Music',
 925                 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
 926                 'duration': 133,
 927                 'upload_date': '20151119',
 928                 'uploader_id': 'IronSoulElf',
 929                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
 930                 'uploader': 'IronSoulElf',
 931                 'creator': 'Todd Haberman,  Daniel Law Heath and Aaron Kaplan',
 932                 'track': 'Dark Walk - Position Music',
 933                 'artist': 'Todd Haberman,  Daniel Law Heath and Aaron Kaplan',
 934                 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
 935             },
 936             'params': {
 937                 'skip_download': True,
 938             },
 939         },
 940         {
 941             # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
 942             'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
 943             'only_matching': True,
 944         },
 945         {
 946             # Video with yt:stretch=17:0
 947             'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
 948             'info_dict': {
 949                 'id': 'Q39EVAstoRM',
 950                 'ext': 'mp4',
 951                 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
 952                 'description': 'md5:ee18a25c350637c8faff806845bddee9',
 953                 'upload_date': '20151107',
 954                 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
 955                 'uploader': 'CH GAMER DROID',
 956             },
 957             'params': {
 958                 'skip_download': True,
 959             },
 960             'skip': 'This video does not exist.',
 961         },
 962         {
 963             # Video licensed under Creative Commons
 964             'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
 965             'info_dict': {
 966                 'id': 'M4gD1WSo5mA',
 967                 'ext': 'mp4',
 968                 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
 969                 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
 970                 'duration': 721,
 971                 'upload_date': '20150127',
 972                 'uploader_id': 'BerkmanCenter',
 973                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
 974                 'uploader': 'The Berkman Klein Center for Internet & Society',
 975                 'license': 'Creative Commons Attribution license (reuse allowed)',
 976             },
 977             'params': {
 978                 'skip_download': True,
 979             },
 980         },
 981         {
 982             # Channel-like uploader_url
 983             'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
 984             'info_dict': {
 985                 'id': 'eQcmzGIKrzg',
 986                 'ext': 'mp4',
 987                 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
 988                 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
 989                 'duration': 4060,
 990                 'upload_date': '20151119',
 991                 'uploader': 'Bernie Sanders',
 992                 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
 993                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
 994                 'license': 'Creative Commons Attribution license (reuse allowed)',
 995             },
 996             'params': {
 997                 'skip_download': True,
 998             },
 999         },
1000         {
1001             'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1002             'only_matching': True,
1003         },
1004         {
1005             # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1006             'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1007             'only_matching': True,
1008         },
1009         {
1010             # Rental video preview
1011             'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1012             'info_dict': {
1013                 'id': 'uGpuVWrhIzE',
1014                 'ext': 'mp4',
1015                 'title': 'Piku - Trailer',
1016                 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1017                 'upload_date': '20150811',
1018                 'uploader': 'FlixMatrix',
1019                 'uploader_id': 'FlixMatrixKaravan',
1020                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1021                 'license': 'Standard YouTube License',
1022             },
1023             'params': {
1024                 'skip_download': True,
1025             },
1026             'skip': 'This video is not available.',
1027         },
1028         {
1029             # YouTube Red video with episode data
1030             'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1031             'info_dict': {
1032                 'id': 'iqKdEhx-dD4',
1033                 'ext': 'mp4',
1034                 'title': 'Isolation - Mind Field (Ep 1)',
1035                 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1036                 'duration': 2085,
1037                 'upload_date': '20170118',
1038                 'uploader': 'Vsauce',
1039                 'uploader_id': 'Vsauce',
1040                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1041                 'series': 'Mind Field',
1042                 'season_number': 1,
1043                 'episode_number': 1,
1044             },
1045             'params': {
1046                 'skip_download': True,
1047             },
1048             'expected_warnings': [
1049                 'Skipping DASH manifest',
1050             ],
1051         },
1052         {
1053             # The following content has been identified by the YouTube community
1054             # as inappropriate or offensive to some audiences.
1055             'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1056             'info_dict': {
1057                 'id': '6SJNVb0GnPI',
1058                 'ext': 'mp4',
1059                 'title': 'Race Differences in Intelligence',
1060                 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1061                 'duration': 965,
1062                 'upload_date': '20140124',
1063                 'uploader': 'New Century Foundation',
1064                 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1065                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1066             },
1067             'params': {
1068                 'skip_download': True,
1069             },
1070         },
1071         {
1072             # itag 212
1073             'url': '1t24XAntNCY',
1074             'only_matching': True,
1075         },
1076         {
1077             # geo restricted to JP
1078             'url': 'sJL6WA-aGkQ',
1079             'only_matching': True,
1080         },
1081         {
1082             'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1083             'only_matching': True,
1084         },
1085         {
1086             'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1087             'only_matching': True,
1088         },
1089         {
1090             # DRM protected
1091             'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1092             'only_matching': True,
1093         },
1094         {
1095             # Video with unsupported adaptive stream type formats
1096             'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1097             'info_dict': {
1098                 'id': 'Z4Vy8R84T1U',
1099                 'ext': 'mp4',
1100                 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1101                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1102                 'duration': 433,
1103                 'upload_date': '20130923',
1104                 'uploader': 'Amelia Putri Harwita',
1105                 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1106                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1107                 'formats': 'maxcount:10',
1108             },
1109             'params': {
1110                 'skip_download': True,
1111                 'youtube_include_dash_manifest': False,
1112             },
1113         },
1114         {
1115             # Youtube Music Auto-generated description
1116             'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1117             'info_dict': {
1118                 'id': 'MgNrAu2pzNs',
1119                 'ext': 'mp4',
1120                 'title': 'Voyeur Girl',
1121                 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1122                 'upload_date': '20190312',
1123                 'uploader': 'Various Artists - Topic',
1124                 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1125                 'artist': 'Stephen',
1126                 'track': 'Voyeur Girl',
1127                 'album': 'it\'s too much love to know my dear',
1128                 'release_date': '20190313',
1129                 'release_year': 2019,
1130             },
1131             'params': {
1132                 'skip_download': True,
1133             },
1134         },
1135         {
1136             # Youtube Music Auto-generated description
1137             # Retrieve 'artist' field from 'Artist:' in video description
1138             # when it is present on youtube music video
1139             'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1140             'info_dict': {
1141                 'id': 'k0jLE7tTwjY',
1142                 'ext': 'mp4',
1143                 'title': 'Latch Feat. Sam Smith',
1144                 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1145                 'upload_date': '20150110',
1146                 'uploader': 'Various Artists - Topic',
1147                 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1148                 'artist': 'Disclosure',
1149                 'track': 'Latch Feat. Sam Smith',
1150                 'album': 'Latch Featuring Sam Smith',
1151                 'release_date': '20121008',
1152                 'release_year': 2012,
1153             },
1154             'params': {
1155                 'skip_download': True,
1156             },
1157         },
1158         {
1159             # Youtube Music Auto-generated description
1160             # handle multiple artists on youtube music video
1161             'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1162             'info_dict': {
1163                 'id': '74qn0eJSjpA',
1164                 'ext': 'mp4',
1165                 'title': 'Eastside',
1166                 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1167                 'upload_date': '20180710',
1168                 'uploader': 'Benny Blanco - Topic',
1169                 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1170                 'artist': 'benny blanco, Halsey, Khalid',
1171                 'track': 'Eastside',
1172                 'album': 'Eastside',
1173                 'release_date': '20180713',
1174                 'release_year': 2018,
1175             },
1176             'params': {
1177                 'skip_download': True,
1178             },
1179         },
1180         {
1181             # Youtube Music Auto-generated description
1182             # handle youtube music video with release_year and no release_date
1183             'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1184             'info_dict': {
1185                 'id': '-hcAI0g-f5M',
1186                 'ext': 'mp4',
1187                 'title': 'Put It On Me',
1188                 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1189                 'upload_date': '20180426',
1190                 'uploader': 'Matt Maeson - Topic',
1191                 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1192                 'artist': 'Matt Maeson',
1193                 'track': 'Put It On Me',
1194                 'album': 'The Hearse',
1195                 'release_date': None,
1196                 'release_year': 2018,
1197             },
1198             'params': {
1199                 'skip_download': True,
1200             },
1201         },
1202     ]
1203
1204     def __init__(self, *args, **kwargs):
1205         super(YoutubeIE, self).__init__(*args, **kwargs)
1206         self._player_cache = {}
1207
1208     def report_video_info_webpage_download(self, video_id):
1209         """Report attempt to download video info webpage."""
1210         self.to_screen('%s: Downloading video info webpage' % video_id)
1211
1212     def report_information_extraction(self, video_id):
1213         """Report attempt to extract video information."""
1214         self.to_screen('%s: Extracting video information' % video_id)
1215
1216     def report_unavailable_format(self, video_id, format):
1217         """Report extracted video URL."""
1218         self.to_screen('%s: Format %s not available' % (video_id, format))
1219
1220     def report_rtmp_download(self):
1221         """Indicate the download will use the RTMP protocol."""
1222         self.to_screen('RTMP download detected')
1223
1224     def _signature_cache_id(self, example_sig):
1225         """ Return a string representation of a signature """
1226         return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1227
1228     def _extract_signature_function(self, video_id, player_url, example_sig):
1229         id_m = re.match(
1230             r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1231             player_url)
1232         if not id_m:
1233             raise ExtractorError('Cannot identify player %r' % player_url)
1234         player_type = id_m.group('ext')
1235         player_id = id_m.group('id')
1236
1237         # Read from filesystem cache
1238         func_id = '%s_%s_%s' % (
1239             player_type, player_id, self._signature_cache_id(example_sig))
1240         assert os.path.basename(func_id) == func_id
1241
1242         cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1243         if cache_spec is not None:
1244             return lambda s: ''.join(s[i] for i in cache_spec)
1245
1246         download_note = (
1247             'Downloading player %s' % player_url
1248             if self._downloader.params.get('verbose') else
1249             'Downloading %s player %s' % (player_type, player_id)
1250         )
1251         if player_type == 'js':
1252             code = self._download_webpage(
1253                 player_url, video_id,
1254                 note=download_note,
1255                 errnote='Download of %s failed' % player_url)
1256             res = self._parse_sig_js(code)
1257         elif player_type == 'swf':
1258             urlh = self._request_webpage(
1259                 player_url, video_id,
1260                 note=download_note,
1261                 errnote='Download of %s failed' % player_url)
1262             code = urlh.read()
1263             res = self._parse_sig_swf(code)
1264         else:
1265             assert False, 'Invalid player type %r' % player_type
1266
1267         test_string = ''.join(map(compat_chr, range(len(example_sig))))
1268         cache_res = res(test_string)
1269         cache_spec = [ord(c) for c in cache_res]
1270
1271         self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1272         return res
1273
1274     def _print_sig_code(self, func, example_sig):
1275         def gen_sig_code(idxs):
1276             def _genslice(start, end, step):
1277                 starts = '' if start == 0 else str(start)
1278                 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1279                 steps = '' if step == 1 else (':%d' % step)
1280                 return 's[%s%s%s]' % (starts, ends, steps)
1281
1282             step = None
1283             # Quelch pyflakes warnings - start will be set when step is set
1284             start = '(Never used)'
1285             for i, prev in zip(idxs[1:], idxs[:-1]):
1286                 if step is not None:
1287                     if i - prev == step:
1288                         continue
1289                     yield _genslice(start, prev, step)
1290                     step = None
1291                     continue
1292                 if i - prev in [-1, 1]:
1293                     step = i - prev
1294                     start = prev
1295                     continue
1296                 else:
1297                     yield 's[%d]' % prev
1298             if step is None:
1299                 yield 's[%d]' % i
1300             else:
1301                 yield _genslice(start, i, step)
1302
1303         test_string = ''.join(map(compat_chr, range(len(example_sig))))
1304         cache_res = func(test_string)
1305         cache_spec = [ord(c) for c in cache_res]
1306         expr_code = ' + '.join(gen_sig_code(cache_spec))
1307         signature_id_tuple = '(%s)' % (
1308             ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1309         code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1310                 '    return %s\n') % (signature_id_tuple, expr_code)
1311         self.to_screen('Extracted signature function:\n' + code)
1312
1313     def _parse_sig_js(self, jscode):
1314         funcname = self._search_regex(
1315             (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1316              r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1317              r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
1318              r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1319              r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1320             jscode, 'Initial JS player signature function name', group='sig')
1321
1322         jsi = JSInterpreter(jscode)
1323         initial_function = jsi.extract_function(funcname)
1324         return lambda s: initial_function([s])
1325
1326     def _parse_sig_swf(self, file_contents):
1327         swfi = SWFInterpreter(file_contents)
1328         TARGET_CLASSNAME = 'SignatureDecipher'
1329         searched_class = swfi.extract_class(TARGET_CLASSNAME)
1330         initial_function = swfi.extract_function(searched_class, 'decipher')
1331         return lambda s: initial_function([s])
1332
1333     def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1334         """Turn the encrypted s field into a working signature"""
1335
1336         if player_url is None:
1337             raise ExtractorError('Cannot decrypt signature without player_url')
1338
1339         if player_url.startswith('//'):
1340             player_url = 'https:' + player_url
1341         elif not re.match(r'https?://', player_url):
1342             player_url = compat_urlparse.urljoin(
1343                 'https://www.youtube.com', player_url)
1344         try:
1345             player_id = (player_url, self._signature_cache_id(s))
1346             if player_id not in self._player_cache:
1347                 func = self._extract_signature_function(
1348                     video_id, player_url, s
1349                 )
1350                 self._player_cache[player_id] = func
1351             func = self._player_cache[player_id]
1352             if self._downloader.params.get('youtube_print_sig_code'):
1353                 self._print_sig_code(func, s)
1354             return func(s)
1355         except Exception as e:
1356             tb = traceback.format_exc()
1357             raise ExtractorError(
1358                 'Signature extraction failed: ' + tb, cause=e)
1359
1360     def _get_subtitles(self, video_id, webpage):
1361         try:
1362             subs_doc = self._download_xml(
1363                 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1364                 video_id, note=False)
1365         except ExtractorError as err:
1366             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1367             return {}
1368
1369         sub_lang_list = {}
1370         for track in subs_doc.findall('track'):
1371             lang = track.attrib['lang_code']
1372             if lang in sub_lang_list:
1373                 continue
1374             sub_formats = []
1375             for ext in self._SUBTITLE_FORMATS:
1376                 params = compat_urllib_parse_urlencode({
1377                     'lang': lang,
1378                     'v': video_id,
1379                     'fmt': ext,
1380                     'name': track.attrib['name'].encode('utf-8'),
1381                 })
1382                 sub_formats.append({
1383                     'url': 'https://www.youtube.com/api/timedtext?' + params,
1384                     'ext': ext,
1385                 })
1386             sub_lang_list[lang] = sub_formats
1387         if not sub_lang_list:
1388             self._downloader.report_warning('video doesn\'t have subtitles')
1389             return {}
1390         return sub_lang_list
1391
1392     def _get_ytplayer_config(self, video_id, webpage):
1393         patterns = (
1394             # User data may contain arbitrary character sequences that may affect
1395             # JSON extraction with regex, e.g. when '};' is contained the second
1396             # regex won't capture the whole JSON. Yet working around by trying more
1397             # concrete regex first keeping in mind proper quoted string handling
1398             # to be implemented in future that will replace this workaround (see
1399             # https://github.com/ytdl-org/youtube-dl/issues/7468,
1400             # https://github.com/ytdl-org/youtube-dl/pull/7599)
1401             r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1402             r';ytplayer\.config\s*=\s*({.+?});',
1403         )
1404         config = self._search_regex(
1405             patterns, webpage, 'ytplayer.config', default=None)
1406         if config:
1407             return self._parse_json(
1408                 uppercase_escape(config), video_id, fatal=False)
1409
1410     def _get_automatic_captions(self, video_id, webpage):
1411         """We need the webpage for getting the captions url, pass it as an
1412            argument to speed up the process."""
1413         self.to_screen('%s: Looking for automatic captions' % video_id)
1414         player_config = self._get_ytplayer_config(video_id, webpage)
1415         err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1416         if not player_config:
1417             self._downloader.report_warning(err_msg)
1418             return {}
1419         try:
1420             args = player_config['args']
1421             caption_url = args.get('ttsurl')
1422             if caption_url:
1423                 timestamp = args['timestamp']
1424                 # We get the available subtitles
1425                 list_params = compat_urllib_parse_urlencode({
1426                     'type': 'list',
1427                     'tlangs': 1,
1428                     'asrs': 1,
1429                 })
1430                 list_url = caption_url + '&' + list_params
1431                 caption_list = self._download_xml(list_url, video_id)
1432                 original_lang_node = caption_list.find('track')
1433                 if original_lang_node is None:
1434                     self._downloader.report_warning('Video doesn\'t have automatic captions')
1435                     return {}
1436                 original_lang = original_lang_node.attrib['lang_code']
1437                 caption_kind = original_lang_node.attrib.get('kind', '')
1438
1439                 sub_lang_list = {}
1440                 for lang_node in caption_list.findall('target'):
1441                     sub_lang = lang_node.attrib['lang_code']
1442                     sub_formats = []
1443                     for ext in self._SUBTITLE_FORMATS:
1444                         params = compat_urllib_parse_urlencode({
1445                             'lang': original_lang,
1446                             'tlang': sub_lang,
1447                             'fmt': ext,
1448                             'ts': timestamp,
1449                             'kind': caption_kind,
1450                         })
1451                         sub_formats.append({
1452                             'url': caption_url + '&' + params,
1453                             'ext': ext,
1454                         })
1455                     sub_lang_list[sub_lang] = sub_formats
1456                 return sub_lang_list
1457
1458             def make_captions(sub_url, sub_langs):
1459                 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1460                 caption_qs = compat_parse_qs(parsed_sub_url.query)
1461                 captions = {}
1462                 for sub_lang in sub_langs:
1463                     sub_formats = []
1464                     for ext in self._SUBTITLE_FORMATS:
1465                         caption_qs.update({
1466                             'tlang': [sub_lang],
1467                             'fmt': [ext],
1468                         })
1469                         sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1470                             query=compat_urllib_parse_urlencode(caption_qs, True)))
1471                         sub_formats.append({
1472                             'url': sub_url,
1473                             'ext': ext,
1474                         })
1475                     captions[sub_lang] = sub_formats
1476                 return captions
1477
1478             # New captions format as of 22.06.2017
1479             player_response = args.get('player_response')
1480             if player_response and isinstance(player_response, compat_str):
1481                 player_response = self._parse_json(
1482                     player_response, video_id, fatal=False)
1483                 if player_response:
1484                     renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1485                     base_url = renderer['captionTracks'][0]['baseUrl']
1486                     sub_lang_list = []
1487                     for lang in renderer['translationLanguages']:
1488                         lang_code = lang.get('languageCode')
1489                         if lang_code:
1490                             sub_lang_list.append(lang_code)
1491                     return make_captions(base_url, sub_lang_list)
1492
1493             # Some videos don't provide ttsurl but rather caption_tracks and
1494             # caption_translation_languages (e.g. 20LmZk1hakA)
1495             # Does not used anymore as of 22.06.2017
1496             caption_tracks = args['caption_tracks']
1497             caption_translation_languages = args['caption_translation_languages']
1498             caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1499             sub_lang_list = []
1500             for lang in caption_translation_languages.split(','):
1501                 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1502                 sub_lang = lang_qs.get('lc', [None])[0]
1503                 if sub_lang:
1504                     sub_lang_list.append(sub_lang)
1505             return make_captions(caption_url, sub_lang_list)
1506         # An extractor error can be raise by the download process if there are
1507         # no automatic captions but there are subtitles
1508         except (KeyError, IndexError, ExtractorError):
1509             self._downloader.report_warning(err_msg)
1510             return {}
1511
1512     def _mark_watched(self, video_id, video_info, player_response):
1513         playback_url = url_or_none(try_get(
1514             player_response,
1515             lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1516             video_info, lambda x: x['videostats_playback_base_url'][0]))
1517         if not playback_url:
1518             return
1519         parsed_playback_url = compat_urlparse.urlparse(playback_url)
1520         qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1521
1522         # cpn generation algorithm is reverse engineered from base.js.
1523         # In fact it works even with dummy cpn.
1524         CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1525         cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1526
1527         qs.update({
1528             'ver': ['2'],
1529             'cpn': [cpn],
1530         })
1531         playback_url = compat_urlparse.urlunparse(
1532             parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1533
1534         self._download_webpage(
1535             playback_url, video_id, 'Marking watched',
1536             'Unable to mark watched', fatal=False)
1537
1538     @staticmethod
1539     def _extract_urls(webpage):
1540         # Embedded YouTube player
1541         entries = [
1542             unescapeHTML(mobj.group('url'))
1543             for mobj in re.finditer(r'''(?x)
1544             (?:
1545                 <iframe[^>]+?src=|
1546                 data-video-url=|
1547                 <embed[^>]+?src=|
1548                 embedSWF\(?:\s*|
1549                 <object[^>]+data=|
1550                 new\s+SWFObject\(
1551             )
1552             (["\'])
1553                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1554                 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1555             \1''', webpage)]
1556
1557         # lazyYT YouTube embed
1558         entries.extend(list(map(
1559             unescapeHTML,
1560             re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1561
1562         # Wordpress "YouTube Video Importer" plugin
1563         matches = re.findall(r'''(?x)<div[^>]+
1564             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1565             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1566         entries.extend(m[-1] for m in matches)
1567
1568         return entries
1569
1570     @staticmethod
1571     def _extract_url(webpage):
1572         urls = YoutubeIE._extract_urls(webpage)
1573         return urls[0] if urls else None
1574
1575     @classmethod
1576     def extract_id(cls, url):
1577         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1578         if mobj is None:
1579             raise ExtractorError('Invalid URL: %s' % url)
1580         video_id = mobj.group(2)
1581         return video_id
1582
1583     def _extract_annotations(self, video_id):
1584         return self._download_webpage(
1585             'https://www.youtube.com/annotations_invideo', video_id,
1586             note='Downloading annotations',
1587             errnote='Unable to download video annotations', fatal=False,
1588             query={
1589                 'features': 1,
1590                 'legacy': 1,
1591                 'video_id': video_id,
1592             })
1593
1594     @staticmethod
1595     def _extract_chapters(description, duration):
1596         if not description:
1597             return None
1598         chapter_lines = re.findall(
1599             r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1600             description)
1601         if not chapter_lines:
1602             return None
1603         chapters = []
1604         for next_num, (chapter_line, time_point) in enumerate(
1605                 chapter_lines, start=1):
1606             start_time = parse_duration(time_point)
1607             if start_time is None:
1608                 continue
1609             if start_time > duration:
1610                 break
1611             end_time = (duration if next_num == len(chapter_lines)
1612                         else parse_duration(chapter_lines[next_num][1]))
1613             if end_time is None:
1614                 continue
1615             if end_time > duration:
1616                 end_time = duration
1617             if start_time > end_time:
1618                 break
1619             chapter_title = re.sub(
1620                 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1621             chapter_title = re.sub(r'\s+', ' ', chapter_title)
1622             chapters.append({
1623                 'start_time': start_time,
1624                 'end_time': end_time,
1625                 'title': chapter_title,
1626             })
1627         return chapters
1628
1629     def _real_extract(self, url):
1630         url, smuggled_data = unsmuggle_url(url, {})
1631
1632         proto = (
1633             'http' if self._downloader.params.get('prefer_insecure', False)
1634             else 'https')
1635
1636         start_time = None
1637         end_time = None
1638         parsed_url = compat_urllib_parse_urlparse(url)
1639         for component in [parsed_url.fragment, parsed_url.query]:
1640             query = compat_parse_qs(component)
1641             if start_time is None and 't' in query:
1642                 start_time = parse_duration(query['t'][0])
1643             if start_time is None and 'start' in query:
1644                 start_time = parse_duration(query['start'][0])
1645             if end_time is None and 'end' in query:
1646                 end_time = parse_duration(query['end'][0])
1647
1648         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1649         mobj = re.search(self._NEXT_URL_RE, url)
1650         if mobj:
1651             url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1652         video_id = self.extract_id(url)
1653
1654         # Get video webpage
1655         url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1656         video_webpage = self._download_webpage(url, video_id)
1657
1658         # Attempt to extract SWF player URL
1659         mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1660         if mobj is not None:
1661             player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1662         else:
1663             player_url = None
1664
1665         dash_mpds = []
1666
1667         def add_dash_mpd(video_info):
1668             dash_mpd = video_info.get('dashmpd')
1669             if dash_mpd and dash_mpd[0] not in dash_mpds:
1670                 dash_mpds.append(dash_mpd[0])
1671
1672         def add_dash_mpd_pr(pl_response):
1673             dash_mpd = url_or_none(try_get(
1674                 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1675                 compat_str))
1676             if dash_mpd and dash_mpd not in dash_mpds:
1677                 dash_mpds.append(dash_mpd)
1678
1679         is_live = None
1680         view_count = None
1681
1682         def extract_view_count(v_info):
1683             return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1684
1685         def extract_token(v_info):
1686             return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1687
1688         player_response = {}
1689
1690         # Get video info
1691         embed_webpage = None
1692         if re.search(r'player-age-gate-content">', video_webpage) is not None:
1693             age_gate = True
1694             # We simulate the access to the video from www.youtube.com/v/{video_id}
1695             # this can be viewed without login into Youtube
1696             url = proto + '://www.youtube.com/embed/%s' % video_id
1697             embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1698             data = compat_urllib_parse_urlencode({
1699                 'video_id': video_id,
1700                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1701                 'sts': self._search_regex(
1702                     r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1703             })
1704             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1705             video_info_webpage = self._download_webpage(
1706                 video_info_url, video_id,
1707                 note='Refetching age-gated info webpage',
1708                 errnote='unable to download video info webpage')
1709             video_info = compat_parse_qs(video_info_webpage)
1710             add_dash_mpd(video_info)
1711         else:
1712             age_gate = False
1713             video_info = None
1714             sts = None
1715             # Try looking directly into the video webpage
1716             ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1717             if ytplayer_config:
1718                 args = ytplayer_config['args']
1719                 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1720                     # Convert to the same format returned by compat_parse_qs
1721                     video_info = dict((k, [v]) for k, v in args.items())
1722                     add_dash_mpd(video_info)
1723                 # Rental video is not rented but preview is available (e.g.
1724                 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1725                 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1726                 if not video_info and args.get('ypc_vid'):
1727                     return self.url_result(
1728                         args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1729                 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1730                     is_live = True
1731                 sts = ytplayer_config.get('sts')
1732                 if not player_response:
1733                     pl_response = str_or_none(args.get('player_response'))
1734                     if pl_response:
1735                         pl_response = self._parse_json(pl_response, video_id, fatal=False)
1736                         if isinstance(pl_response, dict):
1737                             player_response = pl_response
1738             if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1739                 add_dash_mpd_pr(player_response)
1740                 # We also try looking in get_video_info since it may contain different dashmpd
1741                 # URL that points to a DASH manifest with possibly different itag set (some itags
1742                 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1743                 # manifest pointed by get_video_info's dashmpd).
1744                 # The general idea is to take a union of itags of both DASH manifests (for example
1745                 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1746                 self.report_video_info_webpage_download(video_id)
1747                 for el in ('embedded', 'detailpage', 'vevo', ''):
1748                     query = {
1749                         'video_id': video_id,
1750                         'ps': 'default',
1751                         'eurl': '',
1752                         'gl': 'US',
1753                         'hl': 'en',
1754                     }
1755                     if el:
1756                         query['el'] = el
1757                     if sts:
1758                         query['sts'] = sts
1759                     video_info_webpage = self._download_webpage(
1760                         '%s://www.youtube.com/get_video_info' % proto,
1761                         video_id, note=False,
1762                         errnote='unable to download video info webpage',
1763                         fatal=False, query=query)
1764                     if not video_info_webpage:
1765                         continue
1766                     get_video_info = compat_parse_qs(video_info_webpage)
1767                     if not player_response:
1768                         pl_response = get_video_info.get('player_response', [None])[0]
1769                         if isinstance(pl_response, dict):
1770                             player_response = pl_response
1771                             add_dash_mpd_pr(player_response)
1772                     add_dash_mpd(get_video_info)
1773                     if view_count is None:
1774                         view_count = extract_view_count(get_video_info)
1775                     if not video_info:
1776                         video_info = get_video_info
1777                     get_token = extract_token(get_video_info)
1778                     if get_token:
1779                         # Different get_video_info requests may report different results, e.g.
1780                         # some may report video unavailability, but some may serve it without
1781                         # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1782                         # the original webpage as well as el=info and el=embedded get_video_info
1783                         # requests report video unavailability due to geo restriction while
1784                         # el=detailpage succeeds and returns valid data). This is probably
1785                         # due to YouTube measures against IP ranges of hosting providers.
1786                         # Working around by preferring the first succeeded video_info containing
1787                         # the token if no such video_info yet was found.
1788                         token = extract_token(video_info)
1789                         if not token:
1790                             video_info = get_video_info
1791                         break
1792
1793         def extract_unavailable_message():
1794             return self._html_search_regex(
1795                 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1796                 video_webpage, 'unavailable message', default=None)
1797
1798         if not video_info:
1799             unavailable_message = extract_unavailable_message()
1800             if not unavailable_message:
1801                 unavailable_message = 'Unable to extract video data'
1802             raise ExtractorError(
1803                 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1804
1805         video_details = try_get(
1806             player_response, lambda x: x['videoDetails'], dict) or {}
1807
1808         # title
1809         if 'title' in video_info:
1810             video_title = video_info['title'][0]
1811         elif 'title' in player_response:
1812             video_title = video_details['title']
1813         else:
1814             self._downloader.report_warning('Unable to extract video title')
1815             video_title = '_'
1816
1817         # description
1818         description_original = video_description = get_element_by_id("eow-description", video_webpage)
1819         if video_description:
1820
1821             def replace_url(m):
1822                 redir_url = compat_urlparse.urljoin(url, m.group(1))
1823                 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1824                 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1825                     qs = compat_parse_qs(parsed_redir_url.query)
1826                     q = qs.get('q')
1827                     if q and q[0]:
1828                         return q[0]
1829                 return redir_url
1830
1831             description_original = video_description = re.sub(r'''(?x)
1832                 <a\s+
1833                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
1834                     (?:title|href)="([^"]+)"\s+
1835                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
1836                     class="[^"]*"[^>]*>
1837                 [^<]+\.{3}\s*
1838                 </a>
1839             ''', replace_url, video_description)
1840             video_description = clean_html(video_description)
1841         else:
1842             fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1843             if fd_mobj:
1844                 video_description = unescapeHTML(fd_mobj.group(1))
1845             else:
1846                 video_description = ''
1847
1848         if not smuggled_data.get('force_singlefeed', False):
1849             if not self._downloader.params.get('noplaylist'):
1850                 multifeed_metadata_list = try_get(
1851                     player_response,
1852                     lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1853                     compat_str) or try_get(
1854                     video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1855                 if multifeed_metadata_list:
1856                     entries = []
1857                     feed_ids = []
1858                     for feed in multifeed_metadata_list.split(','):
1859                         # Unquote should take place before split on comma (,) since textual
1860                         # fields may contain comma as well (see
1861                         # https://github.com/ytdl-org/youtube-dl/issues/8536)
1862                         feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1863                         entries.append({
1864                             '_type': 'url_transparent',
1865                             'ie_key': 'Youtube',
1866                             'url': smuggle_url(
1867                                 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1868                                 {'force_singlefeed': True}),
1869                             'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1870                         })
1871                         feed_ids.append(feed_data['id'][0])
1872                     self.to_screen(
1873                         'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1874                         % (', '.join(feed_ids), video_id))
1875                     return self.playlist_result(entries, video_id, video_title, video_description)
1876             else:
1877                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1878
1879         if view_count is None:
1880             view_count = extract_view_count(video_info)
1881         if view_count is None and video_details:
1882             view_count = int_or_none(video_details.get('viewCount'))
1883
1884         # Check for "rental" videos
1885         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1886             raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1887
1888         def _extract_filesize(media_url):
1889             return int_or_none(self._search_regex(
1890                 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1891
1892         if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1893             self.report_rtmp_download()
1894             formats = [{
1895                 'format_id': '_rtmp',
1896                 'protocol': 'rtmp',
1897                 'url': video_info['conn'][0],
1898                 'player_url': player_url,
1899             }]
1900         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):
1901             encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1902             if 'rtmpe%3Dyes' in encoded_url_map:
1903                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1904             formats_spec = {}
1905             fmt_list = video_info.get('fmt_list', [''])[0]
1906             if fmt_list:
1907                 for fmt in fmt_list.split(','):
1908                     spec = fmt.split('/')
1909                     if len(spec) > 1:
1910                         width_height = spec[1].split('x')
1911                         if len(width_height) == 2:
1912                             formats_spec[spec[0]] = {
1913                                 'resolution': spec[1],
1914                                 'width': int_or_none(width_height[0]),
1915                                 'height': int_or_none(width_height[1]),
1916                             }
1917             q = qualities(['small', 'medium', 'hd720'])
1918             streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1919             if streaming_formats:
1920                 for fmt in streaming_formats:
1921                     itag = str_or_none(fmt.get('itag'))
1922                     if not itag:
1923                         continue
1924                     quality = fmt.get('quality')
1925                     quality_label = fmt.get('qualityLabel') or quality
1926                     formats_spec[itag] = {
1927                         'asr': int_or_none(fmt.get('audioSampleRate')),
1928                         'filesize': int_or_none(fmt.get('contentLength')),
1929                         'format_note': quality_label,
1930                         'fps': int_or_none(fmt.get('fps')),
1931                         'height': int_or_none(fmt.get('height')),
1932                         'quality': q(quality),
1933                         # bitrate for itag 43 is always 2147483647
1934                         'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1935                         'width': int_or_none(fmt.get('width')),
1936                     }
1937             formats = []
1938             for url_data_str in encoded_url_map.split(','):
1939                 url_data = compat_parse_qs(url_data_str)
1940                 if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
1941                     continue
1942                 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1943                 # Unsupported FORMAT_STREAM_TYPE_OTF
1944                 if stream_type == 3:
1945                     continue
1946                 format_id = url_data['itag'][0]
1947                 url = url_data['url'][0]
1948
1949                 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1950                     ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1951                     jsplayer_url_json = self._search_regex(
1952                         ASSETS_RE,
1953                         embed_webpage if age_gate else video_webpage,
1954                         'JS player URL (1)', default=None)
1955                     if not jsplayer_url_json and not age_gate:
1956                         # We need the embed website after all
1957                         if embed_webpage is None:
1958                             embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1959                             embed_webpage = self._download_webpage(
1960                                 embed_url, video_id, 'Downloading embed webpage')
1961                         jsplayer_url_json = self._search_regex(
1962                             ASSETS_RE, embed_webpage, 'JS player URL')
1963
1964                     player_url = json.loads(jsplayer_url_json)
1965                     if player_url is None:
1966                         player_url_json = self._search_regex(
1967                             r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1968                             video_webpage, 'age gate player URL')
1969                         player_url = json.loads(player_url_json)
1970
1971                 if 'sig' in url_data:
1972                     url += '&signature=' + url_data['sig'][0]
1973                 elif 's' in url_data:
1974                     encrypted_sig = url_data['s'][0]
1975
1976                     if self._downloader.params.get('verbose'):
1977                         if player_url is None:
1978                             player_version = 'unknown'
1979                             player_desc = 'unknown'
1980                         else:
1981                             if player_url.endswith('swf'):
1982                                 player_version = self._search_regex(
1983                                     r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1984                                     'flash player', fatal=False)
1985                                 player_desc = 'flash player %s' % player_version
1986                             else:
1987                                 player_version = self._search_regex(
1988                                     [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1989                                      r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1990                                     player_url,
1991                                     'html5 player', fatal=False)
1992                                 player_desc = 'html5 player %s' % player_version
1993
1994                         parts_sizes = self._signature_cache_id(encrypted_sig)
1995                         self.to_screen('{%s} signature length %s, %s' %
1996                                        (format_id, parts_sizes, player_desc))
1997
1998                     signature = self._decrypt_signature(
1999                         encrypted_sig, video_id, player_url, age_gate)
2000                     sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2001                     url += '&%s=%s' % (sp, signature)
2002                 if 'ratebypass' not in url:
2003                     url += '&ratebypass=yes'
2004
2005                 dct = {
2006                     'format_id': format_id,
2007                     'url': url,
2008                     'player_url': player_url,
2009                 }
2010                 if format_id in self._formats:
2011                     dct.update(self._formats[format_id])
2012                 if format_id in formats_spec:
2013                     dct.update(formats_spec[format_id])
2014
2015                 # Some itags are not included in DASH manifest thus corresponding formats will
2016                 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2017                 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2018                 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2019                 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2020
2021                 filesize = int_or_none(url_data.get(
2022                     'clen', [None])[0]) or _extract_filesize(url)
2023
2024                 quality = url_data.get('quality', [None])[0]
2025
2026                 more_fields = {
2027                     'filesize': filesize,
2028                     'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2029                     'width': width,
2030                     'height': height,
2031                     'fps': int_or_none(url_data.get('fps', [None])[0]),
2032                     'format_note': url_data.get('quality_label', [None])[0] or quality,
2033                     'quality': q(quality),
2034                 }
2035                 for key, value in more_fields.items():
2036                     if value:
2037                         dct[key] = value
2038                 type_ = url_data.get('type', [None])[0]
2039                 if type_:
2040                     type_split = type_.split(';')
2041                     kind_ext = type_split[0].split('/')
2042                     if len(kind_ext) == 2:
2043                         kind, _ = kind_ext
2044                         dct['ext'] = mimetype2ext(type_split[0])
2045                         if kind in ('audio', 'video'):
2046                             codecs = None
2047                             for mobj in re.finditer(
2048                                     r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2049                                 if mobj.group('key') == 'codecs':
2050                                     codecs = mobj.group('val')
2051                                     break
2052                             if codecs:
2053                                 dct.update(parse_codecs(codecs))
2054                 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2055                     dct['downloader_options'] = {
2056                         # Youtube throttles chunks >~10M
2057                         'http_chunk_size': 10485760,
2058                     }
2059                 formats.append(dct)
2060         else:
2061             manifest_url = (
2062                 url_or_none(try_get(
2063                     player_response,
2064                     lambda x: x['streamingData']['hlsManifestUrl'],
2065                     compat_str))
2066                 or url_or_none(try_get(
2067                     video_info, lambda x: x['hlsvp'][0], compat_str)))
2068             if manifest_url:
2069                 formats = []
2070                 m3u8_formats = self._extract_m3u8_formats(
2071                     manifest_url, video_id, 'mp4', fatal=False)
2072                 for a_format in m3u8_formats:
2073                     itag = self._search_regex(
2074                         r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2075                     if itag:
2076                         a_format['format_id'] = itag
2077                         if itag in self._formats:
2078                             dct = self._formats[itag].copy()
2079                             dct.update(a_format)
2080                             a_format = dct
2081                     a_format['player_url'] = player_url
2082                     # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2083                     a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2084                     formats.append(a_format)
2085             else:
2086                 error_message = clean_html(video_info.get('reason', [None])[0])
2087                 if not error_message:
2088                     error_message = extract_unavailable_message()
2089                 if error_message:
2090                     raise ExtractorError(error_message, expected=True)
2091                 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2092
2093         # uploader
2094         video_uploader = try_get(
2095             video_info, lambda x: x['author'][0],
2096             compat_str) or str_or_none(video_details.get('author'))
2097         if video_uploader:
2098             video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2099         else:
2100             self._downloader.report_warning('unable to extract uploader name')
2101
2102         # uploader_id
2103         video_uploader_id = None
2104         video_uploader_url = None
2105         mobj = re.search(
2106             r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2107             video_webpage)
2108         if mobj is not None:
2109             video_uploader_id = mobj.group('uploader_id')
2110             video_uploader_url = mobj.group('uploader_url')
2111         else:
2112             self._downloader.report_warning('unable to extract uploader nickname')
2113
2114         channel_id = (
2115             str_or_none(video_details.get('channelId'))
2116             or self._html_search_meta(
2117                 'channelId', video_webpage, 'channel id', default=None)
2118             or self._search_regex(
2119                 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2120                 video_webpage, 'channel id', default=None, group='id'))
2121         channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2122
2123         # thumbnail image
2124         # We try first to get a high quality image:
2125         m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2126                             video_webpage, re.DOTALL)
2127         if m_thumb is not None:
2128             video_thumbnail = m_thumb.group(1)
2129         elif 'thumbnail_url' not in video_info:
2130             self._downloader.report_warning('unable to extract video thumbnail')
2131             video_thumbnail = None
2132         else:   # don't panic if we can't find it
2133             video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2134
2135         # upload date
2136         upload_date = self._html_search_meta(
2137             'datePublished', video_webpage, 'upload date', default=None)
2138         if not upload_date:
2139             upload_date = self._search_regex(
2140                 [r'(?s)id="eow-date.*?>(.*?)</span>',
2141                  r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2142                 video_webpage, 'upload date', default=None)
2143         upload_date = unified_strdate(upload_date)
2144
2145         video_license = self._html_search_regex(
2146             r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2147             video_webpage, 'license', default=None)
2148
2149         m_music = re.search(
2150             r'''(?x)
2151                 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2152                 <ul[^>]*>\s*
2153                 <li>(?P<title>.+?)
2154                 by (?P<creator>.+?)
2155                 (?:
2156                     \(.+?\)|
2157                     <a[^>]*
2158                         (?:
2159                             \bhref=["\']/red[^>]*>|             # drop possible
2160                             >\s*Listen ad-free with YouTube Red # YouTube Red ad
2161                         )
2162                     .*?
2163                 )?</li
2164             ''',
2165             video_webpage)
2166         if m_music:
2167             video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2168             video_creator = clean_html(m_music.group('creator'))
2169         else:
2170             video_alt_title = video_creator = None
2171
2172         def extract_meta(field):
2173             return self._html_search_regex(
2174                 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2175                 video_webpage, field, default=None)
2176
2177         track = extract_meta('Song')
2178         artist = extract_meta('Artist')
2179         album = extract_meta('Album')
2180
2181         # Youtube Music Auto-generated description
2182         release_date = release_year = None
2183         if video_description:
2184             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)
2185             if mobj:
2186                 if not track:
2187                     track = mobj.group('track').strip()
2188                 if not artist:
2189                     artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2190                 if not album:
2191                     album = mobj.group('album'.strip())
2192                 release_year = mobj.group('release_year')
2193                 release_date = mobj.group('release_date')
2194                 if release_date:
2195                     release_date = release_date.replace('-', '')
2196                     if not release_year:
2197                         release_year = int(release_date[:4])
2198                 if release_year:
2199                     release_year = int(release_year)
2200
2201         m_episode = re.search(
2202             r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2203             video_webpage)
2204         if m_episode:
2205             series = unescapeHTML(m_episode.group('series'))
2206             season_number = int(m_episode.group('season'))
2207             episode_number = int(m_episode.group('episode'))
2208         else:
2209             series = season_number = episode_number = None
2210
2211         m_cat_container = self._search_regex(
2212             r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2213             video_webpage, 'categories', default=None)
2214         if m_cat_container:
2215             category = self._html_search_regex(
2216                 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2217                 default=None)
2218             video_categories = None if category is None else [category]
2219         else:
2220             video_categories = None
2221
2222         video_tags = [
2223             unescapeHTML(m.group('content'))
2224             for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2225
2226         def _extract_count(count_name):
2227             return str_to_int(self._search_regex(
2228                 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2229                 % re.escape(count_name),
2230                 video_webpage, count_name, default=None))
2231
2232         like_count = _extract_count('like')
2233         dislike_count = _extract_count('dislike')
2234
2235         if view_count is None:
2236             view_count = str_to_int(self._search_regex(
2237                 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2238                 'view count', default=None))
2239
2240         average_rating = (
2241             float_or_none(video_details.get('averageRating'))
2242             or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2243
2244         # subtitles
2245         video_subtitles = self.extract_subtitles(video_id, video_webpage)
2246         automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2247
2248         video_duration = try_get(
2249             video_info, lambda x: int_or_none(x['length_seconds'][0]))
2250         if not video_duration:
2251             video_duration = int_or_none(video_details.get('lengthSeconds'))
2252         if not video_duration:
2253             video_duration = parse_duration(self._html_search_meta(
2254                 'duration', video_webpage, 'video duration'))
2255
2256         # annotations
2257         video_annotations = None
2258         if self._downloader.params.get('writeannotations', False):
2259             video_annotations = self._extract_annotations(video_id)
2260
2261         chapters = self._extract_chapters(description_original, video_duration)
2262
2263         # Look for the DASH manifest
2264         if self._downloader.params.get('youtube_include_dash_manifest', True):
2265             dash_mpd_fatal = True
2266             for mpd_url in dash_mpds:
2267                 dash_formats = {}
2268                 try:
2269                     def decrypt_sig(mobj):
2270                         s = mobj.group(1)
2271                         dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2272                         return '/signature/%s' % dec_s
2273
2274                     mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2275
2276                     for df in self._extract_mpd_formats(
2277                             mpd_url, video_id, fatal=dash_mpd_fatal,
2278                             formats_dict=self._formats):
2279                         if not df.get('filesize'):
2280                             df['filesize'] = _extract_filesize(df['url'])
2281                         # Do not overwrite DASH format found in some previous DASH manifest
2282                         if df['format_id'] not in dash_formats:
2283                             dash_formats[df['format_id']] = df
2284                         # Additional DASH manifests may end up in HTTP Error 403 therefore
2285                         # allow them to fail without bug report message if we already have
2286                         # some DASH manifest succeeded. This is temporary workaround to reduce
2287                         # burst of bug reports until we figure out the reason and whether it
2288                         # can be fixed at all.
2289                         dash_mpd_fatal = False
2290                 except (ExtractorError, KeyError) as e:
2291                     self.report_warning(
2292                         'Skipping DASH manifest: %r' % e, video_id)
2293                 if dash_formats:
2294                     # Remove the formats we found through non-DASH, they
2295                     # contain less info and it can be wrong, because we use
2296                     # fixed values (for example the resolution). See
2297                     # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2298                     # example.
2299                     formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2300                     formats.extend(dash_formats.values())
2301
2302         # Check for malformed aspect ratio
2303         stretched_m = re.search(
2304             r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2305             video_webpage)
2306         if stretched_m:
2307             w = float(stretched_m.group('w'))
2308             h = float(stretched_m.group('h'))
2309             # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2310             # We will only process correct ratios.
2311             if w > 0 and h > 0:
2312                 ratio = w / h
2313                 for f in formats:
2314                     if f.get('vcodec') != 'none':
2315                         f['stretched_ratio'] = ratio
2316
2317         if not formats:
2318             token = extract_token(video_info)
2319             if not token:
2320                 if 'reason' in video_info:
2321                     if 'The uploader has not made this video available in your country.' in video_info['reason']:
2322                         regions_allowed = self._html_search_meta(
2323                             'regionsAllowed', video_webpage, default=None)
2324                         countries = regions_allowed.split(',') if regions_allowed else None
2325                         self.raise_geo_restricted(
2326                             msg=video_info['reason'][0], countries=countries)
2327                     reason = video_info['reason'][0]
2328                     if 'Invalid parameters' in reason:
2329                         unavailable_message = extract_unavailable_message()
2330                         if unavailable_message:
2331                             reason = unavailable_message
2332                     raise ExtractorError(
2333                         'YouTube said: %s' % reason,
2334                         expected=True, video_id=video_id)
2335                 else:
2336                     raise ExtractorError(
2337                         '"token" parameter not in video info for unknown reason',
2338                         video_id=video_id)
2339
2340         if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2341             raise ExtractorError('This video is DRM protected.', expected=True)
2342
2343         self._sort_formats(formats)
2344
2345         self.mark_watched(video_id, video_info, player_response)
2346
2347         return {
2348             'id': video_id,
2349             'uploader': video_uploader,
2350             'uploader_id': video_uploader_id,
2351             'uploader_url': video_uploader_url,
2352             'channel_id': channel_id,
2353             'channel_url': channel_url,
2354             'upload_date': upload_date,
2355             'license': video_license,
2356             'creator': video_creator or artist,
2357             'title': video_title,
2358             'alt_title': video_alt_title or track,
2359             'thumbnail': video_thumbnail,
2360             'description': video_description,
2361             'categories': video_categories,
2362             'tags': video_tags,
2363             'subtitles': video_subtitles,
2364             'automatic_captions': automatic_captions,
2365             'duration': video_duration,
2366             'age_limit': 18 if age_gate else 0,
2367             'annotations': video_annotations,
2368             'chapters': chapters,
2369             'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2370             'view_count': view_count,
2371             'like_count': like_count,
2372             'dislike_count': dislike_count,
2373             'average_rating': average_rating,
2374             'formats': formats,
2375             'is_live': is_live,
2376             'start_time': start_time,
2377             'end_time': end_time,
2378             'series': series,
2379             'season_number': season_number,
2380             'episode_number': episode_number,
2381             'track': track,
2382             'artist': artist,
2383             'album': album,
2384             'release_date': release_date,
2385             'release_year': release_year,
2386         }
2387
2388
2389 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2390     IE_DESC = 'YouTube.com playlists'
2391     _VALID_URL = r"""(?x)(?:
2392                         (?:https?://)?
2393                         (?:\w+\.)?
2394                         (?:
2395                             (?:
2396                                 youtube\.com|
2397                                 invidio\.us
2398                             )
2399                             /
2400                             (?:
2401                                (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2402                                \? (?:.*?[&;])*? (?:p|a|list)=
2403                             |  p/
2404                             )|
2405                             youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2406                         )
2407                         (
2408                             (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2409                             # Top tracks, they can also include dots
2410                             |(?:MC)[\w\.]*
2411                         )
2412                         .*
2413                      |
2414                         (%(playlist_id)s)
2415                      )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2416     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2417     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2418     IE_NAME = 'youtube:playlist'
2419     _TESTS = [{
2420         'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2421         'info_dict': {
2422             'title': 'ytdl test PL',
2423             'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2424         },
2425         'playlist_count': 3,
2426     }, {
2427         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2428         'info_dict': {
2429             'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2430             'title': 'YDL_Empty_List',
2431         },
2432         'playlist_count': 0,
2433         'skip': 'This playlist is private',
2434     }, {
2435         'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2436         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2437         'info_dict': {
2438             'title': '29C3: Not my department',
2439             'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2440         },
2441         'playlist_count': 95,
2442     }, {
2443         'note': 'issue #673',
2444         'url': 'PLBB231211A4F62143',
2445         'info_dict': {
2446             'title': '[OLD]Team Fortress 2 (Class-based LP)',
2447             'id': 'PLBB231211A4F62143',
2448         },
2449         'playlist_mincount': 26,
2450     }, {
2451         'note': 'Large playlist',
2452         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2453         'info_dict': {
2454             'title': 'Uploads from Cauchemar',
2455             'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2456         },
2457         'playlist_mincount': 799,
2458     }, {
2459         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2460         'info_dict': {
2461             'title': 'YDL_safe_search',
2462             'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2463         },
2464         'playlist_count': 2,
2465         'skip': 'This playlist is private',
2466     }, {
2467         'note': 'embedded',
2468         'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2469         'playlist_count': 4,
2470         'info_dict': {
2471             'title': 'JODA15',
2472             'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2473         }
2474     }, {
2475         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2476         'playlist_mincount': 485,
2477         'info_dict': {
2478             'title': '2017 華語最新單曲 (2/24更新)',
2479             'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2480         }
2481     }, {
2482         'note': 'Embedded SWF player',
2483         'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2484         'playlist_count': 4,
2485         'info_dict': {
2486             'title': 'JODA7',
2487             'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2488         }
2489     }, {
2490         'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2491         'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2492         'info_dict': {
2493             'title': 'Uploads from Interstellar Movie',
2494             'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2495         },
2496         'playlist_mincount': 21,
2497     }, {
2498         # Playlist URL that does not actually serve a playlist
2499         'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2500         'info_dict': {
2501             'id': 'FqZTN594JQw',
2502             'ext': 'webm',
2503             'title': "Smiley's People 01 detective, Adventure Series, Action",
2504             'uploader': 'STREEM',
2505             'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2506             'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2507             'upload_date': '20150526',
2508             'license': 'Standard YouTube License',
2509             'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2510             'categories': ['People & Blogs'],
2511             'tags': list,
2512             'view_count': int,
2513             'like_count': int,
2514             'dislike_count': int,
2515         },
2516         'params': {
2517             'skip_download': True,
2518         },
2519         'add_ie': [YoutubeIE.ie_key()],
2520     }, {
2521         'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2522         'info_dict': {
2523             'id': 'yeWKywCrFtk',
2524             'ext': 'mp4',
2525             'title': 'Small Scale Baler and Braiding Rugs',
2526             'uploader': 'Backus-Page House Museum',
2527             'uploader_id': 'backuspagemuseum',
2528             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2529             'upload_date': '20161008',
2530             'license': 'Standard YouTube License',
2531             'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2532             'categories': ['Nonprofits & Activism'],
2533             'tags': list,
2534             'like_count': int,
2535             'dislike_count': int,
2536         },
2537         'params': {
2538             'noplaylist': True,
2539             'skip_download': True,
2540         },
2541     }, {
2542         'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2543         'only_matching': True,
2544     }, {
2545         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2546         'only_matching': True,
2547     }, {
2548         # music album playlist
2549         'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2550         'only_matching': True,
2551     }, {
2552         'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2553         'only_matching': True,
2554     }]
2555
2556     def _real_initialize(self):
2557         self._login()
2558
2559     def _extract_mix(self, playlist_id):
2560         # The mixes are generated from a single video
2561         # the id of the playlist is just 'RD' + video_id
2562         ids = []
2563         last_id = playlist_id[-11:]
2564         for n in itertools.count(1):
2565             url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2566             webpage = self._download_webpage(
2567                 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2568             new_ids = orderedSet(re.findall(
2569                 r'''(?xs)data-video-username=".*?".*?
2570                            href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2571                 webpage))
2572             # Fetch new pages until all the videos are repeated, it seems that
2573             # there are always 51 unique videos.
2574             new_ids = [_id for _id in new_ids if _id not in ids]
2575             if not new_ids:
2576                 break
2577             ids.extend(new_ids)
2578             last_id = ids[-1]
2579
2580         url_results = self._ids_to_results(ids)
2581
2582         search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2583         title_span = (
2584             search_title('playlist-title')
2585             or search_title('title long-title')
2586             or search_title('title'))
2587         title = clean_html(title_span)
2588
2589         return self.playlist_result(url_results, playlist_id, title)
2590
2591     def _extract_playlist(self, playlist_id):
2592         url = self._TEMPLATE_URL % playlist_id
2593         page = self._download_webpage(url, playlist_id)
2594
2595         # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2596         for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2597             match = match.strip()
2598             # Check if the playlist exists or is private
2599             mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2600             if mobj:
2601                 reason = mobj.group('reason')
2602                 message = 'This playlist %s' % reason
2603                 if 'private' in reason:
2604                     message += ', use --username or --netrc to access it'
2605                 message += '.'
2606                 raise ExtractorError(message, expected=True)
2607             elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2608                 raise ExtractorError(
2609                     'Invalid parameters. Maybe URL is incorrect.',
2610                     expected=True)
2611             elif re.match(r'[^<]*Choose your language[^<]*', match):
2612                 continue
2613             else:
2614                 self.report_warning('Youtube gives an alert message: ' + match)
2615
2616         playlist_title = self._html_search_regex(
2617             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2618             page, 'title', default=None)
2619
2620         _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2621         uploader = self._search_regex(
2622             r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2623             page, 'uploader', default=None)
2624         mobj = re.search(
2625             r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2626             page)
2627         if mobj:
2628             uploader_id = mobj.group('uploader_id')
2629             uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2630         else:
2631             uploader_id = uploader_url = None
2632
2633         has_videos = True
2634
2635         if not playlist_title:
2636             try:
2637                 # Some playlist URLs don't actually serve a playlist (e.g.
2638                 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2639                 next(self._entries(page, playlist_id))
2640             except StopIteration:
2641                 has_videos = False
2642
2643         playlist = self.playlist_result(
2644             self._entries(page, playlist_id), playlist_id, playlist_title)
2645         playlist.update({
2646             'uploader': uploader,
2647             'uploader_id': uploader_id,
2648             'uploader_url': uploader_url,
2649         })
2650
2651         return has_videos, playlist
2652
2653     def _check_download_just_video(self, url, playlist_id):
2654         # Check if it's a video-specific URL
2655         query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2656         video_id = query_dict.get('v', [None])[0] or self._search_regex(
2657             r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2658             'video id', default=None)
2659         if video_id:
2660             if self._downloader.params.get('noplaylist'):
2661                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2662                 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2663             else:
2664                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2665                 return video_id, None
2666         return None, None
2667
2668     def _real_extract(self, url):
2669         # Extract playlist id
2670         mobj = re.match(self._VALID_URL, url)
2671         if mobj is None:
2672             raise ExtractorError('Invalid URL: %s' % url)
2673         playlist_id = mobj.group(1) or mobj.group(2)
2674
2675         video_id, video = self._check_download_just_video(url, playlist_id)
2676         if video:
2677             return video
2678
2679         if playlist_id.startswith(('RD', 'UL', 'PU')):
2680             # Mixes require a custom extraction process
2681             return self._extract_mix(playlist_id)
2682
2683         has_videos, playlist = self._extract_playlist(playlist_id)
2684         if has_videos or not video_id:
2685             return playlist
2686
2687         # Some playlist URLs don't actually serve a playlist (see
2688         # https://github.com/ytdl-org/youtube-dl/issues/10537).
2689         # Fallback to plain video extraction if there is a video id
2690         # along with playlist id.
2691         return self.url_result(video_id, 'Youtube', video_id=video_id)
2692
2693
2694 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2695     IE_DESC = 'YouTube.com channels'
2696     _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2697     _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2698     _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2699     IE_NAME = 'youtube:channel'
2700     _TESTS = [{
2701         'note': 'paginated channel',
2702         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2703         'playlist_mincount': 91,
2704         'info_dict': {
2705             'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2706             'title': 'Uploads from lex will',
2707         }
2708     }, {
2709         'note': 'Age restricted channel',
2710         # from https://www.youtube.com/user/DeusExOfficial
2711         'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2712         'playlist_mincount': 64,
2713         'info_dict': {
2714             'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2715             'title': 'Uploads from Deus Ex',
2716         },
2717     }, {
2718         'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2719         'only_matching': True,
2720     }]
2721
2722     @classmethod
2723     def suitable(cls, url):
2724         return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2725                 else super(YoutubeChannelIE, cls).suitable(url))
2726
2727     def _build_template_url(self, url, channel_id):
2728         return self._TEMPLATE_URL % channel_id
2729
2730     def _real_extract(self, url):
2731         channel_id = self._match_id(url)
2732
2733         url = self._build_template_url(url, channel_id)
2734
2735         # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2736         # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2737         # otherwise fallback on channel by page extraction
2738         channel_page = self._download_webpage(
2739             url + '?view=57', channel_id,
2740             'Downloading channel page', fatal=False)
2741         if channel_page is False:
2742             channel_playlist_id = False
2743         else:
2744             channel_playlist_id = self._html_search_meta(
2745                 'channelId', channel_page, 'channel id', default=None)
2746             if not channel_playlist_id:
2747                 channel_url = self._html_search_meta(
2748                     ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2749                     channel_page, 'channel url', default=None)
2750                 if channel_url:
2751                     channel_playlist_id = self._search_regex(
2752                         r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2753                         channel_url, 'channel id', default=None)
2754         if channel_playlist_id and channel_playlist_id.startswith('UC'):
2755             playlist_id = 'UU' + channel_playlist_id[2:]
2756             return self.url_result(
2757                 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2758
2759         channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2760         autogenerated = re.search(r'''(?x)
2761                 class="[^"]*?(?:
2762                     channel-header-autogenerated-label|
2763                     yt-channel-title-autogenerated
2764                 )[^"]*"''', channel_page) is not None
2765
2766         if autogenerated:
2767             # The videos are contained in a single page
2768             # the ajax pages can't be used, they are empty
2769             entries = [
2770                 self.url_result(
2771                     video_id, 'Youtube', video_id=video_id,
2772                     video_title=video_title)
2773                 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2774             return self.playlist_result(entries, channel_id)
2775
2776         try:
2777             next(self._entries(channel_page, channel_id))
2778         except StopIteration:
2779             alert_message = self._html_search_regex(
2780                 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2781                 channel_page, 'alert', default=None, group='alert')
2782             if alert_message:
2783                 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2784
2785         return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2786
2787
2788 class YoutubeUserIE(YoutubeChannelIE):
2789     IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2790     _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_-]+)'
2791     _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2792     IE_NAME = 'youtube:user'
2793
2794     _TESTS = [{
2795         'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2796         'playlist_mincount': 320,
2797         'info_dict': {
2798             'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2799             'title': 'Uploads from The Linux Foundation',
2800         }
2801     }, {
2802         # Only available via https://www.youtube.com/c/12minuteathlete/videos
2803         # but not https://www.youtube.com/user/12minuteathlete/videos
2804         'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2805         'playlist_mincount': 249,
2806         'info_dict': {
2807             'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2808             'title': 'Uploads from 12 Minute Athlete',
2809         }
2810     }, {
2811         'url': 'ytuser:phihag',
2812         'only_matching': True,
2813     }, {
2814         'url': 'https://www.youtube.com/c/gametrailers',
2815         'only_matching': True,
2816     }, {
2817         'url': 'https://www.youtube.com/gametrailers',
2818         'only_matching': True,
2819     }, {
2820         # This channel is not available, geo restricted to JP
2821         'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2822         'only_matching': True,
2823     }]
2824
2825     @classmethod
2826     def suitable(cls, url):
2827         # Don't return True if the url can be extracted with other youtube
2828         # extractor, the regex would is too permissive and it would match.
2829         other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2830         if any(ie.suitable(url) for ie in other_yt_ies):
2831             return False
2832         else:
2833             return super(YoutubeUserIE, cls).suitable(url)
2834
2835     def _build_template_url(self, url, channel_id):
2836         mobj = re.match(self._VALID_URL, url)
2837         return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2838
2839
2840 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2841     IE_DESC = 'YouTube.com live streams'
2842     _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2843     IE_NAME = 'youtube:live'
2844
2845     _TESTS = [{
2846         'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2847         'info_dict': {
2848             'id': 'a48o2S1cPoo',
2849             'ext': 'mp4',
2850             'title': 'The Young Turks - Live Main Show',
2851             'uploader': 'The Young Turks',
2852             'uploader_id': 'TheYoungTurks',
2853             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2854             'upload_date': '20150715',
2855             'license': 'Standard YouTube License',
2856             'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2857             'categories': ['News & Politics'],
2858             'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2859             'like_count': int,
2860             'dislike_count': int,
2861         },
2862         'params': {
2863             'skip_download': True,
2864         },
2865     }, {
2866         'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2867         'only_matching': True,
2868     }, {
2869         'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2870         'only_matching': True,
2871     }, {
2872         'url': 'https://www.youtube.com/TheYoungTurks/live',
2873         'only_matching': True,
2874     }]
2875
2876     def _real_extract(self, url):
2877         mobj = re.match(self._VALID_URL, url)
2878         channel_id = mobj.group('id')
2879         base_url = mobj.group('base_url')
2880         webpage = self._download_webpage(url, channel_id, fatal=False)
2881         if webpage:
2882             page_type = self._og_search_property(
2883                 'type', webpage, 'page type', default='')
2884             video_id = self._html_search_meta(
2885                 'videoId', webpage, 'video id', default=None)
2886             if page_type.startswith('video') and video_id and re.match(
2887                     r'^[0-9A-Za-z_-]{11}$', video_id):
2888                 return self.url_result(video_id, YoutubeIE.ie_key())
2889         return self.url_result(base_url)
2890
2891
2892 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2893     IE_DESC = 'YouTube.com user/channel playlists'
2894     _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2895     IE_NAME = 'youtube:playlists'
2896
2897     _TESTS = [{
2898         'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2899         'playlist_mincount': 4,
2900         'info_dict': {
2901             'id': 'ThirstForScience',
2902             'title': 'Thirst for Science',
2903         },
2904     }, {
2905         # with "Load more" button
2906         'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2907         'playlist_mincount': 70,
2908         'info_dict': {
2909             'id': 'igorkle1',
2910             'title': 'Игорь Клейнер',
2911         },
2912     }, {
2913         'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2914         'playlist_mincount': 17,
2915         'info_dict': {
2916             'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2917             'title': 'Chem Player',
2918         },
2919     }]
2920
2921
2922 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2923     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2924
2925
2926 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2927     IE_DESC = 'YouTube.com searches'
2928     # there doesn't appear to be a real limit, for example if you search for
2929     # 'python' you get more than 8.000.000 results
2930     _MAX_RESULTS = float('inf')
2931     IE_NAME = 'youtube:search'
2932     _SEARCH_KEY = 'ytsearch'
2933     _EXTRA_QUERY_ARGS = {}
2934     _TESTS = []
2935
2936     def _get_n_results(self, query, n):
2937         """Get a specified number of results for a query"""
2938
2939         videos = []
2940         limit = n
2941
2942         url_query = {
2943             'search_query': query.encode('utf-8'),
2944         }
2945         url_query.update(self._EXTRA_QUERY_ARGS)
2946         result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2947
2948         for pagenum in itertools.count(1):
2949             data = self._download_json(
2950                 result_url, video_id='query "%s"' % query,
2951                 note='Downloading page %s' % pagenum,
2952                 errnote='Unable to download API page',
2953                 query={'spf': 'navigate'})
2954             html_content = data[1]['body']['content']
2955
2956             if 'class="search-message' in html_content:
2957                 raise ExtractorError(
2958                     '[youtube] No video results', expected=True)
2959
2960             new_videos = list(self._process_page(html_content))
2961             videos += new_videos
2962             if not new_videos or len(videos) > limit:
2963                 break
2964             next_link = self._html_search_regex(
2965                 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2966                 html_content, 'next link', default=None)
2967             if next_link is None:
2968                 break
2969             result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2970
2971         if len(videos) > n:
2972             videos = videos[:n]
2973         return self.playlist_result(videos, query)
2974
2975
2976 class YoutubeSearchDateIE(YoutubeSearchIE):
2977     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2978     _SEARCH_KEY = 'ytsearchdate'
2979     IE_DESC = 'YouTube.com searches, newest videos first'
2980     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2981
2982
2983 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2984     IE_DESC = 'YouTube.com search URLs'
2985     IE_NAME = 'youtube:search_url'
2986     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2987     _TESTS = [{
2988         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2989         'playlist_mincount': 5,
2990         'info_dict': {
2991             'title': 'youtube-dl test video',
2992         }
2993     }, {
2994         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2995         'only_matching': True,
2996     }]
2997
2998     def _real_extract(self, url):
2999         mobj = re.match(self._VALID_URL, url)
3000         query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3001         webpage = self._download_webpage(url, query)
3002         return self.playlist_result(self._process_page(webpage), playlist_title=query)
3003
3004
3005 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3006     IE_DESC = 'YouTube.com (multi-season) shows'
3007     _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3008     IE_NAME = 'youtube:show'
3009     _TESTS = [{
3010         'url': 'https://www.youtube.com/show/airdisasters',
3011         'playlist_mincount': 5,
3012         'info_dict': {
3013             'id': 'airdisasters',
3014             'title': 'Air Disasters',
3015         }
3016     }]
3017
3018     def _real_extract(self, url):
3019         playlist_id = self._match_id(url)
3020         return super(YoutubeShowIE, self)._real_extract(
3021             'https://www.youtube.com/show/%s/playlists' % playlist_id)
3022
3023
3024 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3025     """
3026     Base class for feed extractors
3027     Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3028     """
3029     _LOGIN_REQUIRED = True
3030
3031     @property
3032     def IE_NAME(self):
3033         return 'youtube:%s' % self._FEED_NAME
3034
3035     def _real_initialize(self):
3036         self._login()
3037
3038     def _entries(self, page):
3039         # The extraction process is the same as for playlists, but the regex
3040         # for the video ids doesn't contain an index
3041         ids = []
3042         more_widget_html = content_html = page
3043         for page_num in itertools.count(1):
3044             matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3045
3046             # 'recommended' feed has infinite 'load more' and each new portion spins
3047             # the same videos in (sometimes) slightly different order, so we'll check
3048             # for unicity and break when portion has no new videos
3049             new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3050             if not new_ids:
3051                 break
3052
3053             ids.extend(new_ids)
3054
3055             for entry in self._ids_to_results(new_ids):
3056                 yield entry
3057
3058             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3059             if not mobj:
3060                 break
3061
3062             more = self._download_json(
3063                 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3064                 'Downloading page #%s' % page_num,
3065                 transform_source=uppercase_escape)
3066             content_html = more['content_html']
3067             more_widget_html = more['load_more_widget_html']
3068
3069     def _real_extract(self, url):
3070         page = self._download_webpage(
3071             'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3072             self._PLAYLIST_TITLE)
3073         return self.playlist_result(
3074             self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3075
3076
3077 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3078     IE_NAME = 'youtube:watchlater'
3079     IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3080     _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3081
3082     _TESTS = [{
3083         'url': 'https://www.youtube.com/playlist?list=WL',
3084         'only_matching': True,
3085     }, {
3086         'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3087         'only_matching': True,
3088     }]
3089
3090     def _real_extract(self, url):
3091         _, video = self._check_download_just_video(url, 'WL')
3092         if video:
3093             return video
3094         _, playlist = self._extract_playlist('WL')
3095         return playlist
3096
3097
3098 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3099     IE_NAME = 'youtube:favorites'
3100     IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3101     _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3102     _LOGIN_REQUIRED = True
3103
3104     def _real_extract(self, url):
3105         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3106         playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3107         return self.url_result(playlist_id, 'YoutubePlaylist')
3108
3109
3110 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3111     IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3112     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3113     _FEED_NAME = 'recommended'
3114     _PLAYLIST_TITLE = 'Youtube Recommended videos'
3115
3116
3117 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3118     IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3119     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3120     _FEED_NAME = 'subscriptions'
3121     _PLAYLIST_TITLE = 'Youtube Subscriptions'
3122
3123
3124 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3125     IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3126     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3127     _FEED_NAME = 'history'
3128     _PLAYLIST_TITLE = 'Youtube History'
3129
3130
3131 class YoutubeTruncatedURLIE(InfoExtractor):
3132     IE_NAME = 'youtube:truncated_url'
3133     IE_DESC = False  # Do not list
3134     _VALID_URL = r'''(?x)
3135         (?:https?://)?
3136         (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3137         (?:watch\?(?:
3138             feature=[a-z_]+|
3139             annotation_id=annotation_[^&]+|
3140             x-yt-cl=[0-9]+|
3141             hl=[^&]*|
3142             t=[0-9]+
3143         )?
3144         |
3145             attribution_link\?a=[^&]+
3146         )
3147         $
3148     '''
3149
3150     _TESTS = [{
3151         'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3152         'only_matching': True,
3153     }, {
3154         'url': 'https://www.youtube.com/watch?',
3155         'only_matching': True,
3156     }, {
3157         'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3158         'only_matching': True,
3159     }, {
3160         'url': 'https://www.youtube.com/watch?feature=foo',
3161         'only_matching': True,
3162     }, {
3163         'url': 'https://www.youtube.com/watch?hl=en-GB',
3164         'only_matching': True,
3165     }, {
3166         'url': 'https://www.youtube.com/watch?t=2372',
3167         'only_matching': True,
3168     }]
3169
3170     def _real_extract(self, url):
3171         raise ExtractorError(
3172             'Did you forget to quote the URL? Remember that & is a meta '
3173             'character in most shells, so you want to put the URL in quotes, '
3174             'like  youtube-dl '
3175             '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3176             ' or simply  youtube-dl BaW_jenozKc  .',
3177             expected=True)
3178
3179
3180 class YoutubeTruncatedIDIE(InfoExtractor):
3181     IE_NAME = 'youtube:truncated_id'
3182     IE_DESC = False  # Do not list
3183     _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3184
3185     _TESTS = [{
3186         'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3187         'only_matching': True,
3188     }]
3189
3190     def _real_extract(self, url):
3191         video_id = self._match_id(url)
3192         raise ExtractorError(
3193             'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3194             expected=True)