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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from .subtitles import SubtitlesInfoExtractor
  10 from ..utils import (
  11     clean_html,
  12     compat_HTTPError,
  13     compat_urllib_parse,
  14     compat_urllib_request,
  15     compat_urlparse,
  16     ExtractorError,
  17     get_element_by_attribute,
  18     InAdvancePagedList,
  19     int_or_none,
  20     RegexNotFoundError,
  21     std_headers,
  22     unsmuggle_url,
  23     urlencode_postdata,
  24 )
  25
  26
  27 class VimeoBaseInfoExtractor(InfoExtractor):
  28     _NETRC_MACHINE = 'vimeo'
  29     _LOGIN_REQUIRED = False
  30
  31     def _login(self):
  32         (username, password) = self._get_login_info()
  33         if username is None:
  34             if self._LOGIN_REQUIRED:
  35                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  36             return
  37         self.report_login()
  38         login_url = 'https://vimeo.com/log_in'
  39         webpage = self._download_webpage(login_url, None, False)
  40         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
  41         data = urlencode_postdata({
  42             'email': username,
  43             'password': password,
  44             'action': 'login',
  45             'service': 'vimeo',
  46             'token': token,
  47         })
  48         login_request = compat_urllib_request.Request(login_url, data)
  49         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  50         login_request.add_header('Cookie', 'xsrft=%s' % token)
  51         self._download_webpage(login_request, None, False, 'Wrong login info')
  52
  53
  54 class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
  55     """Information extractor for vimeo.com."""
  56
  57     # _VALID_URL matches Vimeo URLs
  58     _VALID_URL = r'''(?x)
  59         (?P<proto>(?:https?:)?//)?
  60         (?:(?:www|(?P<player>player))\.)?
  61         vimeo(?P<pro>pro)?\.com/
  62         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  63         (?:.*?/)?
  64         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  65         (?:videos?/)?
  66         (?P<id>[0-9]+)
  67         /?(?:[?&].*)?(?:[#].*)?$'''
  68     IE_NAME = 'vimeo'
  69     _TESTS = [
  70         {
  71             'url': 'http://vimeo.com/56015672#at=0',
  72             'md5': '8879b6cc097e987f02484baf890129e5',
  73             'info_dict': {
  74                 'id': '56015672',
  75                 'ext': 'mp4',
  76                 "upload_date": "20121220",
  77                 "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  78                 "uploader_id": "user7108434",
  79                 "uploader": "Filippo Valsorda",
  80                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  81                 "duration": 10,
  82             },
  83         },
  84         {
  85             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  86             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  87             'note': 'Vimeo Pro video (#1197)',
  88             'info_dict': {
  89                 'id': '68093876',
  90                 'ext': 'mp4',
  91                 'uploader_id': 'openstreetmapus',
  92                 'uploader': 'OpenStreetMap US',
  93                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
  94                 'description': 'md5:380943ec71b89736ff4bf27183233d09',
  95                 'duration': 1595,
  96             },
  97         },
  98         {
  99             'url': 'http://player.vimeo.com/video/54469442',
 100             'md5': '619b811a4417aa4abe78dc653becf511',
 101             'note': 'Videos that embed the url in the player page',
 102             'info_dict': {
 103                 'id': '54469442',
 104                 'ext': 'mp4',
 105                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 106                 'uploader': 'The BLN & Business of Software',
 107                 'uploader_id': 'theblnbusinessofsoftware',
 108                 'duration': 3610,
 109                 'description': None,
 110             },
 111         },
 112         {
 113             'url': 'http://vimeo.com/68375962',
 114             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 115             'note': 'Video protected with password',
 116             'info_dict': {
 117                 'id': '68375962',
 118                 'ext': 'mp4',
 119                 'title': 'youtube-dl password protected test video',
 120                 'upload_date': '20130614',
 121                 'uploader_id': 'user18948128',
 122                 'uploader': 'Jaime Marquínez Ferrándiz',
 123                 'duration': 10,
 124                 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.',
 125             },
 126             'params': {
 127                 'videopassword': 'youtube-dl',
 128             },
 129         },
 130         {
 131             'url': 'http://vimeo.com/channels/keypeele/75629013',
 132             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 133             'note': 'Video is freely available via original URL '
 134                     'and protected with password when accessed via http://vimeo.com/75629013',
 135             'info_dict': {
 136                 'id': '75629013',
 137                 'ext': 'mp4',
 138                 'title': 'Key & Peele: Terrorist Interrogation',
 139                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 140                 'uploader_id': 'atencio',
 141                 'uploader': 'Peter Atencio',
 142                 'duration': 187,
 143             },
 144         },
 145         {
 146             'url': 'http://vimeo.com/76979871',
 147             'md5': '3363dd6ffebe3784d56f4132317fd446',
 148             'note': 'Video with subtitles',
 149             'info_dict': {
 150                 'id': '76979871',
 151                 'ext': 'mp4',
 152                 'title': 'The New Vimeo Player (You Know, For Videos)',
 153                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 154                 'upload_date': '20131015',
 155                 'uploader_id': 'staff',
 156                 'uploader': 'Vimeo Staff',
 157                 'duration': 62,
 158             }
 159         },
 160     ]
 161
 162     def _verify_video_password(self, url, video_id, webpage):
 163         password = self._downloader.params.get('videopassword', None)
 164         if password is None:
 165             raise ExtractorError('This video is protected by a password, use the --video-password option')
 166         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 167         data = compat_urllib_parse.urlencode({
 168             'password': password,
 169             'token': token,
 170         })
 171         # I didn't manage to use the password with https
 172         if url.startswith('https'):
 173             pass_url = url.replace('https', 'http')
 174         else:
 175             pass_url = url
 176         password_request = compat_urllib_request.Request(pass_url + '/password', data)
 177         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 178         password_request.add_header('Cookie', 'xsrft=%s' % token)
 179         self._download_webpage(password_request, video_id,
 180                                'Verifying the password',
 181                                'Wrong password')
 182
 183     def _verify_player_video_password(self, url, video_id):
 184         password = self._downloader.params.get('videopassword', None)
 185         if password is None:
 186             raise ExtractorError('This video is protected by a password, use the --video-password option')
 187         data = compat_urllib_parse.urlencode({'password': password})
 188         pass_url = url + '/check-password'
 189         password_request = compat_urllib_request.Request(pass_url, data)
 190         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 191         return self._download_json(
 192             password_request, video_id,
 193             'Verifying the password',
 194             'Wrong password')
 195
 196     def _real_initialize(self):
 197         self._login()
 198
 199     def _real_extract(self, url):
 200         url, data = unsmuggle_url(url)
 201         headers = std_headers
 202         if data is not None:
 203             headers = headers.copy()
 204             headers.update(data)
 205         if 'Referer' not in headers:
 206             headers['Referer'] = url
 207
 208         # Extract ID from URL
 209         mobj = re.match(self._VALID_URL, url)
 210         video_id = mobj.group('id')
 211         orig_url = url
 212         if mobj.group('pro') or mobj.group('player'):
 213             url = 'http://player.vimeo.com/video/' + video_id
 214
 215         # Retrieve video webpage to extract further information
 216         request = compat_urllib_request.Request(url, None, headers)
 217         try:
 218             webpage = self._download_webpage(request, video_id)
 219         except ExtractorError as ee:
 220             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 221                 errmsg = ee.cause.read()
 222                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 223                     raise ExtractorError(
 224                         'Cannot download embed-only video without embedding '
 225                         'URL. Please call youtube-dl with the URL of the page '
 226                         'that embeds this video.',
 227                         expected=True)
 228             raise
 229
 230         # Now we begin extracting as much information as we can from what we
 231         # retrieved. First we extract the information common to all extractors,
 232         # and latter we extract those that are Vimeo specific.
 233         self.report_extraction(video_id)
 234
 235         # Extract the config JSON
 236         try:
 237             try:
 238                 config_url = self._html_search_regex(
 239                     r' data-config-url="(.+?)"', webpage, 'config URL')
 240                 config_json = self._download_webpage(config_url, video_id)
 241                 config = json.loads(config_json)
 242             except RegexNotFoundError:
 243                 # For pro videos or player.vimeo.com urls
 244                 # We try to find out to which variable is assigned the config dic
 245                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 246                 if m_variable_name is not None:
 247                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 248                 else:
 249                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 250                 config = self._search_regex(config_re, webpage, 'info section',
 251                     flags=re.DOTALL)
 252                 config = json.loads(config)
 253         except Exception as e:
 254             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 255                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 256
 257             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 258                 self._verify_video_password(url, video_id, webpage)
 259                 return self._real_extract(url)
 260             else:
 261                 raise ExtractorError('Unable to extract info section',
 262                                      cause=e)
 263         else:
 264             if config.get('view') == 4:
 265                 config = self._verify_player_video_password(url, video_id)
 266
 267         # Extract title
 268         video_title = config["video"]["title"]
 269
 270         # Extract uploader and uploader_id
 271         video_uploader = config["video"]["owner"]["name"]
 272         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 273
 274         # Extract video thumbnail
 275         video_thumbnail = config["video"].get("thumbnail")
 276         if video_thumbnail is None:
 277             video_thumbs = config["video"].get("thumbs")
 278             if video_thumbs and isinstance(video_thumbs, dict):
 279                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 280
 281         # Extract video description
 282
 283         video_description = self._html_search_regex(
 284             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 285             webpage, 'description', default=None)
 286         if not video_description:
 287             video_description = self._html_search_meta(
 288                 'description', webpage, default=None)
 289         if not video_description and mobj.group('pro'):
 290             orig_webpage = self._download_webpage(
 291                 orig_url, video_id,
 292                 note='Downloading webpage for description',
 293                 fatal=False)
 294             if orig_webpage:
 295                 video_description = self._html_search_meta(
 296                     'description', orig_webpage, default=None)
 297         if not video_description and not mobj.group('player'):
 298             self._downloader.report_warning('Cannot find video description')
 299
 300         # Extract video duration
 301         video_duration = int_or_none(config["video"].get("duration"))
 302
 303         # Extract upload date
 304         video_upload_date = None
 305         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 306         if mobj is not None:
 307             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 308
 309         try:
 310             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 311             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 312             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 313         except RegexNotFoundError:
 314             # This info is only available in vimeo.com/{id} urls
 315             view_count = None
 316             like_count = None
 317             comment_count = None
 318
 319         # Vimeo specific: extract request signature and timestamp
 320         sig = config['request']['signature']
 321         timestamp = config['request']['timestamp']
 322
 323         # Vimeo specific: extract video codec and quality information
 324         # First consider quality, then codecs, then take everything
 325         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 326         files = {'hd': [], 'sd': [], 'other': []}
 327         config_files = config["video"].get("files") or config["request"].get("files")
 328         for codec_name, codec_extension in codecs:
 329             for quality in config_files.get(codec_name, []):
 330                 format_id = '-'.join((codec_name, quality)).lower()
 331                 key = quality if quality in files else 'other'
 332                 video_url = None
 333                 if isinstance(config_files[codec_name], dict):
 334                     file_info = config_files[codec_name][quality]
 335                     video_url = file_info.get('url')
 336                 else:
 337                     file_info = {}
 338                 if video_url is None:
 339                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 340                         % (video_id, sig, timestamp, quality, codec_name.upper())
 341
 342                 files[key].append({
 343                     'ext': codec_extension,
 344                     'url': video_url,
 345                     'format_id': format_id,
 346                     'width': file_info.get('width'),
 347                     'height': file_info.get('height'),
 348                 })
 349         formats = []
 350         for key in ('other', 'sd', 'hd'):
 351             formats += files[key]
 352         if len(formats) == 0:
 353             raise ExtractorError('No known codec found')
 354
 355         subtitles = {}
 356         text_tracks = config['request'].get('text_tracks')
 357         if text_tracks:
 358             for tt in text_tracks:
 359                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
 360
 361         video_subtitles = self.extract_subtitles(video_id, subtitles)
 362         if self._downloader.params.get('listsubtitles', False):
 363             self._list_available_subtitles(video_id, subtitles)
 364             return
 365
 366         return {
 367             'id': video_id,
 368             'uploader': video_uploader,
 369             'uploader_id': video_uploader_id,
 370             'upload_date': video_upload_date,
 371             'title': video_title,
 372             'thumbnail': video_thumbnail,
 373             'description': video_description,
 374             'duration': video_duration,
 375             'formats': formats,
 376             'webpage_url': url,
 377             'view_count': view_count,
 378             'like_count': like_count,
 379             'comment_count': comment_count,
 380             'subtitles': video_subtitles,
 381         }
 382
 383
 384 class VimeoChannelIE(InfoExtractor):
 385     IE_NAME = 'vimeo:channel'
 386     _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 387     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 388     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 389     _TESTS = [{
 390         'url': 'http://vimeo.com/channels/tributes',
 391         'info_dict': {
 392             'title': 'Vimeo Tributes',
 393         },
 394         'playlist_mincount': 25,
 395     }]
 396
 397     def _page_url(self, base_url, pagenum):
 398         return '%s/videos/page:%d/' % (base_url, pagenum)
 399
 400     def _extract_list_title(self, webpage):
 401         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 402
 403     def _extract_videos(self, list_id, base_url):
 404         video_ids = []
 405         for pagenum in itertools.count(1):
 406             webpage = self._download_webpage(
 407                 self._page_url(base_url, pagenum), list_id,
 408                 'Downloading page %s' % pagenum)
 409             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 410             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 411                 break
 412
 413         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 414                    for video_id in video_ids]
 415         return {'_type': 'playlist',
 416                 'id': list_id,
 417                 'title': self._extract_list_title(webpage),
 418                 'entries': entries,
 419                 }
 420
 421     def _real_extract(self, url):
 422         mobj = re.match(self._VALID_URL, url)
 423         channel_id = mobj.group('id')
 424         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 425
 426
 427 class VimeoUserIE(VimeoChannelIE):
 428     IE_NAME = 'vimeo:user'
 429     _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 430     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 431     _TESTS = [{
 432         'url': 'http://vimeo.com/nkistudio/videos',
 433         'info_dict': {
 434             'title': 'Nki',
 435         },
 436         'playlist_mincount': 66,
 437     }]
 438
 439     def _real_extract(self, url):
 440         mobj = re.match(self._VALID_URL, url)
 441         name = mobj.group('name')
 442         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 443
 444
 445 class VimeoAlbumIE(VimeoChannelIE):
 446     IE_NAME = 'vimeo:album'
 447     _VALID_URL = r'https?://vimeo\.com/album/(?P<id>\d+)'
 448     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 449     _TESTS = [{
 450         'url': 'http://vimeo.com/album/2632481',
 451         'info_dict': {
 452             'title': 'Staff Favorites: November 2013',
 453         },
 454         'playlist_mincount': 13,
 455     }]
 456
 457     def _page_url(self, base_url, pagenum):
 458         return '%s/page:%d/' % (base_url, pagenum)
 459
 460     def _real_extract(self, url):
 461         mobj = re.match(self._VALID_URL, url)
 462         album_id = mobj.group('id')
 463         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 464
 465
 466 class VimeoGroupsIE(VimeoAlbumIE):
 467     IE_NAME = 'vimeo:group'
 468     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 469     _TESTS = [{
 470         'url': 'http://vimeo.com/groups/rolexawards',
 471         'info_dict': {
 472             'title': 'Rolex Awards for Enterprise',
 473         },
 474         'playlist_mincount': 73,
 475     }]
 476
 477     def _extract_list_title(self, webpage):
 478         return self._og_search_title(webpage)
 479
 480     def _real_extract(self, url):
 481         mobj = re.match(self._VALID_URL, url)
 482         name = mobj.group('name')
 483         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 484
 485
 486 class VimeoReviewIE(InfoExtractor):
 487     IE_NAME = 'vimeo:review'
 488     IE_DESC = 'Review pages on vimeo'
 489     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 490     _TESTS = [{
 491         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 492         'file': '75524534.mp4',
 493         'md5': 'c507a72f780cacc12b2248bb4006d253',
 494         'info_dict': {
 495             'title': "DICK HARDWICK 'Comedian'",
 496             'uploader': 'Richard Hardwick',
 497         }
 498     }, {
 499         'note': 'video player needs Referer',
 500         'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
 501         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 502         'info_dict': {
 503             'id': '91613211',
 504             'ext': 'mp4',
 505             'title': 'Death by dogma versus assembling agile - Sander Hoogendoorn',
 506             'uploader': 'DevWeek Events',
 507             'duration': 2773,
 508             'thumbnail': 're:^https?://.*\.jpg$',
 509         }
 510     }]
 511
 512     def _real_extract(self, url):
 513         mobj = re.match(self._VALID_URL, url)
 514         video_id = mobj.group('id')
 515         player_url = 'https://player.vimeo.com/player/' + video_id
 516         return self.url_result(player_url, 'Vimeo', video_id)
 517
 518
 519 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 520     IE_NAME = 'vimeo:watchlater'
 521     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 522     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 523     _LOGIN_REQUIRED = True
 524     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 525     _TESTS = [{
 526         'url': 'http://vimeo.com/home/watchlater',
 527         'only_matching': True,
 528     }]
 529
 530     def _real_initialize(self):
 531         self._login()
 532
 533     def _page_url(self, base_url, pagenum):
 534         url = '%s/page:%d/' % (base_url, pagenum)
 535         request = compat_urllib_request.Request(url)
 536         # Set the header to get a partial html page with the ids,
 537         # the normal page doesn't contain them.
 538         request.add_header('X-Requested-With', 'XMLHttpRequest')
 539         return request
 540
 541     def _real_extract(self, url):
 542         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
 543
 544
 545 class VimeoLikesIE(InfoExtractor):
 546     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 547     IE_NAME = 'vimeo:likes'
 548     IE_DESC = 'Vimeo user likes'
 549     _TEST = {
 550         'url': 'https://vimeo.com/user755559/likes/',
 551         'playlist_mincount': 293,
 552         "info_dict": {
 553             "description": "See all the videos urza likes",
 554             "title": 'Videos urza likes',
 555         },
 556     }
 557
 558     def _real_extract(self, url):
 559         user_id = self._match_id(url)
 560         webpage = self._download_webpage(url, user_id)
 561         page_count = self._int(
 562             self._search_regex(
 563                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 564                     .*?</a></li>\s*<li\s+class="pagination_next">
 565                 ''', webpage, 'page count'),
 566             'page count', fatal=True)
 567         PAGE_SIZE = 12
 568         title = self._html_search_regex(
 569             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 570         description = self._html_search_meta('description', webpage)
 571
 572         def _get_page(idx):
 573             page_url = '%s//vimeo.com/user%s/likes/page:%d/sort:date' % (
 574                 self.http_scheme(), user_id, idx + 1)
 575             webpage = self._download_webpage(
 576                 page_url, user_id,
 577                 note='Downloading page %d/%d' % (idx + 1, page_count))
 578             video_list = self._search_regex(
 579                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 580                 webpage, 'video content')
 581             paths = re.findall(
 582                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 583             for path in paths:
 584                 yield {
 585                     '_type': 'url',
 586                     'url': compat_urlparse.urljoin(page_url, path),
 587                 }
 588
 589         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 590
 591         return {
 592             '_type': 'playlist',
 593             'id': 'user%s_likes' % user_id,
 594             'title': title,
 595             'description': description,
 596             'entries': pl,
 597         }