_ 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 ..compat import (
  10     compat_HTTPError,
  11     compat_urllib_request,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     encode_dict,
  16     ExtractorError,
  17     InAdvancePagedList,
  18     int_or_none,
  19     RegexNotFoundError,
  20     smuggle_url,
  21     std_headers,
  22     unified_strdate,
  23     unsmuggle_url,
  24     urlencode_postdata,
  25     unescapeHTML,
  26 )
  27
  28
  29 class VimeoBaseInfoExtractor(InfoExtractor):
  30     _NETRC_MACHINE = 'vimeo'
  31     _LOGIN_REQUIRED = False
  32     _LOGIN_URL = 'https://vimeo.com/log_in'
  33
  34     def _login(self):
  35         (username, password) = self._get_login_info()
  36         if username is None:
  37             if self._LOGIN_REQUIRED:
  38                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  39             return
  40         self.report_login()
  41         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  42         token, vuid = self._extract_xsrft_and_vuid(webpage)
  43         data = urlencode_postdata(encode_dict({
  44             'action': 'login',
  45             'email': username,
  46             'password': password,
  47             'service': 'vimeo',
  48             'token': token,
  49         }))
  50         login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
  51         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  52         login_request.add_header('Referer', self._LOGIN_URL)
  53         self._set_vimeo_cookie('vuid', vuid)
  54         self._download_webpage(login_request, None, False, 'Wrong login info')
  55
  56     def _extract_xsrft_and_vuid(self, webpage):
  57         xsrft = self._search_regex(
  58             r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  59             webpage, 'login token', group='xsrft')
  60         vuid = self._search_regex(
  61             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  62             webpage, 'vuid', group='vuid')
  63         return xsrft, vuid
  64
  65     def _set_vimeo_cookie(self, name, value):
  66         self._set_cookie('vimeo.com', name, value)
  67
  68
  69 class VimeoIE(VimeoBaseInfoExtractor):
  70     """Information extractor for vimeo.com."""
  71
  72     # _VALID_URL matches Vimeo URLs
  73     _VALID_URL = r'''(?x)
  74         https?://
  75         (?:(?:www|(?P<player>player))\.)?
  76         vimeo(?P<pro>pro)?\.com/
  77         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  78         (?:.*?/)?
  79         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  80         (?:videos?/)?
  81         (?P<id>[0-9]+)
  82         /?(?:[?&].*)?(?:[#].*)?$'''
  83     IE_NAME = 'vimeo'
  84     _TESTS = [
  85         {
  86             'url': 'http://vimeo.com/56015672#at=0',
  87             'md5': '8879b6cc097e987f02484baf890129e5',
  88             'info_dict': {
  89                 'id': '56015672',
  90                 'ext': 'mp4',
  91                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  92                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
  93                 'upload_date': '20121220',
  94                 'uploader_id': 'user7108434',
  95                 'uploader': 'Filippo Valsorda',
  96                 'duration': 10,
  97             },
  98         },
  99         {
 100             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 101             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 102             'note': 'Vimeo Pro video (#1197)',
 103             'info_dict': {
 104                 'id': '68093876',
 105                 'ext': 'mp4',
 106                 'uploader_id': 'openstreetmapus',
 107                 'uploader': 'OpenStreetMap US',
 108                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 109                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 110                 'duration': 1595,
 111             },
 112         },
 113         {
 114             'url': 'http://player.vimeo.com/video/54469442',
 115             'md5': '619b811a4417aa4abe78dc653becf511',
 116             'note': 'Videos that embed the url in the player page',
 117             'info_dict': {
 118                 'id': '54469442',
 119                 'ext': 'mp4',
 120                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 121                 'uploader': 'The BLN & Business of Software',
 122                 'uploader_id': 'theblnbusinessofsoftware',
 123                 'duration': 3610,
 124                 'description': None,
 125             },
 126         },
 127         {
 128             'url': 'http://vimeo.com/68375962',
 129             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 130             'note': 'Video protected with password',
 131             'info_dict': {
 132                 'id': '68375962',
 133                 'ext': 'mp4',
 134                 'title': 'youtube-dl password protected test video',
 135                 'upload_date': '20130614',
 136                 'uploader_id': 'user18948128',
 137                 'uploader': 'Jaime Marquínez Ferrándiz',
 138                 'duration': 10,
 139                 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people\u2026',
 140             },
 141             'params': {
 142                 'videopassword': 'youtube-dl',
 143             },
 144         },
 145         {
 146             'url': 'http://vimeo.com/channels/keypeele/75629013',
 147             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 148             'note': 'Video is freely available via original URL '
 149                     'and protected with password when accessed via http://vimeo.com/75629013',
 150             'info_dict': {
 151                 'id': '75629013',
 152                 'ext': 'mp4',
 153                 'title': 'Key & Peele: Terrorist Interrogation',
 154                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 155                 'uploader_id': 'atencio',
 156                 'uploader': 'Peter Atencio',
 157                 'upload_date': '20130927',
 158                 'duration': 187,
 159             },
 160         },
 161         {
 162             'url': 'http://vimeo.com/76979871',
 163             'note': 'Video with subtitles',
 164             'info_dict': {
 165                 'id': '76979871',
 166                 'ext': 'mp4',
 167                 'title': 'The New Vimeo Player (You Know, For Videos)',
 168                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 169                 'upload_date': '20131015',
 170                 'uploader_id': 'staff',
 171                 'uploader': 'Vimeo Staff',
 172                 'duration': 62,
 173             }
 174         },
 175         {
 176             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 177             'url': 'https://player.vimeo.com/video/98044508',
 178             'note': 'The js code contains assignments to the same variable as the config',
 179             'info_dict': {
 180                 'id': '98044508',
 181                 'ext': 'mp4',
 182                 'title': 'Pier Solar OUYA Official Trailer',
 183                 'uploader': 'Tulio Gonçalves',
 184                 'uploader_id': 'user28849593',
 185             },
 186         },
 187         {
 188             'url': 'https://vimeo.com/109815029',
 189             'note': 'Video not completely processed, "failed" seed status',
 190             'only_matching': True,
 191         },
 192     ]
 193
 194     @staticmethod
 195     def _extract_vimeo_url(url, webpage):
 196         # Look for embedded (iframe) Vimeo player
 197         mobj = re.search(
 198             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 199         if mobj:
 200             player_url = unescapeHTML(mobj.group('url'))
 201             surl = smuggle_url(player_url, {'Referer': url})
 202             return surl
 203         # Look for embedded (swf embed) Vimeo player
 204         mobj = re.search(
 205             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 206         if mobj:
 207             return mobj.group(1)
 208
 209     def _verify_video_password(self, url, video_id, webpage):
 210         password = self._downloader.params.get('videopassword', None)
 211         if password is None:
 212             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
 213         token, vuid = self._extract_xsrft_and_vuid(webpage)
 214         data = urlencode_postdata(encode_dict({
 215             'password': password,
 216             'token': token,
 217         }))
 218         if url.startswith('http://'):
 219             # vimeo only supports https now, but the user can give an http url
 220             url = url.replace('http://', 'https://')
 221         password_request = compat_urllib_request.Request(url + '/password', data)
 222         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 223         password_request.add_header('Referer', url)
 224         self._set_vimeo_cookie('vuid', vuid)
 225         return self._download_webpage(
 226             password_request, video_id,
 227             'Verifying the password', 'Wrong password')
 228
 229     def _verify_player_video_password(self, url, video_id):
 230         password = self._downloader.params.get('videopassword', None)
 231         if password is None:
 232             raise ExtractorError('This video is protected by a password, use the --video-password option')
 233         data = urlencode_postdata(encode_dict({'password': password}))
 234         pass_url = url + '/check-password'
 235         password_request = compat_urllib_request.Request(pass_url, data)
 236         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 237         return self._download_json(
 238             password_request, video_id,
 239             'Verifying the password',
 240             'Wrong password')
 241
 242     def _real_initialize(self):
 243         self._login()
 244
 245     def _real_extract(self, url):
 246         url, data = unsmuggle_url(url)
 247         headers = std_headers
 248         if data is not None:
 249             headers = headers.copy()
 250             headers.update(data)
 251         if 'Referer' not in headers:
 252             headers['Referer'] = url
 253
 254         # Extract ID from URL
 255         mobj = re.match(self._VALID_URL, url)
 256         video_id = mobj.group('id')
 257         orig_url = url
 258         if mobj.group('pro') or mobj.group('player'):
 259             url = 'https://player.vimeo.com/video/' + video_id
 260         else:
 261             url = 'https://vimeo.com/' + video_id
 262
 263         # Retrieve video webpage to extract further information
 264         request = compat_urllib_request.Request(url, None, headers)
 265         try:
 266             webpage = self._download_webpage(request, video_id)
 267         except ExtractorError as ee:
 268             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 269                 errmsg = ee.cause.read()
 270                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 271                     raise ExtractorError(
 272                         'Cannot download embed-only video without embedding '
 273                         'URL. Please call youtube-dl with the URL of the page '
 274                         'that embeds this video.',
 275                         expected=True)
 276             raise
 277
 278         # Now we begin extracting as much information as we can from what we
 279         # retrieved. First we extract the information common to all extractors,
 280         # and latter we extract those that are Vimeo specific.
 281         self.report_extraction(video_id)
 282
 283         vimeo_config = self._search_regex(
 284             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 285             'vimeo config', default=None)
 286         if vimeo_config:
 287             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 288             if seed_status.get('state') == 'failed':
 289                 raise ExtractorError(
 290                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 291                     expected=True)
 292
 293         # Extract the config JSON
 294         try:
 295             try:
 296                 config_url = self._html_search_regex(
 297                     r' data-config-url="(.+?)"', webpage,
 298                     'config URL', default=None)
 299                 if not config_url:
 300                     # Sometimes new react-based page is served instead of old one that require
 301                     # different config URL extraction approach (see
 302                     # https://github.com/rg3/youtube-dl/pull/7209)
 303                     vimeo_clip_page_config = self._search_regex(
 304                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 305                         'vimeo clip page config')
 306                     config_url = self._parse_json(
 307                         vimeo_clip_page_config, video_id)['player']['config_url']
 308                 config_json = self._download_webpage(config_url, video_id)
 309                 config = json.loads(config_json)
 310             except RegexNotFoundError:
 311                 # For pro videos or player.vimeo.com urls
 312                 # We try to find out to which variable is assigned the config dic
 313                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 314                 if m_variable_name is not None:
 315                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 316                 else:
 317                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 318                 config = self._search_regex(config_re, webpage, 'info section',
 319                                             flags=re.DOTALL)
 320                 config = json.loads(config)
 321         except Exception as e:
 322             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 323                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 324
 325             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 326                 if data and '_video_password_verified' in data:
 327                     raise ExtractorError('video password verification failed!')
 328                 self._verify_video_password(url, video_id, webpage)
 329                 return self._real_extract(
 330                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 331             else:
 332                 raise ExtractorError('Unable to extract info section',
 333                                      cause=e)
 334         else:
 335             if config.get('view') == 4:
 336                 config = self._verify_player_video_password(url, video_id)
 337
 338         # Extract title
 339         video_title = config["video"]["title"]
 340
 341         # Extract uploader and uploader_id
 342         video_uploader = config["video"]["owner"]["name"]
 343         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 344
 345         # Extract video thumbnail
 346         video_thumbnail = config["video"].get("thumbnail")
 347         if video_thumbnail is None:
 348             video_thumbs = config["video"].get("thumbs")
 349             if video_thumbs and isinstance(video_thumbs, dict):
 350                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 351
 352         # Extract video description
 353
 354         video_description = self._html_search_regex(
 355             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 356             webpage, 'description', default=None)
 357         if not video_description:
 358             video_description = self._html_search_meta(
 359                 'description', webpage, default=None)
 360         if not video_description and mobj.group('pro'):
 361             orig_webpage = self._download_webpage(
 362                 orig_url, video_id,
 363                 note='Downloading webpage for description',
 364                 fatal=False)
 365             if orig_webpage:
 366                 video_description = self._html_search_meta(
 367                     'description', orig_webpage, default=None)
 368         if not video_description and not mobj.group('player'):
 369             self._downloader.report_warning('Cannot find video description')
 370
 371         # Extract video duration
 372         video_duration = int_or_none(config["video"].get("duration"))
 373
 374         # Extract upload date
 375         video_upload_date = None
 376         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 377         if mobj is not None:
 378             video_upload_date = unified_strdate(mobj.group(1))
 379
 380         try:
 381             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 382             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 383             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 384         except RegexNotFoundError:
 385             # This info is only available in vimeo.com/{id} urls
 386             view_count = None
 387             like_count = None
 388             comment_count = None
 389
 390         # Vimeo specific: extract request signature and timestamp
 391         sig = config['request']['signature']
 392         timestamp = config['request']['timestamp']
 393
 394         # Vimeo specific: extract video codec and quality information
 395         # First consider quality, then codecs, then take everything
 396         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 397         files = {'hd': [], 'sd': [], 'other': []}
 398         config_files = config["video"].get("files") or config["request"].get("files")
 399         for codec_name, codec_extension in codecs:
 400             for quality in config_files.get(codec_name, []):
 401                 format_id = '-'.join((codec_name, quality)).lower()
 402                 key = quality if quality in files else 'other'
 403                 video_url = None
 404                 if isinstance(config_files[codec_name], dict):
 405                     file_info = config_files[codec_name][quality]
 406                     video_url = file_info.get('url')
 407                 else:
 408                     file_info = {}
 409                 if video_url is None:
 410                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 411                         % (video_id, sig, timestamp, quality, codec_name.upper())
 412
 413                 files[key].append({
 414                     'ext': codec_extension,
 415                     'url': video_url,
 416                     'format_id': format_id,
 417                     'width': int_or_none(file_info.get('width')),
 418                     'height': int_or_none(file_info.get('height')),
 419                     'tbr': int_or_none(file_info.get('bitrate')),
 420                 })
 421         formats = []
 422         m3u8_url = config_files.get('hls', {}).get('all')
 423         if m3u8_url:
 424             m3u8_formats = self._extract_m3u8_formats(
 425                 m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
 426             if m3u8_formats:
 427                 formats.extend(m3u8_formats)
 428         for key in ('other', 'sd', 'hd'):
 429             formats += files[key]
 430         self._sort_formats(formats)
 431
 432         subtitles = {}
 433         text_tracks = config['request'].get('text_tracks')
 434         if text_tracks:
 435             for tt in text_tracks:
 436                 subtitles[tt['lang']] = [{
 437                     'ext': 'vtt',
 438                     'url': 'https://vimeo.com' + tt['url'],
 439                 }]
 440
 441         return {
 442             'id': video_id,
 443             'uploader': video_uploader,
 444             'uploader_id': video_uploader_id,
 445             'upload_date': video_upload_date,
 446             'title': video_title,
 447             'thumbnail': video_thumbnail,
 448             'description': video_description,
 449             'duration': video_duration,
 450             'formats': formats,
 451             'webpage_url': url,
 452             'view_count': view_count,
 453             'like_count': like_count,
 454             'comment_count': comment_count,
 455             'subtitles': subtitles,
 456         }
 457
 458
 459 class VimeoChannelIE(VimeoBaseInfoExtractor):
 460     IE_NAME = 'vimeo:channel'
 461     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 462     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 463     _TITLE = None
 464     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 465     _TESTS = [{
 466         'url': 'https://vimeo.com/channels/tributes',
 467         'info_dict': {
 468             'id': 'tributes',
 469             'title': 'Vimeo Tributes',
 470         },
 471         'playlist_mincount': 25,
 472     }]
 473
 474     def _page_url(self, base_url, pagenum):
 475         return '%s/videos/page:%d/' % (base_url, pagenum)
 476
 477     def _extract_list_title(self, webpage):
 478         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 479
 480     def _login_list_password(self, page_url, list_id, webpage):
 481         login_form = self._search_regex(
 482             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 483             webpage, 'login form', default=None)
 484         if not login_form:
 485             return webpage
 486
 487         password = self._downloader.params.get('videopassword', None)
 488         if password is None:
 489             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 490         fields = self._hidden_inputs(login_form)
 491         token, vuid = self._extract_xsrft_and_vuid(webpage)
 492         fields['token'] = token
 493         fields['password'] = password
 494         post = urlencode_postdata(encode_dict(fields))
 495         password_path = self._search_regex(
 496             r'action="([^"]+)"', login_form, 'password URL')
 497         password_url = compat_urlparse.urljoin(page_url, password_path)
 498         password_request = compat_urllib_request.Request(password_url, post)
 499         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 500         self._set_vimeo_cookie('vuid', vuid)
 501         self._set_vimeo_cookie('xsrft', token)
 502
 503         return self._download_webpage(
 504             password_request, list_id,
 505             'Verifying the password', 'Wrong password')
 506
 507     def _extract_videos(self, list_id, base_url):
 508         video_ids = []
 509         for pagenum in itertools.count(1):
 510             page_url = self._page_url(base_url, pagenum)
 511             webpage = self._download_webpage(
 512                 page_url, list_id,
 513                 'Downloading page %s' % pagenum)
 514
 515             if pagenum == 1:
 516                 webpage = self._login_list_password(page_url, list_id, webpage)
 517
 518             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 519             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 520                 break
 521
 522         entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo')
 523                    for video_id in video_ids]
 524         return {'_type': 'playlist',
 525                 'id': list_id,
 526                 'title': self._extract_list_title(webpage),
 527                 'entries': entries,
 528                 }
 529
 530     def _real_extract(self, url):
 531         mobj = re.match(self._VALID_URL, url)
 532         channel_id = mobj.group('id')
 533         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 534
 535
 536 class VimeoUserIE(VimeoChannelIE):
 537     IE_NAME = 'vimeo:user'
 538     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 539     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 540     _TESTS = [{
 541         'url': 'https://vimeo.com/nkistudio/videos',
 542         'info_dict': {
 543             'title': 'Nki',
 544             'id': 'nkistudio',
 545         },
 546         'playlist_mincount': 66,
 547     }]
 548
 549     def _real_extract(self, url):
 550         mobj = re.match(self._VALID_URL, url)
 551         name = mobj.group('name')
 552         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 553
 554
 555 class VimeoAlbumIE(VimeoChannelIE):
 556     IE_NAME = 'vimeo:album'
 557     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
 558     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 559     _TESTS = [{
 560         'url': 'https://vimeo.com/album/2632481',
 561         'info_dict': {
 562             'id': '2632481',
 563             'title': 'Staff Favorites: November 2013',
 564         },
 565         'playlist_mincount': 13,
 566     }, {
 567         'note': 'Password-protected album',
 568         'url': 'https://vimeo.com/album/3253534',
 569         'info_dict': {
 570             'title': 'test',
 571             'id': '3253534',
 572         },
 573         'playlist_count': 1,
 574         'params': {
 575             'videopassword': 'youtube-dl',
 576         }
 577     }]
 578
 579     def _page_url(self, base_url, pagenum):
 580         return '%s/page:%d/' % (base_url, pagenum)
 581
 582     def _real_extract(self, url):
 583         album_id = self._match_id(url)
 584         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 585
 586
 587 class VimeoGroupsIE(VimeoAlbumIE):
 588     IE_NAME = 'vimeo:group'
 589     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)'
 590     _TESTS = [{
 591         'url': 'https://vimeo.com/groups/rolexawards',
 592         'info_dict': {
 593             'id': 'rolexawards',
 594             'title': 'Rolex Awards for Enterprise',
 595         },
 596         'playlist_mincount': 73,
 597     }]
 598
 599     def _extract_list_title(self, webpage):
 600         return self._og_search_title(webpage)
 601
 602     def _real_extract(self, url):
 603         mobj = re.match(self._VALID_URL, url)
 604         name = mobj.group('name')
 605         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 606
 607
 608 class VimeoReviewIE(InfoExtractor):
 609     IE_NAME = 'vimeo:review'
 610     IE_DESC = 'Review pages on vimeo'
 611     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 612     _TESTS = [{
 613         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 614         'md5': 'c507a72f780cacc12b2248bb4006d253',
 615         'info_dict': {
 616             'id': '75524534',
 617             'ext': 'mp4',
 618             'title': "DICK HARDWICK 'Comedian'",
 619             'uploader': 'Richard Hardwick',
 620         }
 621     }, {
 622         'note': 'video player needs Referer',
 623         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 624         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 625         'info_dict': {
 626             'id': '91613211',
 627             'ext': 'mp4',
 628             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 629             'uploader': 'DevWeek Events',
 630             'duration': 2773,
 631             'thumbnail': 're:^https?://.*\.jpg$',
 632         }
 633     }]
 634
 635     def _real_extract(self, url):
 636         mobj = re.match(self._VALID_URL, url)
 637         video_id = mobj.group('id')
 638         player_url = 'https://player.vimeo.com/player/' + video_id
 639         return self.url_result(player_url, 'Vimeo', video_id)
 640
 641
 642 class VimeoWatchLaterIE(VimeoChannelIE):
 643     IE_NAME = 'vimeo:watchlater'
 644     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 645     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 646     _TITLE = 'Watch Later'
 647     _LOGIN_REQUIRED = True
 648     _TESTS = [{
 649         'url': 'https://vimeo.com/watchlater',
 650         'only_matching': True,
 651     }]
 652
 653     def _real_initialize(self):
 654         self._login()
 655
 656     def _page_url(self, base_url, pagenum):
 657         url = '%s/page:%d/' % (base_url, pagenum)
 658         request = compat_urllib_request.Request(url)
 659         # Set the header to get a partial html page with the ids,
 660         # the normal page doesn't contain them.
 661         request.add_header('X-Requested-With', 'XMLHttpRequest')
 662         return request
 663
 664     def _real_extract(self, url):
 665         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 666
 667
 668 class VimeoLikesIE(InfoExtractor):
 669     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 670     IE_NAME = 'vimeo:likes'
 671     IE_DESC = 'Vimeo user likes'
 672     _TEST = {
 673         'url': 'https://vimeo.com/user755559/likes/',
 674         'playlist_mincount': 293,
 675         "info_dict": {
 676             'id': 'user755559_likes',
 677             "description": "See all the videos urza likes",
 678             "title": 'Videos urza likes',
 679         },
 680     }
 681
 682     def _real_extract(self, url):
 683         user_id = self._match_id(url)
 684         webpage = self._download_webpage(url, user_id)
 685         page_count = self._int(
 686             self._search_regex(
 687                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 688                     .*?</a></li>\s*<li\s+class="pagination_next">
 689                 ''', webpage, 'page count'),
 690             'page count', fatal=True)
 691         PAGE_SIZE = 12
 692         title = self._html_search_regex(
 693             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 694         description = self._html_search_meta('description', webpage)
 695
 696         def _get_page(idx):
 697             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 698                 user_id, idx + 1)
 699             webpage = self._download_webpage(
 700                 page_url, user_id,
 701                 note='Downloading page %d/%d' % (idx + 1, page_count))
 702             video_list = self._search_regex(
 703                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 704                 webpage, 'video content')
 705             paths = re.findall(
 706                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 707             for path in paths:
 708                 yield {
 709                     '_type': 'url',
 710                     'url': compat_urlparse.urljoin(page_url, path),
 711                 }
 712
 713         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 714
 715         return {
 716             '_type': 'playlist',
 717             'id': 'user%s_likes' % user_id,
 718             'title': title,
 719             'description': description,
 720             'entries': pl,
 721         }