_ 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_parse,
  12     compat_urllib_request,
  13     compat_urlparse,
  14 )
  15 from ..utils import (
  16     encode_dict,
  17     ExtractorError,
  18     InAdvancePagedList,
  19     int_or_none,
  20     RegexNotFoundError,
  21     smuggle_url,
  22     std_headers,
  23     unified_strdate,
  24     unsmuggle_url,
  25     urlencode_postdata,
  26     unescapeHTML,
  27 )
  28
  29
  30 class VimeoBaseInfoExtractor(InfoExtractor):
  31     _NETRC_MACHINE = 'vimeo'
  32     _LOGIN_REQUIRED = False
  33     _LOGIN_URL = 'https://vimeo.com/log_in'
  34
  35     def _login(self):
  36         (username, password) = self._get_login_info()
  37         if username is None:
  38             if self._LOGIN_REQUIRED:
  39                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  40             return
  41         self.report_login()
  42         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  43         token, vuid = self._extract_xsrft_and_vuid(webpage)
  44         data = urlencode_postdata(encode_dict({
  45             'action': 'login',
  46             'email': username,
  47             'password': password,
  48             'service': 'vimeo',
  49             'token': token,
  50         }))
  51         login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
  52         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  53         login_request.add_header('Cookie', 'vuid=%s' % vuid)
  54         login_request.add_header('Referer', self._LOGIN_URL)
  55         self._download_webpage(login_request, None, False, 'Wrong login info')
  56
  57     def _extract_xsrft_and_vuid(self, webpage):
  58         xsrft = self._search_regex(
  59             r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  60             webpage, 'login token', group='xsrft')
  61         vuid = self._search_regex(
  62             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  63             webpage, 'vuid', group='vuid')
  64         return xsrft, vuid
  65
  66
  67 class VimeoIE(VimeoBaseInfoExtractor):
  68     """Information extractor for vimeo.com."""
  69
  70     # _VALID_URL matches Vimeo URLs
  71     _VALID_URL = r'''(?x)
  72         https?://
  73         (?:(?:www|(?P<player>player))\.)?
  74         vimeo(?P<pro>pro)?\.com/
  75         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  76         (?:.*?/)?
  77         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  78         (?:videos?/)?
  79         (?P<id>[0-9]+)
  80         /?(?:[?&].*)?(?:[#].*)?$'''
  81     IE_NAME = 'vimeo'
  82     _TESTS = [
  83         {
  84             'url': 'http://vimeo.com/56015672#at=0',
  85             'md5': '8879b6cc097e987f02484baf890129e5',
  86             'info_dict': {
  87                 'id': '56015672',
  88                 'ext': 'mp4',
  89                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  90                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
  91                 'upload_date': '20121220',
  92                 'uploader_id': 'user7108434',
  93                 'uploader': 'Filippo Valsorda',
  94                 'duration': 10,
  95             },
  96         },
  97         {
  98             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  99             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 100             'note': 'Vimeo Pro video (#1197)',
 101             'info_dict': {
 102                 'id': '68093876',
 103                 'ext': 'mp4',
 104                 'uploader_id': 'openstreetmapus',
 105                 'uploader': 'OpenStreetMap US',
 106                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 107                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 108                 'duration': 1595,
 109             },
 110         },
 111         {
 112             'url': 'http://player.vimeo.com/video/54469442',
 113             'md5': '619b811a4417aa4abe78dc653becf511',
 114             'note': 'Videos that embed the url in the player page',
 115             'info_dict': {
 116                 'id': '54469442',
 117                 'ext': 'mp4',
 118                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 119                 'uploader': 'The BLN & Business of Software',
 120                 'uploader_id': 'theblnbusinessofsoftware',
 121                 'duration': 3610,
 122                 'description': None,
 123             },
 124         },
 125         {
 126             'url': 'http://vimeo.com/68375962',
 127             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 128             'note': 'Video protected with password',
 129             'info_dict': {
 130                 'id': '68375962',
 131                 'ext': 'mp4',
 132                 'title': 'youtube-dl password protected test video',
 133                 'upload_date': '20130614',
 134                 'uploader_id': 'user18948128',
 135                 'uploader': 'Jaime Marquínez Ferrándiz',
 136                 'duration': 10,
 137                 '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',
 138             },
 139             'params': {
 140                 'videopassword': 'youtube-dl',
 141             },
 142         },
 143         {
 144             'url': 'http://vimeo.com/channels/keypeele/75629013',
 145             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 146             'note': 'Video is freely available via original URL '
 147                     'and protected with password when accessed via http://vimeo.com/75629013',
 148             'info_dict': {
 149                 'id': '75629013',
 150                 'ext': 'mp4',
 151                 'title': 'Key & Peele: Terrorist Interrogation',
 152                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 153                 'uploader_id': 'atencio',
 154                 'uploader': 'Peter Atencio',
 155                 'upload_date': '20130927',
 156                 'duration': 187,
 157             },
 158         },
 159         {
 160             'url': 'http://vimeo.com/76979871',
 161             'note': 'Video with subtitles',
 162             'info_dict': {
 163                 'id': '76979871',
 164                 'ext': 'mp4',
 165                 'title': 'The New Vimeo Player (You Know, For Videos)',
 166                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 167                 'upload_date': '20131015',
 168                 'uploader_id': 'staff',
 169                 'uploader': 'Vimeo Staff',
 170                 'duration': 62,
 171             }
 172         },
 173         {
 174             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 175             'url': 'https://player.vimeo.com/video/98044508',
 176             'note': 'The js code contains assignments to the same variable as the config',
 177             'info_dict': {
 178                 'id': '98044508',
 179                 'ext': 'mp4',
 180                 'title': 'Pier Solar OUYA Official Trailer',
 181                 'uploader': 'Tulio Gonçalves',
 182                 'uploader_id': 'user28849593',
 183             },
 184         },
 185         {
 186             'url': 'https://vimeo.com/109815029',
 187             'note': 'Video not completely processed, "failed" seed status',
 188             'only_matching': True,
 189         },
 190     ]
 191
 192     @staticmethod
 193     def _extract_vimeo_url(url, webpage):
 194         # Look for embedded (iframe) Vimeo player
 195         mobj = re.search(
 196             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 197         if mobj:
 198             player_url = unescapeHTML(mobj.group('url'))
 199             surl = smuggle_url(player_url, {'Referer': url})
 200             return surl
 201         # Look for embedded (swf embed) Vimeo player
 202         mobj = re.search(
 203             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 204         if mobj:
 205             return mobj.group(1)
 206
 207     def _verify_video_password(self, url, video_id, webpage):
 208         password = self._downloader.params.get('videopassword', None)
 209         if password is None:
 210             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
 211         token, vuid = self._extract_xsrft_and_vuid(webpage)
 212         data = urlencode_postdata(encode_dict({
 213             'password': password,
 214             'token': token,
 215         }))
 216         if url.startswith('http://'):
 217             # vimeo only supports https now, but the user can give an http url
 218             url = url.replace('http://', 'https://')
 219         password_request = compat_urllib_request.Request(url + '/password', data)
 220         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 221         password_request.add_header('Cookie', 'clip_test2=1; vuid=%s' % vuid)
 222         password_request.add_header('Referer', url)
 223         return self._download_webpage(
 224             password_request, video_id,
 225             'Verifying the password', 'Wrong password')
 226
 227     def _verify_player_video_password(self, url, video_id):
 228         password = self._downloader.params.get('videopassword', None)
 229         if password is None:
 230             raise ExtractorError('This video is protected by a password, use the --video-password option')
 231         data = urlencode_postdata(encode_dict({'password': password}))
 232         pass_url = url + '/check-password'
 233         password_request = compat_urllib_request.Request(pass_url, data)
 234         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 235         return self._download_json(
 236             password_request, video_id,
 237             'Verifying the password',
 238             'Wrong password')
 239
 240     def _real_initialize(self):
 241         self._login()
 242
 243     def _real_extract(self, url):
 244         url, data = unsmuggle_url(url)
 245         headers = std_headers
 246         if data is not None:
 247             headers = headers.copy()
 248             headers.update(data)
 249         if 'Referer' not in headers:
 250             headers['Referer'] = url
 251
 252         # Extract ID from URL
 253         mobj = re.match(self._VALID_URL, url)
 254         video_id = mobj.group('id')
 255         orig_url = url
 256         if mobj.group('pro') or mobj.group('player'):
 257             url = 'https://player.vimeo.com/video/' + video_id
 258         else:
 259             url = 'https://vimeo.com/' + video_id
 260
 261         # Retrieve video webpage to extract further information
 262         request = compat_urllib_request.Request(url, None, headers)
 263         try:
 264             webpage = self._download_webpage(request, video_id)
 265         except ExtractorError as ee:
 266             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 267                 errmsg = ee.cause.read()
 268                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 269                     raise ExtractorError(
 270                         'Cannot download embed-only video without embedding '
 271                         'URL. Please call youtube-dl with the URL of the page '
 272                         'that embeds this video.',
 273                         expected=True)
 274             raise
 275
 276         # Now we begin extracting as much information as we can from what we
 277         # retrieved. First we extract the information common to all extractors,
 278         # and latter we extract those that are Vimeo specific.
 279         self.report_extraction(video_id)
 280
 281         vimeo_config = self._search_regex(
 282             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 283             'vimeo config', default=None)
 284         if vimeo_config:
 285             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 286             if seed_status.get('state') == 'failed':
 287                 raise ExtractorError(
 288                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 289                     expected=True)
 290
 291         # Extract the config JSON
 292         try:
 293             try:
 294                 config_url = self._html_search_regex(
 295                     r' data-config-url="(.+?)"', webpage,
 296                     'config URL', default=None)
 297                 if not config_url:
 298                     # Sometimes new react-based page is served instead of old one that require
 299                     # different config URL extraction approach (see
 300                     # https://github.com/rg3/youtube-dl/pull/7209)
 301                     vimeo_clip_page_config = self._search_regex(
 302                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 303                         'vimeo clip page config')
 304                     config_url = self._parse_json(
 305                         vimeo_clip_page_config, video_id)['player']['config_url']
 306                 config_json = self._download_webpage(config_url, video_id)
 307                 config = json.loads(config_json)
 308             except RegexNotFoundError:
 309                 # For pro videos or player.vimeo.com urls
 310                 # We try to find out to which variable is assigned the config dic
 311                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 312                 if m_variable_name is not None:
 313                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 314                 else:
 315                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 316                 config = self._search_regex(config_re, webpage, 'info section',
 317                                             flags=re.DOTALL)
 318                 config = json.loads(config)
 319         except Exception as e:
 320             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 321                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 322
 323             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 324                 if data and '_video_password_verified' in data:
 325                     raise ExtractorError('video password verification failed!')
 326                 self._verify_video_password(url, video_id, webpage)
 327                 return self._real_extract(
 328                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 329             else:
 330                 raise ExtractorError('Unable to extract info section',
 331                                      cause=e)
 332         else:
 333             if config.get('view') == 4:
 334                 config = self._verify_player_video_password(url, video_id)
 335
 336         # Extract title
 337         video_title = config["video"]["title"]
 338
 339         # Extract uploader and uploader_id
 340         video_uploader = config["video"]["owner"]["name"]
 341         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 342
 343         # Extract video thumbnail
 344         video_thumbnail = config["video"].get("thumbnail")
 345         if video_thumbnail is None:
 346             video_thumbs = config["video"].get("thumbs")
 347             if video_thumbs and isinstance(video_thumbs, dict):
 348                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 349
 350         # Extract video description
 351
 352         video_description = self._html_search_regex(
 353             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 354             webpage, 'description', default=None)
 355         if not video_description:
 356             video_description = self._html_search_meta(
 357                 'description', webpage, default=None)
 358         if not video_description and mobj.group('pro'):
 359             orig_webpage = self._download_webpage(
 360                 orig_url, video_id,
 361                 note='Downloading webpage for description',
 362                 fatal=False)
 363             if orig_webpage:
 364                 video_description = self._html_search_meta(
 365                     'description', orig_webpage, default=None)
 366         if not video_description and not mobj.group('player'):
 367             self._downloader.report_warning('Cannot find video description')
 368
 369         # Extract video duration
 370         video_duration = int_or_none(config["video"].get("duration"))
 371
 372         # Extract upload date
 373         video_upload_date = None
 374         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 375         if mobj is not None:
 376             video_upload_date = unified_strdate(mobj.group(1))
 377
 378         try:
 379             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 380             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 381             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 382         except RegexNotFoundError:
 383             # This info is only available in vimeo.com/{id} urls
 384             view_count = None
 385             like_count = None
 386             comment_count = None
 387
 388         # Vimeo specific: extract request signature and timestamp
 389         sig = config['request']['signature']
 390         timestamp = config['request']['timestamp']
 391
 392         # Vimeo specific: extract video codec and quality information
 393         # First consider quality, then codecs, then take everything
 394         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 395         files = {'hd': [], 'sd': [], 'other': []}
 396         config_files = config["video"].get("files") or config["request"].get("files")
 397         for codec_name, codec_extension in codecs:
 398             for quality in config_files.get(codec_name, []):
 399                 format_id = '-'.join((codec_name, quality)).lower()
 400                 key = quality if quality in files else 'other'
 401                 video_url = None
 402                 if isinstance(config_files[codec_name], dict):
 403                     file_info = config_files[codec_name][quality]
 404                     video_url = file_info.get('url')
 405                 else:
 406                     file_info = {}
 407                 if video_url is None:
 408                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 409                         % (video_id, sig, timestamp, quality, codec_name.upper())
 410
 411                 files[key].append({
 412                     'ext': codec_extension,
 413                     'url': video_url,
 414                     'format_id': format_id,
 415                     'width': int_or_none(file_info.get('width')),
 416                     'height': int_or_none(file_info.get('height')),
 417                     'tbr': int_or_none(file_info.get('bitrate')),
 418                 })
 419         formats = []
 420         m3u8_url = config_files.get('hls', {}).get('all')
 421         if m3u8_url:
 422             m3u8_formats = self._extract_m3u8_formats(
 423                 m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
 424             if m3u8_formats:
 425                 formats.extend(m3u8_formats)
 426         for key in ('other', 'sd', 'hd'):
 427             formats += files[key]
 428         self._sort_formats(formats)
 429
 430         subtitles = {}
 431         text_tracks = config['request'].get('text_tracks')
 432         if text_tracks:
 433             for tt in text_tracks:
 434                 subtitles[tt['lang']] = [{
 435                     'ext': 'vtt',
 436                     'url': 'https://vimeo.com' + tt['url'],
 437                 }]
 438
 439         return {
 440             'id': video_id,
 441             'uploader': video_uploader,
 442             'uploader_id': video_uploader_id,
 443             'upload_date': video_upload_date,
 444             'title': video_title,
 445             'thumbnail': video_thumbnail,
 446             'description': video_description,
 447             'duration': video_duration,
 448             'formats': formats,
 449             'webpage_url': url,
 450             'view_count': view_count,
 451             'like_count': like_count,
 452             'comment_count': comment_count,
 453             'subtitles': subtitles,
 454         }
 455
 456
 457 class VimeoChannelIE(VimeoBaseInfoExtractor):
 458     IE_NAME = 'vimeo:channel'
 459     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 460     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 461     _TITLE = None
 462     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 463     _TESTS = [{
 464         'url': 'https://vimeo.com/channels/tributes',
 465         'info_dict': {
 466             'id': 'tributes',
 467             'title': 'Vimeo Tributes',
 468         },
 469         'playlist_mincount': 25,
 470     }]
 471
 472     def _page_url(self, base_url, pagenum):
 473         return '%s/videos/page:%d/' % (base_url, pagenum)
 474
 475     def _extract_list_title(self, webpage):
 476         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 477
 478     def _login_list_password(self, page_url, list_id, webpage):
 479         login_form = self._search_regex(
 480             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 481             webpage, 'login form', default=None)
 482         if not login_form:
 483             return webpage
 484
 485         password = self._downloader.params.get('videopassword', None)
 486         if password is None:
 487             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 488         fields = self._hidden_inputs(login_form)
 489         token, vuid = self._extract_xsrft_and_vuid(webpage)
 490         fields['token'] = token
 491         fields['password'] = password
 492         post = urlencode_postdata(encode_dict(fields))
 493         password_path = self._search_regex(
 494             r'action="([^"]+)"', login_form, 'password URL')
 495         password_url = compat_urlparse.urljoin(page_url, password_path)
 496         password_request = compat_urllib_request.Request(password_url, post)
 497         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 498         password_request.add_header('Cookie', 'vuid=%s' % vuid)
 499         self._set_cookie('vimeo.com', 'xsrft', token)
 500
 501         return self._download_webpage(
 502             password_request, list_id,
 503             'Verifying the password', 'Wrong password')
 504
 505     def _extract_videos(self, list_id, base_url):
 506         video_ids = []
 507         for pagenum in itertools.count(1):
 508             page_url = self._page_url(base_url, pagenum)
 509             webpage = self._download_webpage(
 510                 page_url, list_id,
 511                 'Downloading page %s' % pagenum)
 512
 513             if pagenum == 1:
 514                 webpage = self._login_list_password(page_url, list_id, webpage)
 515
 516             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 517             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 518                 break
 519
 520         entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo')
 521                    for video_id in video_ids]
 522         return {'_type': 'playlist',
 523                 'id': list_id,
 524                 'title': self._extract_list_title(webpage),
 525                 'entries': entries,
 526                 }
 527
 528     def _real_extract(self, url):
 529         mobj = re.match(self._VALID_URL, url)
 530         channel_id = mobj.group('id')
 531         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 532
 533
 534 class VimeoUserIE(VimeoChannelIE):
 535     IE_NAME = 'vimeo:user'
 536     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 537     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 538     _TESTS = [{
 539         'url': 'https://vimeo.com/nkistudio/videos',
 540         'info_dict': {
 541             'title': 'Nki',
 542             'id': 'nkistudio',
 543         },
 544         'playlist_mincount': 66,
 545     }]
 546
 547     def _real_extract(self, url):
 548         mobj = re.match(self._VALID_URL, url)
 549         name = mobj.group('name')
 550         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 551
 552
 553 class VimeoAlbumIE(VimeoChannelIE):
 554     IE_NAME = 'vimeo:album'
 555     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
 556     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 557     _TESTS = [{
 558         'url': 'https://vimeo.com/album/2632481',
 559         'info_dict': {
 560             'id': '2632481',
 561             'title': 'Staff Favorites: November 2013',
 562         },
 563         'playlist_mincount': 13,
 564     }, {
 565         'note': 'Password-protected album',
 566         'url': 'https://vimeo.com/album/3253534',
 567         'info_dict': {
 568             'title': 'test',
 569             'id': '3253534',
 570         },
 571         'playlist_count': 1,
 572         'params': {
 573             'videopassword': 'youtube-dl',
 574         }
 575     }]
 576
 577     def _page_url(self, base_url, pagenum):
 578         return '%s/page:%d/' % (base_url, pagenum)
 579
 580     def _real_extract(self, url):
 581         album_id = self._match_id(url)
 582         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 583
 584
 585 class VimeoGroupsIE(VimeoAlbumIE):
 586     IE_NAME = 'vimeo:group'
 587     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)'
 588     _TESTS = [{
 589         'url': 'https://vimeo.com/groups/rolexawards',
 590         'info_dict': {
 591             'id': 'rolexawards',
 592             'title': 'Rolex Awards for Enterprise',
 593         },
 594         'playlist_mincount': 73,
 595     }]
 596
 597     def _extract_list_title(self, webpage):
 598         return self._og_search_title(webpage)
 599
 600     def _real_extract(self, url):
 601         mobj = re.match(self._VALID_URL, url)
 602         name = mobj.group('name')
 603         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 604
 605
 606 class VimeoReviewIE(InfoExtractor):
 607     IE_NAME = 'vimeo:review'
 608     IE_DESC = 'Review pages on vimeo'
 609     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 610     _TESTS = [{
 611         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 612         'md5': 'c507a72f780cacc12b2248bb4006d253',
 613         'info_dict': {
 614             'id': '75524534',
 615             'ext': 'mp4',
 616             'title': "DICK HARDWICK 'Comedian'",
 617             'uploader': 'Richard Hardwick',
 618         }
 619     }, {
 620         'note': 'video player needs Referer',
 621         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 622         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 623         'info_dict': {
 624             'id': '91613211',
 625             'ext': 'mp4',
 626             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 627             'uploader': 'DevWeek Events',
 628             'duration': 2773,
 629             'thumbnail': 're:^https?://.*\.jpg$',
 630         }
 631     }]
 632
 633     def _real_extract(self, url):
 634         mobj = re.match(self._VALID_URL, url)
 635         video_id = mobj.group('id')
 636         player_url = 'https://player.vimeo.com/player/' + video_id
 637         return self.url_result(player_url, 'Vimeo', video_id)
 638
 639
 640 class VimeoWatchLaterIE(VimeoChannelIE):
 641     IE_NAME = 'vimeo:watchlater'
 642     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 643     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 644     _TITLE = 'Watch Later'
 645     _LOGIN_REQUIRED = True
 646     _TESTS = [{
 647         'url': 'https://vimeo.com/watchlater',
 648         'only_matching': True,
 649     }]
 650
 651     def _real_initialize(self):
 652         self._login()
 653
 654     def _page_url(self, base_url, pagenum):
 655         url = '%s/page:%d/' % (base_url, pagenum)
 656         request = compat_urllib_request.Request(url)
 657         # Set the header to get a partial html page with the ids,
 658         # the normal page doesn't contain them.
 659         request.add_header('X-Requested-With', 'XMLHttpRequest')
 660         return request
 661
 662     def _real_extract(self, url):
 663         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 664
 665
 666 class VimeoLikesIE(InfoExtractor):
 667     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 668     IE_NAME = 'vimeo:likes'
 669     IE_DESC = 'Vimeo user likes'
 670     _TEST = {
 671         'url': 'https://vimeo.com/user755559/likes/',
 672         'playlist_mincount': 293,
 673         "info_dict": {
 674             'id': 'user755559_likes',
 675             "description": "See all the videos urza likes",
 676             "title": 'Videos urza likes',
 677         },
 678     }
 679
 680     def _real_extract(self, url):
 681         user_id = self._match_id(url)
 682         webpage = self._download_webpage(url, user_id)
 683         page_count = self._int(
 684             self._search_regex(
 685                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 686                     .*?</a></li>\s*<li\s+class="pagination_next">
 687                 ''', webpage, 'page count'),
 688             'page count', fatal=True)
 689         PAGE_SIZE = 12
 690         title = self._html_search_regex(
 691             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 692         description = self._html_search_meta('description', webpage)
 693
 694         def _get_page(idx):
 695             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 696                 user_id, idx + 1)
 697             webpage = self._download_webpage(
 698                 page_url, user_id,
 699                 note='Downloading page %d/%d' % (idx + 1, page_count))
 700             video_list = self._search_regex(
 701                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 702                 webpage, 'video content')
 703             paths = re.findall(
 704                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 705             for path in paths:
 706                 yield {
 707                     '_type': 'url',
 708                     'url': compat_urlparse.urljoin(page_url, path),
 709                 }
 710
 711         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 712
 713         return {
 714             '_type': 'playlist',
 715             'id': 'user%s_likes' % user_id,
 716             'title': title,
 717             'description': description,
 718             'entries': pl,
 719         }