_ 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     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 = self._extract_xsrft(webpage)
  43         vuid = self._search_regex(
  44             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  45             webpage, 'vuid', group='vuid')
  46         data = urlencode_postdata({
  47             'action': 'login',
  48             'email': username,
  49             'password': password,
  50             'service': 'vimeo',
  51             'token': token,
  52         })
  53         login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
  54         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  55         login_request.add_header('Cookie', 'vuid=%s' % vuid)
  56         login_request.add_header('Referer', self._LOGIN_URL)
  57         self._download_webpage(login_request, None, False, 'Wrong login info')
  58
  59     def _extract_xsrft(self, webpage):
  60         return self._search_regex(
  61             r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  62             webpage, 'login token', group='xsrft')
  63
  64
  65 class VimeoIE(VimeoBaseInfoExtractor):
  66     """Information extractor for vimeo.com."""
  67
  68     # _VALID_URL matches Vimeo URLs
  69     _VALID_URL = r'''(?x)
  70         https?://
  71         (?:(?:www|(?P<player>player))\.)?
  72         vimeo(?P<pro>pro)?\.com/
  73         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  74         (?:.*?/)?
  75         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  76         (?:videos?/)?
  77         (?P<id>[0-9]+)
  78         /?(?:[?&].*)?(?:[#].*)?$'''
  79     IE_NAME = 'vimeo'
  80     _TESTS = [
  81         {
  82             'url': 'http://vimeo.com/56015672#at=0',
  83             'md5': '8879b6cc097e987f02484baf890129e5',
  84             'info_dict': {
  85                 'id': '56015672',
  86                 'ext': 'mp4',
  87                 "upload_date": "20121220",
  88                 "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",
  89                 "uploader_id": "user7108434",
  90                 "uploader": "Filippo Valsorda",
  91                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  92                 "duration": 10,
  93             },
  94         },
  95         {
  96             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  97             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  98             'note': 'Vimeo Pro video (#1197)',
  99             'info_dict': {
 100                 'id': '68093876',
 101                 'ext': 'mp4',
 102                 'uploader_id': 'openstreetmapus',
 103                 'uploader': 'OpenStreetMap US',
 104                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 105                 'description': 'md5:380943ec71b89736ff4bf27183233d09',
 106                 'duration': 1595,
 107             },
 108         },
 109         {
 110             'url': 'http://player.vimeo.com/video/54469442',
 111             'md5': '619b811a4417aa4abe78dc653becf511',
 112             'note': 'Videos that embed the url in the player page',
 113             'info_dict': {
 114                 'id': '54469442',
 115                 'ext': 'mp4',
 116                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 117                 'uploader': 'The BLN & Business of Software',
 118                 'uploader_id': 'theblnbusinessofsoftware',
 119                 'duration': 3610,
 120                 'description': None,
 121             },
 122         },
 123         {
 124             'url': 'http://vimeo.com/68375962',
 125             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 126             'note': 'Video protected with password',
 127             'info_dict': {
 128                 'id': '68375962',
 129                 'ext': 'mp4',
 130                 'title': 'youtube-dl password protected test video',
 131                 'upload_date': '20130614',
 132                 'uploader_id': 'user18948128',
 133                 'uploader': 'Jaime Marquínez Ferrándiz',
 134                 'duration': 10,
 135                 '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.',
 136             },
 137             'params': {
 138                 'videopassword': 'youtube-dl',
 139             },
 140         },
 141         {
 142             'url': 'http://vimeo.com/channels/keypeele/75629013',
 143             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 144             'note': 'Video is freely available via original URL '
 145                     'and protected with password when accessed via http://vimeo.com/75629013',
 146             'info_dict': {
 147                 'id': '75629013',
 148                 'ext': 'mp4',
 149                 'title': 'Key & Peele: Terrorist Interrogation',
 150                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 151                 'uploader_id': 'atencio',
 152                 'uploader': 'Peter Atencio',
 153                 'upload_date': '20130927',
 154                 'duration': 187,
 155             },
 156         },
 157         {
 158             'url': 'http://vimeo.com/76979871',
 159             'md5': '3363dd6ffebe3784d56f4132317fd446',
 160             'note': 'Video with subtitles',
 161             'info_dict': {
 162                 'id': '76979871',
 163                 'ext': 'mp4',
 164                 'title': 'The New Vimeo Player (You Know, For Videos)',
 165                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 166                 'upload_date': '20131015',
 167                 'uploader_id': 'staff',
 168                 'uploader': 'Vimeo Staff',
 169                 'duration': 62,
 170             }
 171         },
 172         {
 173             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 174             'url': 'https://player.vimeo.com/video/98044508',
 175             'note': 'The js code contains assignments to the same variable as the config',
 176             'info_dict': {
 177                 'id': '98044508',
 178                 'ext': 'mp4',
 179                 'title': 'Pier Solar OUYA Official Trailer',
 180                 'uploader': 'Tulio Gonçalves',
 181                 'uploader_id': 'user28849593',
 182             },
 183         },
 184     ]
 185
 186     @staticmethod
 187     def _extract_vimeo_url(url, webpage):
 188         # Look for embedded (iframe) Vimeo player
 189         mobj = re.search(
 190             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 191         if mobj:
 192             player_url = unescapeHTML(mobj.group('url'))
 193             surl = smuggle_url(player_url, {'Referer': url})
 194             return surl
 195         # Look for embedded (swf embed) Vimeo player
 196         mobj = re.search(
 197             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 198         if mobj:
 199             return mobj.group(1)
 200
 201     def _verify_video_password(self, url, video_id, webpage):
 202         password = self._downloader.params.get('videopassword', None)
 203         if password is None:
 204             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
 205         token = self._extract_xsrft(webpage)
 206         data = urlencode_postdata({
 207             'password': password,
 208             'token': token,
 209         })
 210         if url.startswith('http://'):
 211             # vimeo only supports https now, but the user can give an http url
 212             url = url.replace('http://', 'https://')
 213         password_request = compat_urllib_request.Request(url + '/password', data)
 214         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 215         password_request.add_header('Referer', url)
 216         return self._download_webpage(
 217             password_request, video_id,
 218             'Verifying the password', 'Wrong password')
 219
 220     def _verify_player_video_password(self, url, video_id):
 221         password = self._downloader.params.get('videopassword', None)
 222         if password is None:
 223             raise ExtractorError('This video is protected by a password, use the --video-password option')
 224         data = compat_urllib_parse.urlencode({'password': password})
 225         pass_url = url + '/check-password'
 226         password_request = compat_urllib_request.Request(pass_url, data)
 227         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 228         return self._download_json(
 229             password_request, video_id,
 230             'Verifying the password',
 231             'Wrong password')
 232
 233     def _real_initialize(self):
 234         self._login()
 235
 236     def _real_extract(self, url):
 237         url, data = unsmuggle_url(url)
 238         headers = std_headers
 239         if data is not None:
 240             headers = headers.copy()
 241             headers.update(data)
 242         if 'Referer' not in headers:
 243             headers['Referer'] = url
 244
 245         # Extract ID from URL
 246         mobj = re.match(self._VALID_URL, url)
 247         video_id = mobj.group('id')
 248         orig_url = url
 249         if mobj.group('pro') or mobj.group('player'):
 250             url = 'https://player.vimeo.com/video/' + video_id
 251         else:
 252             url = 'https://vimeo.com/' + video_id
 253
 254         # Retrieve video webpage to extract further information
 255         request = compat_urllib_request.Request(url, None, headers)
 256         try:
 257             webpage = self._download_webpage(request, video_id)
 258         except ExtractorError as ee:
 259             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 260                 errmsg = ee.cause.read()
 261                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 262                     raise ExtractorError(
 263                         'Cannot download embed-only video without embedding '
 264                         'URL. Please call youtube-dl with the URL of the page '
 265                         'that embeds this video.',
 266                         expected=True)
 267             raise
 268
 269         # Now we begin extracting as much information as we can from what we
 270         # retrieved. First we extract the information common to all extractors,
 271         # and latter we extract those that are Vimeo specific.
 272         self.report_extraction(video_id)
 273
 274         vimeo_config = self._search_regex(
 275             r'vimeo\.config\s*=\s*({.+?});', webpage,
 276             'vimeo config', default=None)
 277         if vimeo_config:
 278             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 279             if seed_status.get('state') == 'failed':
 280                 raise ExtractorError(
 281                     '%s returned error: %s' % (self.IE_NAME, seed_status['title']),
 282                     expected=True)
 283
 284         # Extract the config JSON
 285         try:
 286             try:
 287                 config_url = self._html_search_regex(
 288                     r' data-config-url="(.+?)"', webpage, 'config URL')
 289                 config_json = self._download_webpage(config_url, video_id)
 290                 config = json.loads(config_json)
 291             except RegexNotFoundError:
 292                 # For pro videos or player.vimeo.com urls
 293                 # We try to find out to which variable is assigned the config dic
 294                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 295                 if m_variable_name is not None:
 296                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 297                 else:
 298                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 299                 config = self._search_regex(config_re, webpage, 'info section',
 300                                             flags=re.DOTALL)
 301                 config = json.loads(config)
 302         except Exception as e:
 303             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 304                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 305
 306             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 307                 if data and '_video_password_verified' in data:
 308                     raise ExtractorError('video password verification failed!')
 309                 self._verify_video_password(url, video_id, webpage)
 310                 return self._real_extract(
 311                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 312             else:
 313                 raise ExtractorError('Unable to extract info section',
 314                                      cause=e)
 315         else:
 316             if config.get('view') == 4:
 317                 config = self._verify_player_video_password(url, video_id)
 318
 319         # Extract title
 320         video_title = config["video"]["title"]
 321
 322         # Extract uploader and uploader_id
 323         video_uploader = config["video"]["owner"]["name"]
 324         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 325
 326         # Extract video thumbnail
 327         video_thumbnail = config["video"].get("thumbnail")
 328         if video_thumbnail is None:
 329             video_thumbs = config["video"].get("thumbs")
 330             if video_thumbs and isinstance(video_thumbs, dict):
 331                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 332
 333         # Extract video description
 334
 335         video_description = self._html_search_regex(
 336             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 337             webpage, 'description', default=None)
 338         if not video_description:
 339             video_description = self._html_search_meta(
 340                 'description', webpage, default=None)
 341         if not video_description and mobj.group('pro'):
 342             orig_webpage = self._download_webpage(
 343                 orig_url, video_id,
 344                 note='Downloading webpage for description',
 345                 fatal=False)
 346             if orig_webpage:
 347                 video_description = self._html_search_meta(
 348                     'description', orig_webpage, default=None)
 349         if not video_description and not mobj.group('player'):
 350             self._downloader.report_warning('Cannot find video description')
 351
 352         # Extract video duration
 353         video_duration = int_or_none(config["video"].get("duration"))
 354
 355         # Extract upload date
 356         video_upload_date = None
 357         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 358         if mobj is not None:
 359             video_upload_date = unified_strdate(mobj.group(1))
 360
 361         try:
 362             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 363             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 364             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 365         except RegexNotFoundError:
 366             # This info is only available in vimeo.com/{id} urls
 367             view_count = None
 368             like_count = None
 369             comment_count = None
 370
 371         # Vimeo specific: extract request signature and timestamp
 372         sig = config['request']['signature']
 373         timestamp = config['request']['timestamp']
 374
 375         # Vimeo specific: extract video codec and quality information
 376         # First consider quality, then codecs, then take everything
 377         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 378         files = {'hd': [], 'sd': [], 'other': []}
 379         config_files = config["video"].get("files") or config["request"].get("files")
 380         for codec_name, codec_extension in codecs:
 381             for quality in config_files.get(codec_name, []):
 382                 format_id = '-'.join((codec_name, quality)).lower()
 383                 key = quality if quality in files else 'other'
 384                 video_url = None
 385                 if isinstance(config_files[codec_name], dict):
 386                     file_info = config_files[codec_name][quality]
 387                     video_url = file_info.get('url')
 388                 else:
 389                     file_info = {}
 390                 if video_url is None:
 391                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 392                         % (video_id, sig, timestamp, quality, codec_name.upper())
 393
 394                 files[key].append({
 395                     'ext': codec_extension,
 396                     'url': video_url,
 397                     'format_id': format_id,
 398                     'width': file_info.get('width'),
 399                     'height': file_info.get('height'),
 400                     'tbr': file_info.get('bitrate'),
 401                 })
 402         formats = []
 403         hls = config_files.get("hls")
 404         if hls:
 405             formats = self._extract_m3u8_formats(hls['all'], video_id, m3u8_id='hls')
 406         for key in ('other', 'sd', 'hd'):
 407             formats += files[key]
 408         if len(formats) == 0:
 409             raise ExtractorError('No known codec found')
 410
 411         subtitles = {}
 412         text_tracks = config['request'].get('text_tracks')
 413         if text_tracks:
 414             for tt in text_tracks:
 415                 subtitles[tt['lang']] = [{
 416                     'ext': 'vtt',
 417                     'url': 'https://vimeo.com' + tt['url'],
 418                 }]
 419
 420         return {
 421             'id': video_id,
 422             'uploader': video_uploader,
 423             'uploader_id': video_uploader_id,
 424             'upload_date': video_upload_date,
 425             'title': video_title,
 426             'thumbnail': video_thumbnail,
 427             'description': video_description,
 428             'duration': video_duration,
 429             'formats': formats,
 430             'webpage_url': url,
 431             'view_count': view_count,
 432             'like_count': like_count,
 433             'comment_count': comment_count,
 434             'subtitles': subtitles,
 435         }
 436
 437
 438 class VimeoChannelIE(VimeoBaseInfoExtractor):
 439     IE_NAME = 'vimeo:channel'
 440     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 441     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 442     _TITLE = None
 443     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 444     _TESTS = [{
 445         'url': 'https://vimeo.com/channels/tributes',
 446         'info_dict': {
 447             'id': 'tributes',
 448             'title': 'Vimeo Tributes',
 449         },
 450         'playlist_mincount': 25,
 451     }]
 452
 453     def _page_url(self, base_url, pagenum):
 454         return '%s/videos/page:%d/' % (base_url, pagenum)
 455
 456     def _extract_list_title(self, webpage):
 457         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 458
 459     def _login_list_password(self, page_url, list_id, webpage):
 460         login_form = self._search_regex(
 461             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 462             webpage, 'login form', default=None)
 463         if not login_form:
 464             return webpage
 465
 466         password = self._downloader.params.get('videopassword', None)
 467         if password is None:
 468             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 469         fields = self._hidden_inputs(login_form)
 470         token = self._extract_xsrft(webpage)
 471         fields['token'] = token
 472         fields['password'] = password
 473         post = urlencode_postdata(fields)
 474         password_path = self._search_regex(
 475             r'action="([^"]+)"', login_form, 'password URL')
 476         password_url = compat_urlparse.urljoin(page_url, password_path)
 477         password_request = compat_urllib_request.Request(password_url, post)
 478         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 479         self._set_cookie('vimeo.com', 'xsrft', token)
 480
 481         return self._download_webpage(
 482             password_request, list_id,
 483             'Verifying the password', 'Wrong password')
 484
 485     def _extract_videos(self, list_id, base_url):
 486         video_ids = []
 487         for pagenum in itertools.count(1):
 488             page_url = self._page_url(base_url, pagenum)
 489             webpage = self._download_webpage(
 490                 page_url, list_id,
 491                 'Downloading page %s' % pagenum)
 492
 493             if pagenum == 1:
 494                 webpage = self._login_list_password(page_url, list_id, webpage)
 495
 496             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 497             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 498                 break
 499
 500         entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo')
 501                    for video_id in video_ids]
 502         return {'_type': 'playlist',
 503                 'id': list_id,
 504                 'title': self._extract_list_title(webpage),
 505                 'entries': entries,
 506                 }
 507
 508     def _real_extract(self, url):
 509         mobj = re.match(self._VALID_URL, url)
 510         channel_id = mobj.group('id')
 511         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 512
 513
 514 class VimeoUserIE(VimeoChannelIE):
 515     IE_NAME = 'vimeo:user'
 516     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 517     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 518     _TESTS = [{
 519         'url': 'https://vimeo.com/nkistudio/videos',
 520         'info_dict': {
 521             'title': 'Nki',
 522             'id': 'nkistudio',
 523         },
 524         'playlist_mincount': 66,
 525     }]
 526
 527     def _real_extract(self, url):
 528         mobj = re.match(self._VALID_URL, url)
 529         name = mobj.group('name')
 530         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 531
 532
 533 class VimeoAlbumIE(VimeoChannelIE):
 534     IE_NAME = 'vimeo:album'
 535     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
 536     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 537     _TESTS = [{
 538         'url': 'https://vimeo.com/album/2632481',
 539         'info_dict': {
 540             'id': '2632481',
 541             'title': 'Staff Favorites: November 2013',
 542         },
 543         'playlist_mincount': 13,
 544     }, {
 545         'note': 'Password-protected album',
 546         'url': 'https://vimeo.com/album/3253534',
 547         'info_dict': {
 548             'title': 'test',
 549             'id': '3253534',
 550         },
 551         'playlist_count': 1,
 552         'params': {
 553             'videopassword': 'youtube-dl',
 554         }
 555     }]
 556
 557     def _page_url(self, base_url, pagenum):
 558         return '%s/page:%d/' % (base_url, pagenum)
 559
 560     def _real_extract(self, url):
 561         album_id = self._match_id(url)
 562         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 563
 564
 565 class VimeoGroupsIE(VimeoAlbumIE):
 566     IE_NAME = 'vimeo:group'
 567     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)'
 568     _TESTS = [{
 569         'url': 'https://vimeo.com/groups/rolexawards',
 570         'info_dict': {
 571             'id': 'rolexawards',
 572             'title': 'Rolex Awards for Enterprise',
 573         },
 574         'playlist_mincount': 73,
 575     }]
 576
 577     def _extract_list_title(self, webpage):
 578         return self._og_search_title(webpage)
 579
 580     def _real_extract(self, url):
 581         mobj = re.match(self._VALID_URL, url)
 582         name = mobj.group('name')
 583         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 584
 585
 586 class VimeoReviewIE(InfoExtractor):
 587     IE_NAME = 'vimeo:review'
 588     IE_DESC = 'Review pages on vimeo'
 589     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 590     _TESTS = [{
 591         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 592         'md5': 'c507a72f780cacc12b2248bb4006d253',
 593         'info_dict': {
 594             'id': '75524534',
 595             'ext': 'mp4',
 596             'title': "DICK HARDWICK 'Comedian'",
 597             'uploader': 'Richard Hardwick',
 598         }
 599     }, {
 600         'note': 'video player needs Referer',
 601         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 602         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 603         'info_dict': {
 604             'id': '91613211',
 605             'ext': 'mp4',
 606             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 607             'uploader': 'DevWeek Events',
 608             'duration': 2773,
 609             'thumbnail': 're:^https?://.*\.jpg$',
 610         }
 611     }]
 612
 613     def _real_extract(self, url):
 614         mobj = re.match(self._VALID_URL, url)
 615         video_id = mobj.group('id')
 616         player_url = 'https://player.vimeo.com/player/' + video_id
 617         return self.url_result(player_url, 'Vimeo', video_id)
 618
 619
 620 class VimeoWatchLaterIE(VimeoChannelIE):
 621     IE_NAME = 'vimeo:watchlater'
 622     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 623     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 624     _TITLE = 'Watch Later'
 625     _LOGIN_REQUIRED = True
 626     _TESTS = [{
 627         'url': 'https://vimeo.com/watchlater',
 628         'only_matching': True,
 629     }]
 630
 631     def _real_initialize(self):
 632         self._login()
 633
 634     def _page_url(self, base_url, pagenum):
 635         url = '%s/page:%d/' % (base_url, pagenum)
 636         request = compat_urllib_request.Request(url)
 637         # Set the header to get a partial html page with the ids,
 638         # the normal page doesn't contain them.
 639         request.add_header('X-Requested-With', 'XMLHttpRequest')
 640         return request
 641
 642     def _real_extract(self, url):
 643         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 644
 645
 646 class VimeoLikesIE(InfoExtractor):
 647     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 648     IE_NAME = 'vimeo:likes'
 649     IE_DESC = 'Vimeo user likes'
 650     _TEST = {
 651         'url': 'https://vimeo.com/user755559/likes/',
 652         'playlist_mincount': 293,
 653         "info_dict": {
 654             'id': 'user755559_likes',
 655             "description": "See all the videos urza likes",
 656             "title": 'Videos urza likes',
 657         },
 658     }
 659
 660     def _real_extract(self, url):
 661         user_id = self._match_id(url)
 662         webpage = self._download_webpage(url, user_id)
 663         page_count = self._int(
 664             self._search_regex(
 665                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 666                     .*?</a></li>\s*<li\s+class="pagination_next">
 667                 ''', webpage, 'page count'),
 668             'page count', fatal=True)
 669         PAGE_SIZE = 12
 670         title = self._html_search_regex(
 671             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 672         description = self._html_search_meta('description', webpage)
 673
 674         def _get_page(idx):
 675             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 676                 user_id, idx + 1)
 677             webpage = self._download_webpage(
 678                 page_url, user_id,
 679                 note='Downloading page %d/%d' % (idx + 1, page_count))
 680             video_list = self._search_regex(
 681                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 682                 webpage, 'video content')
 683             paths = re.findall(
 684                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 685             for path in paths:
 686                 yield {
 687                     '_type': 'url',
 688                     'url': compat_urlparse.urljoin(page_url, path),
 689                 }
 690
 691         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 692
 693         return {
 694             '_type': 'playlist',
 695             'id': 'user%s_likes' % user_id,
 696             'title': title,
 697             'description': description,
 698             'entries': pl,
 699         }