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