youtube_dl/extractor/vimeo.py

   1 # coding: 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_str,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     InAdvancePagedList,
  18     int_or_none,
  19     NO_DEFAULT,
  20     RegexNotFoundError,
  21     sanitized_Request,
  22     smuggle_url,
  23     std_headers,
  24     unified_strdate,
  25     unsmuggle_url,
  26     urlencode_postdata,
  27     unescapeHTML,
  28     parse_filesize,
  29     try_get,
  30 )
  31
  32
  33 class VimeoBaseInfoExtractor(InfoExtractor):
  34     _NETRC_MACHINE = 'vimeo'
  35     _LOGIN_REQUIRED = False
  36     _LOGIN_URL = 'https://vimeo.com/log_in'
  37
  38     def _login(self):
  39         (username, password) = self._get_login_info()
  40         if username is None:
  41             if self._LOGIN_REQUIRED:
  42                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  43             return
  44         self.report_login()
  45         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  46         token, vuid = self._extract_xsrft_and_vuid(webpage)
  47         data = urlencode_postdata({
  48             'action': 'login',
  49             'email': username,
  50             'password': password,
  51             'service': 'vimeo',
  52             'token': token,
  53         })
  54         login_request = sanitized_Request(self._LOGIN_URL, data)
  55         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  56         login_request.add_header('Referer', self._LOGIN_URL)
  57         self._set_vimeo_cookie('vuid', vuid)
  58         self._download_webpage(login_request, None, False, 'Wrong login info')
  59
  60     def _verify_video_password(self, url, video_id, webpage):
  61         password = self._downloader.params.get('videopassword')
  62         if password is None:
  63             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  64         token, vuid = self._extract_xsrft_and_vuid(webpage)
  65         data = urlencode_postdata({
  66             'password': password,
  67             'token': token,
  68         })
  69         if url.startswith('http://'):
  70             # vimeo only supports https now, but the user can give an http url
  71             url = url.replace('http://', 'https://')
  72         password_request = sanitized_Request(url + '/password', data)
  73         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  74         password_request.add_header('Referer', url)
  75         self._set_vimeo_cookie('vuid', vuid)
  76         return self._download_webpage(
  77             password_request, video_id,
  78             'Verifying the password', 'Wrong password')
  79
  80     def _extract_xsrft_and_vuid(self, webpage):
  81         xsrft = self._search_regex(
  82             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  83             webpage, 'login token', group='xsrft')
  84         vuid = self._search_regex(
  85             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  86             webpage, 'vuid', group='vuid')
  87         return xsrft, vuid
  88
  89     def _set_vimeo_cookie(self, name, value):
  90         self._set_cookie('vimeo.com', name, value)
  91
  92     def _vimeo_sort_formats(self, formats):
  93         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
  94         # at the same time without actual units specified. This lead to wrong sorting.
  95         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'format_id'))
  96
  97     def _parse_config(self, config, video_id):
  98         # Extract title
  99         video_title = config['video']['title']
 100
 101         # Extract uploader, uploader_url and uploader_id
 102         video_uploader = config['video'].get('owner', {}).get('name')
 103         video_uploader_url = config['video'].get('owner', {}).get('url')
 104         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 105
 106         # Extract video thumbnail
 107         video_thumbnail = config['video'].get('thumbnail')
 108         if video_thumbnail is None:
 109             video_thumbs = config['video'].get('thumbs')
 110             if video_thumbs and isinstance(video_thumbs, dict):
 111                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 112
 113         # Extract video duration
 114         video_duration = int_or_none(config['video'].get('duration'))
 115
 116         formats = []
 117         config_files = config['video'].get('files') or config['request'].get('files', {})
 118         for f in config_files.get('progressive', []):
 119             video_url = f.get('url')
 120             if not video_url:
 121                 continue
 122             formats.append({
 123                 'url': video_url,
 124                 'format_id': 'http-%s' % f.get('quality'),
 125                 'width': int_or_none(f.get('width')),
 126                 'height': int_or_none(f.get('height')),
 127                 'fps': int_or_none(f.get('fps')),
 128                 'tbr': int_or_none(f.get('bitrate')),
 129             })
 130         m3u8_url = config_files.get('hls', {}).get('url')
 131         if m3u8_url:
 132             formats.extend(self._extract_m3u8_formats(
 133                 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
 134
 135         subtitles = {}
 136         text_tracks = config['request'].get('text_tracks')
 137         if text_tracks:
 138             for tt in text_tracks:
 139                 subtitles[tt['lang']] = [{
 140                     'ext': 'vtt',
 141                     'url': 'https://vimeo.com' + tt['url'],
 142                 }]
 143
 144         return {
 145             'title': video_title,
 146             'uploader': video_uploader,
 147             'uploader_id': video_uploader_id,
 148             'uploader_url': video_uploader_url,
 149             'thumbnail': video_thumbnail,
 150             'duration': video_duration,
 151             'formats': formats,
 152             'subtitles': subtitles,
 153         }
 154
 155
 156 class VimeoIE(VimeoBaseInfoExtractor):
 157     """Information extractor for vimeo.com."""
 158
 159     # _VALID_URL matches Vimeo URLs
 160     _VALID_URL = r'''(?x)
 161                     https?://
 162                         (?:
 163                             (?:
 164                                 www|
 165                                 (?P<player>player)
 166                             )
 167                             \.
 168                         )?
 169                         vimeo(?P<pro>pro)?\.com/
 170                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 171                         (?:.*?/)?
 172                         (?:
 173                             (?:
 174                                 play_redirect_hls|
 175                                 moogaloop\.swf)\?clip_id=
 176                             )?
 177                         (?:videos?/)?
 178                         (?P<id>[0-9]+)
 179                         (?:/[\da-f]+)?
 180                         /?(?:[?&].*)?(?:[#].*)?$
 181                     '''
 182     IE_NAME = 'vimeo'
 183     _TESTS = [
 184         {
 185             'url': 'http://vimeo.com/56015672#at=0',
 186             'md5': '8879b6cc097e987f02484baf890129e5',
 187             'info_dict': {
 188                 'id': '56015672',
 189                 'ext': 'mp4',
 190                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 191                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 192                 'upload_date': '20121220',
 193                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user7108434',
 194                 'uploader_id': 'user7108434',
 195                 'uploader': 'Filippo Valsorda',
 196                 'duration': 10,
 197             },
 198         },
 199         {
 200             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 201             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 202             'note': 'Vimeo Pro video (#1197)',
 203             'info_dict': {
 204                 'id': '68093876',
 205                 'ext': 'mp4',
 206                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 207                 'uploader_id': 'openstreetmapus',
 208                 'uploader': 'OpenStreetMap US',
 209                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 210                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 211                 'duration': 1595,
 212             },
 213         },
 214         {
 215             'url': 'http://player.vimeo.com/video/54469442',
 216             'md5': '619b811a4417aa4abe78dc653becf511',
 217             'note': 'Videos that embed the url in the player page',
 218             'info_dict': {
 219                 'id': '54469442',
 220                 'ext': 'mp4',
 221                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 222                 'uploader': 'The BLN & Business of Software',
 223                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 224                 'uploader_id': 'theblnbusinessofsoftware',
 225                 'duration': 3610,
 226                 'description': None,
 227             },
 228         },
 229         {
 230             'url': 'http://vimeo.com/68375962',
 231             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 232             'note': 'Video protected with password',
 233             'info_dict': {
 234                 'id': '68375962',
 235                 'ext': 'mp4',
 236                 'title': 'youtube-dl password protected test video',
 237                 'upload_date': '20130614',
 238                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user18948128',
 239                 'uploader_id': 'user18948128',
 240                 'uploader': 'Jaime Marquínez Ferrándiz',
 241                 'duration': 10,
 242                 'description': 'This is "youtube-dl password protected test video" by  on Vimeo, the home for high quality videos and the people who love them.',
 243             },
 244             'params': {
 245                 'videopassword': 'youtube-dl',
 246             },
 247         },
 248         {
 249             'url': 'http://vimeo.com/channels/keypeele/75629013',
 250             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 251             'info_dict': {
 252                 'id': '75629013',
 253                 'ext': 'mp4',
 254                 'title': 'Key & Peele: Terrorist Interrogation',
 255                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 256                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/atencio',
 257                 'uploader_id': 'atencio',
 258                 'uploader': 'Peter Atencio',
 259                 'upload_date': '20130927',
 260                 'duration': 187,
 261             },
 262         },
 263         {
 264             'url': 'http://vimeo.com/76979871',
 265             'note': 'Video with subtitles',
 266             'info_dict': {
 267                 'id': '76979871',
 268                 'ext': 'mp4',
 269                 'title': 'The New Vimeo Player (You Know, For Videos)',
 270                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 271                 'upload_date': '20131015',
 272                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/staff',
 273                 'uploader_id': 'staff',
 274                 'uploader': 'Vimeo Staff',
 275                 'duration': 62,
 276             }
 277         },
 278         {
 279             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 280             'url': 'https://player.vimeo.com/video/98044508',
 281             'note': 'The js code contains assignments to the same variable as the config',
 282             'info_dict': {
 283                 'id': '98044508',
 284                 'ext': 'mp4',
 285                 'title': 'Pier Solar OUYA Official Trailer',
 286                 'uploader': 'Tulio Gonçalves',
 287                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user28849593',
 288                 'uploader_id': 'user28849593',
 289             },
 290         },
 291         {
 292             # contains original format
 293             'url': 'https://vimeo.com/33951933',
 294             'md5': '2d9f5475e0537f013d0073e812ab89e6',
 295             'info_dict': {
 296                 'id': '33951933',
 297                 'ext': 'mp4',
 298                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 299                 'uploader': 'The DMCI',
 300                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/dmci',
 301                 'uploader_id': 'dmci',
 302                 'upload_date': '20111220',
 303                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 304             },
 305         },
 306         {
 307             # only available via https://vimeo.com/channels/tributes/6213729 and
 308             # not via https://vimeo.com/6213729
 309             'url': 'https://vimeo.com/channels/tributes/6213729',
 310             'info_dict': {
 311                 'id': '6213729',
 312                 'ext': 'mp4',
 313                 'title': 'Vimeo Tribute: The Shining',
 314                 'uploader': 'Casey Donahue',
 315                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 316                 'uploader_id': 'caseydonahue',
 317                 'upload_date': '20090821',
 318                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 319             },
 320             'params': {
 321                 'skip_download': True,
 322             },
 323             'expected_warnings': ['Unable to download JSON metadata'],
 324         },
 325         {
 326             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 327             'only_matching': True,
 328         },
 329         {
 330             'url': 'https://vimeo.com/109815029',
 331             'note': 'Video not completely processed, "failed" seed status',
 332             'only_matching': True,
 333         },
 334         {
 335             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 336             'only_matching': True,
 337         },
 338         {
 339             'url': 'https://vimeo.com/album/2632481/video/79010983',
 340             'only_matching': True,
 341         },
 342         {
 343             # source file returns 403: Forbidden
 344             'url': 'https://vimeo.com/7809605',
 345             'only_matching': True,
 346         },
 347         {
 348             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 349             'only_matching': True,
 350         }
 351     ]
 352
 353     @staticmethod
 354     def _smuggle_referrer(url, referrer_url):
 355         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 356
 357     @staticmethod
 358     def _extract_urls(url, webpage):
 359         urls = []
 360         # Look for embedded (iframe) Vimeo player
 361         for mobj in re.finditer(
 362                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1',
 363                 webpage):
 364             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 365         PLAIN_EMBED_RE = (
 366             # Look for embedded (swf embed) Vimeo player
 367             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 368             # Look more for non-standard embedded Vimeo player
 369             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 370         )
 371         for embed_re in PLAIN_EMBED_RE:
 372             for mobj in re.finditer(embed_re, webpage):
 373                 urls.append(mobj.group('url'))
 374         return urls
 375
 376     @staticmethod
 377     def _extract_url(url, webpage):
 378         urls = VimeoIE._extract_urls(url, webpage)
 379         return urls[0] if urls else None
 380
 381     def _verify_player_video_password(self, url, video_id):
 382         password = self._downloader.params.get('videopassword')
 383         if password is None:
 384             raise ExtractorError('This video is protected by a password, use the --video-password option')
 385         data = urlencode_postdata({'password': password})
 386         pass_url = url + '/check-password'
 387         password_request = sanitized_Request(pass_url, data)
 388         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 389         password_request.add_header('Referer', url)
 390         return self._download_json(
 391             password_request, video_id,
 392             'Verifying the password', 'Wrong password')
 393
 394     def _real_initialize(self):
 395         self._login()
 396
 397     def _real_extract(self, url):
 398         url, data = unsmuggle_url(url, {})
 399         headers = std_headers.copy()
 400         if 'http_headers' in data:
 401             headers.update(data['http_headers'])
 402         if 'Referer' not in headers:
 403             headers['Referer'] = url
 404
 405         # Extract ID from URL
 406         mobj = re.match(self._VALID_URL, url)
 407         video_id = mobj.group('id')
 408         orig_url = url
 409         if mobj.group('pro') or mobj.group('player'):
 410             url = 'https://player.vimeo.com/video/' + video_id
 411         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 412             url = 'https://vimeo.com/' + video_id
 413
 414         # Retrieve video webpage to extract further information
 415         request = sanitized_Request(url, headers=headers)
 416         try:
 417             webpage = self._download_webpage(request, video_id)
 418         except ExtractorError as ee:
 419             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 420                 errmsg = ee.cause.read()
 421                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 422                     raise ExtractorError(
 423                         'Cannot download embed-only video without embedding '
 424                         'URL. Please call youtube-dl with the URL of the page '
 425                         'that embeds this video.',
 426                         expected=True)
 427             raise
 428
 429         # Now we begin extracting as much information as we can from what we
 430         # retrieved. First we extract the information common to all extractors,
 431         # and latter we extract those that are Vimeo specific.
 432         self.report_extraction(video_id)
 433
 434         vimeo_config = self._search_regex(
 435             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 436             'vimeo config', default=None)
 437         if vimeo_config:
 438             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 439             if seed_status.get('state') == 'failed':
 440                 raise ExtractorError(
 441                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 442                     expected=True)
 443
 444         # Extract the config JSON
 445         try:
 446             try:
 447                 config_url = self._html_search_regex(
 448                     r' data-config-url="(.+?)"', webpage,
 449                     'config URL', default=None)
 450                 if not config_url:
 451                     # Sometimes new react-based page is served instead of old one that require
 452                     # different config URL extraction approach (see
 453                     # https://github.com/rg3/youtube-dl/pull/7209)
 454                     vimeo_clip_page_config = self._search_regex(
 455                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 456                         'vimeo clip page config')
 457                     config_url = self._parse_json(
 458                         vimeo_clip_page_config, video_id)['player']['config_url']
 459                 config_json = self._download_webpage(config_url, video_id)
 460                 config = json.loads(config_json)
 461             except RegexNotFoundError:
 462                 # For pro videos or player.vimeo.com urls
 463                 # We try to find out to which variable is assigned the config dic
 464                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 465                 if m_variable_name is not None:
 466                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 467                 else:
 468                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 469                 config = self._search_regex(config_re, webpage, 'info section',
 470                                             flags=re.DOTALL)
 471                 config = json.loads(config)
 472         except Exception as e:
 473             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 474                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 475
 476             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 477                 if '_video_password_verified' in data:
 478                     raise ExtractorError('video password verification failed!')
 479                 self._verify_video_password(url, video_id, webpage)
 480                 return self._real_extract(
 481                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 482             else:
 483                 raise ExtractorError('Unable to extract info section',
 484                                      cause=e)
 485         else:
 486             if config.get('view') == 4:
 487                 config = self._verify_player_video_password(url, video_id)
 488
 489         def is_rented():
 490             if '>You rented this title.<' in webpage:
 491                 return True
 492             if config.get('user', {}).get('purchased'):
 493                 return True
 494             label = try_get(
 495                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 496             if label and label.startswith('You rented this'):
 497                 return True
 498             return False
 499
 500         if is_rented():
 501             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 502             if feature_id and not data.get('force_feature_id', False):
 503                 return self.url_result(smuggle_url(
 504                     'https://player.vimeo.com/player/%s' % feature_id,
 505                     {'force_feature_id': True}), 'Vimeo')
 506
 507         # Extract video description
 508
 509         video_description = self._html_search_regex(
 510             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 511             webpage, 'description', default=None)
 512         if not video_description:
 513             video_description = self._html_search_meta(
 514                 'description', webpage, default=None)
 515         if not video_description and mobj.group('pro'):
 516             orig_webpage = self._download_webpage(
 517                 orig_url, video_id,
 518                 note='Downloading webpage for description',
 519                 fatal=False)
 520             if orig_webpage:
 521                 video_description = self._html_search_meta(
 522                     'description', orig_webpage, default=None)
 523         if not video_description and not mobj.group('player'):
 524             self._downloader.report_warning('Cannot find video description')
 525
 526         # Extract upload date
 527         video_upload_date = None
 528         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 529         if mobj is not None:
 530             video_upload_date = unified_strdate(mobj.group(1))
 531
 532         try:
 533             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 534             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 535             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 536         except RegexNotFoundError:
 537             # This info is only available in vimeo.com/{id} urls
 538             view_count = None
 539             like_count = None
 540             comment_count = None
 541
 542         formats = []
 543         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 544             'X-Requested-With': 'XMLHttpRequest'})
 545         download_data = self._download_json(download_request, video_id, fatal=False)
 546         if download_data:
 547             source_file = download_data.get('source_file')
 548             if isinstance(source_file, dict):
 549                 download_url = source_file.get('download_url')
 550                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 551                     source_name = source_file.get('public_name', 'Original')
 552                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 553                         ext = source_file.get('extension', determine_ext(download_url)).lower()
 554                         formats.append({
 555                             'url': download_url,
 556                             'ext': ext,
 557                             'width': int_or_none(source_file.get('width')),
 558                             'height': int_or_none(source_file.get('height')),
 559                             'filesize': parse_filesize(source_file.get('size')),
 560                             'format_id': source_name,
 561                             'preference': 1,
 562                         })
 563
 564         info_dict = self._parse_config(config, video_id)
 565         formats.extend(info_dict['formats'])
 566         self._vimeo_sort_formats(formats)
 567         info_dict.update({
 568             'id': video_id,
 569             'formats': formats,
 570             'upload_date': video_upload_date,
 571             'description': video_description,
 572             'webpage_url': url,
 573             'view_count': view_count,
 574             'like_count': like_count,
 575             'comment_count': comment_count,
 576         })
 577
 578         return info_dict
 579
 580
 581 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 582     IE_NAME = 'vimeo:ondemand'
 583     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 584     _TESTS = [{
 585         # ondemand video not available via https://vimeo.com/id
 586         'url': 'https://vimeo.com/ondemand/20704',
 587         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 588         'info_dict': {
 589             'id': '105442900',
 590             'ext': 'mp4',
 591             'title': 'המעבדה - במאי יותם פלדמן',
 592             'uploader': 'גם סרטים',
 593             'uploader_url': 're:https?://(?:www\.)?vimeo\.com/gumfilms',
 594             'uploader_id': 'gumfilms',
 595         },
 596     }, {
 597         # requires Referer to be passed along with og:video:url
 598         'url': 'https://vimeo.com/ondemand/36938/126682985',
 599         'info_dict': {
 600             'id': '126682985',
 601             'ext': 'mp4',
 602             'title': 'Rävlock, rätt läte på rätt plats',
 603             'uploader': 'Lindroth & Norin',
 604             'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user14430847',
 605             'uploader_id': 'user14430847',
 606         },
 607         'params': {
 608             'skip_download': True,
 609         },
 610     }, {
 611         'url': 'https://vimeo.com/ondemand/nazmaalik',
 612         'only_matching': True,
 613     }, {
 614         'url': 'https://vimeo.com/ondemand/141692381',
 615         'only_matching': True,
 616     }, {
 617         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 618         'only_matching': True,
 619     }]
 620
 621     def _real_extract(self, url):
 622         video_id = self._match_id(url)
 623         webpage = self._download_webpage(url, video_id)
 624         return self.url_result(
 625             # Some videos require Referer to be passed along with og:video:url
 626             # similarly to generic vimeo embeds (e.g.
 627             # https://vimeo.com/ondemand/36938/126682985).
 628             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 629             VimeoIE.ie_key())
 630
 631
 632 class VimeoChannelIE(VimeoBaseInfoExtractor):
 633     IE_NAME = 'vimeo:channel'
 634     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 635     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 636     _TITLE = None
 637     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 638     _TESTS = [{
 639         'url': 'https://vimeo.com/channels/tributes',
 640         'info_dict': {
 641             'id': 'tributes',
 642             'title': 'Vimeo Tributes',
 643         },
 644         'playlist_mincount': 25,
 645     }]
 646
 647     def _page_url(self, base_url, pagenum):
 648         return '%s/videos/page:%d/' % (base_url, pagenum)
 649
 650     def _extract_list_title(self, webpage):
 651         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 652
 653     def _login_list_password(self, page_url, list_id, webpage):
 654         login_form = self._search_regex(
 655             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 656             webpage, 'login form', default=None)
 657         if not login_form:
 658             return webpage
 659
 660         password = self._downloader.params.get('videopassword')
 661         if password is None:
 662             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 663         fields = self._hidden_inputs(login_form)
 664         token, vuid = self._extract_xsrft_and_vuid(webpage)
 665         fields['token'] = token
 666         fields['password'] = password
 667         post = urlencode_postdata(fields)
 668         password_path = self._search_regex(
 669             r'action="([^"]+)"', login_form, 'password URL')
 670         password_url = compat_urlparse.urljoin(page_url, password_path)
 671         password_request = sanitized_Request(password_url, post)
 672         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 673         self._set_vimeo_cookie('vuid', vuid)
 674         self._set_vimeo_cookie('xsrft', token)
 675
 676         return self._download_webpage(
 677             password_request, list_id,
 678             'Verifying the password', 'Wrong password')
 679
 680     def _title_and_entries(self, list_id, base_url):
 681         for pagenum in itertools.count(1):
 682             page_url = self._page_url(base_url, pagenum)
 683             webpage = self._download_webpage(
 684                 page_url, list_id,
 685                 'Downloading page %s' % pagenum)
 686
 687             if pagenum == 1:
 688                 webpage = self._login_list_password(page_url, list_id, webpage)
 689                 yield self._extract_list_title(webpage)
 690
 691             # Try extracting href first since not all videos are available via
 692             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 693             clips = re.findall(
 694                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)', webpage)
 695             if clips:
 696                 for video_id, video_url in clips:
 697                     yield self.url_result(
 698                         compat_urlparse.urljoin(base_url, video_url),
 699                         VimeoIE.ie_key(), video_id=video_id)
 700             # More relaxed fallback
 701             else:
 702                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 703                     yield self.url_result(
 704                         'https://vimeo.com/%s' % video_id,
 705                         VimeoIE.ie_key(), video_id=video_id)
 706
 707             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 708                 break
 709
 710     def _extract_videos(self, list_id, base_url):
 711         title_and_entries = self._title_and_entries(list_id, base_url)
 712         list_title = next(title_and_entries)
 713         return self.playlist_result(title_and_entries, list_id, list_title)
 714
 715     def _real_extract(self, url):
 716         mobj = re.match(self._VALID_URL, url)
 717         channel_id = mobj.group('id')
 718         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 719
 720
 721 class VimeoUserIE(VimeoChannelIE):
 722     IE_NAME = 'vimeo:user'
 723     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 724     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 725     _TESTS = [{
 726         'url': 'https://vimeo.com/nkistudio/videos',
 727         'info_dict': {
 728             'title': 'Nki',
 729             'id': 'nkistudio',
 730         },
 731         'playlist_mincount': 66,
 732     }]
 733
 734     def _real_extract(self, url):
 735         mobj = re.match(self._VALID_URL, url)
 736         name = mobj.group('name')
 737         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 738
 739
 740 class VimeoAlbumIE(VimeoChannelIE):
 741     IE_NAME = 'vimeo:album'
 742     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 743     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 744     _TESTS = [{
 745         'url': 'https://vimeo.com/album/2632481',
 746         'info_dict': {
 747             'id': '2632481',
 748             'title': 'Staff Favorites: November 2013',
 749         },
 750         'playlist_mincount': 13,
 751     }, {
 752         'note': 'Password-protected album',
 753         'url': 'https://vimeo.com/album/3253534',
 754         'info_dict': {
 755             'title': 'test',
 756             'id': '3253534',
 757         },
 758         'playlist_count': 1,
 759         'params': {
 760             'videopassword': 'youtube-dl',
 761         }
 762     }, {
 763         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 764         'only_matching': True,
 765     }, {
 766         # TODO: respect page number
 767         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 768         'only_matching': True,
 769     }]
 770
 771     def _page_url(self, base_url, pagenum):
 772         return '%s/page:%d/' % (base_url, pagenum)
 773
 774     def _real_extract(self, url):
 775         album_id = self._match_id(url)
 776         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 777
 778
 779 class VimeoGroupsIE(VimeoAlbumIE):
 780     IE_NAME = 'vimeo:group'
 781     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 782     _TESTS = [{
 783         'url': 'https://vimeo.com/groups/rolexawards',
 784         'info_dict': {
 785             'id': 'rolexawards',
 786             'title': 'Rolex Awards for Enterprise',
 787         },
 788         'playlist_mincount': 73,
 789     }]
 790
 791     def _extract_list_title(self, webpage):
 792         return self._og_search_title(webpage)
 793
 794     def _real_extract(self, url):
 795         mobj = re.match(self._VALID_URL, url)
 796         name = mobj.group('name')
 797         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 798
 799
 800 class VimeoReviewIE(VimeoBaseInfoExtractor):
 801     IE_NAME = 'vimeo:review'
 802     IE_DESC = 'Review pages on vimeo'
 803     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 804     _TESTS = [{
 805         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 806         'md5': 'c507a72f780cacc12b2248bb4006d253',
 807         'info_dict': {
 808             'id': '75524534',
 809             'ext': 'mp4',
 810             'title': "DICK HARDWICK 'Comedian'",
 811             'uploader': 'Richard Hardwick',
 812             'uploader_id': 'user21297594',
 813         }
 814     }, {
 815         'note': 'video player needs Referer',
 816         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 817         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 818         'info_dict': {
 819             'id': '91613211',
 820             'ext': 'mp4',
 821             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 822             'uploader': 'DevWeek Events',
 823             'duration': 2773,
 824             'thumbnail': 're:^https?://.*\.jpg$',
 825             'uploader_id': 'user22258446',
 826         }
 827     }, {
 828         'note': 'Password protected',
 829         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 830         'info_dict': {
 831             'id': '138823582',
 832             'ext': 'mp4',
 833             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 834             'uploader': 'TMB',
 835             'uploader_id': 'user37284429',
 836         },
 837         'params': {
 838             'videopassword': 'holygrail',
 839         },
 840         'skip': 'video gone',
 841     }]
 842
 843     def _real_initialize(self):
 844         self._login()
 845
 846     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 847         webpage = self._download_webpage(webpage_url, video_id)
 848         data = self._parse_json(self._search_regex(
 849             r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 850             default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 851         config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 852         if config_url is None:
 853             self._verify_video_password(webpage_url, video_id, webpage)
 854             config_url = self._get_config_url(
 855                 webpage_url, video_id, video_password_verified=True)
 856         return config_url
 857
 858     def _real_extract(self, url):
 859         video_id = self._match_id(url)
 860         config_url = self._get_config_url(url, video_id)
 861         config = self._download_json(config_url, video_id)
 862         info_dict = self._parse_config(config, video_id)
 863         self._vimeo_sort_formats(info_dict['formats'])
 864         info_dict['id'] = video_id
 865         return info_dict
 866
 867
 868 class VimeoWatchLaterIE(VimeoChannelIE):
 869     IE_NAME = 'vimeo:watchlater'
 870     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 871     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 872     _TITLE = 'Watch Later'
 873     _LOGIN_REQUIRED = True
 874     _TESTS = [{
 875         'url': 'https://vimeo.com/watchlater',
 876         'only_matching': True,
 877     }]
 878
 879     def _real_initialize(self):
 880         self._login()
 881
 882     def _page_url(self, base_url, pagenum):
 883         url = '%s/page:%d/' % (base_url, pagenum)
 884         request = sanitized_Request(url)
 885         # Set the header to get a partial html page with the ids,
 886         # the normal page doesn't contain them.
 887         request.add_header('X-Requested-With', 'XMLHttpRequest')
 888         return request
 889
 890     def _real_extract(self, url):
 891         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 892
 893
 894 class VimeoLikesIE(InfoExtractor):
 895     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 896     IE_NAME = 'vimeo:likes'
 897     IE_DESC = 'Vimeo user likes'
 898     _TEST = {
 899         'url': 'https://vimeo.com/user755559/likes/',
 900         'playlist_mincount': 293,
 901         'info_dict': {
 902             'id': 'user755559_likes',
 903             'description': 'See all the videos urza likes',
 904             'title': 'Videos urza likes',
 905         },
 906     }
 907
 908     def _real_extract(self, url):
 909         user_id = self._match_id(url)
 910         webpage = self._download_webpage(url, user_id)
 911         page_count = self._int(
 912             self._search_regex(
 913                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 914                     .*?</a></li>\s*<li\s+class="pagination_next">
 915                 ''', webpage, 'page count'),
 916             'page count', fatal=True)
 917         PAGE_SIZE = 12
 918         title = self._html_search_regex(
 919             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 920         description = self._html_search_meta('description', webpage)
 921
 922         def _get_page(idx):
 923             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 924                 user_id, idx + 1)
 925             webpage = self._download_webpage(
 926                 page_url, user_id,
 927                 note='Downloading page %d/%d' % (idx + 1, page_count))
 928             video_list = self._search_regex(
 929                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 930                 webpage, 'video content')
 931             paths = re.findall(
 932                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 933             for path in paths:
 934                 yield {
 935                     '_type': 'url',
 936                     'url': compat_urlparse.urljoin(page_url, path),
 937                 }
 938
 939         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 940
 941         return {
 942             '_type': 'playlist',
 943             'id': 'user%s_likes' % user_id,
 944             'title': title,
 945             'description': description,
 946             'entries': pl,
 947         }