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', 'tbr', 'format_id'))
  96
  97     def _parse_config(self, config, video_id):
  98         video_data = config['video']
  99         # Extract title
 100         video_title = video_data['title']
 101
 102         # Extract uploader, uploader_url and uploader_id
 103         video_uploader = video_data.get('owner', {}).get('name')
 104         video_uploader_url = video_data.get('owner', {}).get('url')
 105         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 106
 107         # Extract video thumbnail
 108         video_thumbnail = video_data.get('thumbnail')
 109         if video_thumbnail is None:
 110             video_thumbs = video_data.get('thumbs')
 111             if video_thumbs and isinstance(video_thumbs, dict):
 112                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 113
 114         # Extract video duration
 115         video_duration = int_or_none(video_data.get('duration'))
 116
 117         formats = []
 118         config_files = video_data.get('files') or config['request'].get('files', {})
 119         for f in config_files.get('progressive', []):
 120             video_url = f.get('url')
 121             if not video_url:
 122                 continue
 123             formats.append({
 124                 'url': video_url,
 125                 'format_id': 'http-%s' % f.get('quality'),
 126                 'width': int_or_none(f.get('width')),
 127                 'height': int_or_none(f.get('height')),
 128                 'fps': int_or_none(f.get('fps')),
 129                 'tbr': int_or_none(f.get('bitrate')),
 130             })
 131
 132         for files_type in ('hls', 'dash'):
 133             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 134                 manifest_url = cdn_data.get('url')
 135                 if not manifest_url:
 136                     continue
 137                 format_id = '%s-%s' % (files_type, cdn_name)
 138                 if files_type == 'hls':
 139                     formats.extend(self._extract_m3u8_formats(
 140                         manifest_url, video_id, 'mp4',
 141                         'm3u8_native', m3u8_id=format_id,
 142                         note='Downloading %s m3u8 information' % cdn_name,
 143                         fatal=False))
 144                 elif files_type == 'dash':
 145                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 146                     mpd_manifest_urls = []
 147                     if re.search(mpd_pattern, manifest_url):
 148                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 149                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 150                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 151                     else:
 152                         mpd_manifest_urls = [(format_id, manifest_url)]
 153                     for f_id, m_url in mpd_manifest_urls:
 154                         formats.extend(self._extract_mpd_formats(
 155                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 156                             'Downloading %s MPD information' % cdn_name,
 157                             fatal=False))
 158
 159         subtitles = {}
 160         text_tracks = config['request'].get('text_tracks')
 161         if text_tracks:
 162             for tt in text_tracks:
 163                 subtitles[tt['lang']] = [{
 164                     'ext': 'vtt',
 165                     'url': 'https://vimeo.com' + tt['url'],
 166                 }]
 167
 168         return {
 169             'title': video_title,
 170             'uploader': video_uploader,
 171             'uploader_id': video_uploader_id,
 172             'uploader_url': video_uploader_url,
 173             'thumbnail': video_thumbnail,
 174             'duration': video_duration,
 175             'formats': formats,
 176             'subtitles': subtitles,
 177         }
 178
 179
 180 class VimeoIE(VimeoBaseInfoExtractor):
 181     """Information extractor for vimeo.com."""
 182
 183     # _VALID_URL matches Vimeo URLs
 184     _VALID_URL = r'''(?x)
 185                     https?://
 186                         (?:
 187                             (?:
 188                                 www|
 189                                 (?P<player>player)
 190                             )
 191                             \.
 192                         )?
 193                         vimeo(?P<pro>pro)?\.com/
 194                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 195                         (?:.*?/)?
 196                         (?:
 197                             (?:
 198                                 play_redirect_hls|
 199                                 moogaloop\.swf)\?clip_id=
 200                             )?
 201                         (?:videos?/)?
 202                         (?P<id>[0-9]+)
 203                         (?:/[\da-f]+)?
 204                         /?(?:[?&].*)?(?:[#].*)?$
 205                     '''
 206     IE_NAME = 'vimeo'
 207     _TESTS = [
 208         {
 209             'url': 'http://vimeo.com/56015672#at=0',
 210             'md5': '8879b6cc097e987f02484baf890129e5',
 211             'info_dict': {
 212                 'id': '56015672',
 213                 'ext': 'mp4',
 214                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 215                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 216                 'upload_date': '20121220',
 217                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 218                 'uploader_id': 'user7108434',
 219                 'uploader': 'Filippo Valsorda',
 220                 'duration': 10,
 221                 'license': 'by-sa',
 222             },
 223         },
 224         {
 225             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 226             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 227             'note': 'Vimeo Pro video (#1197)',
 228             'info_dict': {
 229                 'id': '68093876',
 230                 'ext': 'mp4',
 231                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 232                 'uploader_id': 'openstreetmapus',
 233                 'uploader': 'OpenStreetMap US',
 234                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 235                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 236                 'duration': 1595,
 237             },
 238         },
 239         {
 240             'url': 'http://player.vimeo.com/video/54469442',
 241             'md5': '619b811a4417aa4abe78dc653becf511',
 242             'note': 'Videos that embed the url in the player page',
 243             'info_dict': {
 244                 'id': '54469442',
 245                 'ext': 'mp4',
 246                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 247                 'uploader': 'The BLN & Business of Software',
 248                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 249                 'uploader_id': 'theblnbusinessofsoftware',
 250                 'duration': 3610,
 251                 'description': None,
 252             },
 253         },
 254         {
 255             'url': 'http://vimeo.com/68375962',
 256             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 257             'note': 'Video protected with password',
 258             'info_dict': {
 259                 'id': '68375962',
 260                 'ext': 'mp4',
 261                 'title': 'youtube-dl password protected test video',
 262                 'upload_date': '20130614',
 263                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 264                 'uploader_id': 'user18948128',
 265                 'uploader': 'Jaime Marquínez Ferrándiz',
 266                 'duration': 10,
 267                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 268             },
 269             'params': {
 270                 'videopassword': 'youtube-dl',
 271             },
 272         },
 273         {
 274             'url': 'http://vimeo.com/channels/keypeele/75629013',
 275             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 276             'info_dict': {
 277                 'id': '75629013',
 278                 'ext': 'mp4',
 279                 'title': 'Key & Peele: Terrorist Interrogation',
 280                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 281                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 282                 'uploader_id': 'atencio',
 283                 'uploader': 'Peter Atencio',
 284                 'upload_date': '20130927',
 285                 'duration': 187,
 286             },
 287         },
 288         {
 289             'url': 'http://vimeo.com/76979871',
 290             'note': 'Video with subtitles',
 291             'info_dict': {
 292                 'id': '76979871',
 293                 'ext': 'mp4',
 294                 'title': 'The New Vimeo Player (You Know, For Videos)',
 295                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 296                 'upload_date': '20131015',
 297                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 298                 'uploader_id': 'staff',
 299                 'uploader': 'Vimeo Staff',
 300                 'duration': 62,
 301             }
 302         },
 303         {
 304             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 305             'url': 'https://player.vimeo.com/video/98044508',
 306             'note': 'The js code contains assignments to the same variable as the config',
 307             'info_dict': {
 308                 'id': '98044508',
 309                 'ext': 'mp4',
 310                 'title': 'Pier Solar OUYA Official Trailer',
 311                 'uploader': 'Tulio Gonçalves',
 312                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 313                 'uploader_id': 'user28849593',
 314             },
 315         },
 316         {
 317             # contains original format
 318             'url': 'https://vimeo.com/33951933',
 319             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 320             'info_dict': {
 321                 'id': '33951933',
 322                 'ext': 'mp4',
 323                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 324                 'uploader': 'The DMCI',
 325                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 326                 'uploader_id': 'dmci',
 327                 'upload_date': '20111220',
 328                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 329             },
 330         },
 331         {
 332             # only available via https://vimeo.com/channels/tributes/6213729 and
 333             # not via https://vimeo.com/6213729
 334             'url': 'https://vimeo.com/channels/tributes/6213729',
 335             'info_dict': {
 336                 'id': '6213729',
 337                 'ext': 'mov',
 338                 'title': 'Vimeo Tribute: The Shining',
 339                 'uploader': 'Casey Donahue',
 340                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 341                 'uploader_id': 'caseydonahue',
 342                 'upload_date': '20090821',
 343                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 344             },
 345             'params': {
 346                 'skip_download': True,
 347             },
 348             'expected_warnings': ['Unable to download JSON metadata'],
 349         },
 350         {
 351             # redirects to ondemand extractor and should be passed through it
 352             # for successful extraction
 353             'url': 'https://vimeo.com/73445910',
 354             'info_dict': {
 355                 'id': '73445910',
 356                 'ext': 'mp4',
 357                 'title': 'The Reluctant Revolutionary',
 358                 'uploader': '10Ft Films',
 359                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 360                 'uploader_id': 'tenfootfilms',
 361             },
 362             'params': {
 363                 'skip_download': True,
 364             },
 365         },
 366         {
 367             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 368             'only_matching': True,
 369         },
 370         {
 371             'url': 'https://vimeo.com/109815029',
 372             'note': 'Video not completely processed, "failed" seed status',
 373             'only_matching': True,
 374         },
 375         {
 376             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 377             'only_matching': True,
 378         },
 379         {
 380             'url': 'https://vimeo.com/album/2632481/video/79010983',
 381             'only_matching': True,
 382         },
 383         {
 384             # source file returns 403: Forbidden
 385             'url': 'https://vimeo.com/7809605',
 386             'only_matching': True,
 387         },
 388         {
 389             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 390             'only_matching': True,
 391         }
 392     ]
 393
 394     @staticmethod
 395     def _smuggle_referrer(url, referrer_url):
 396         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 397
 398     @staticmethod
 399     def _extract_urls(url, webpage):
 400         urls = []
 401         # Look for embedded (iframe) Vimeo player
 402         for mobj in re.finditer(
 403                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1',
 404                 webpage):
 405             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 406         PLAIN_EMBED_RE = (
 407             # Look for embedded (swf embed) Vimeo player
 408             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 409             # Look more for non-standard embedded Vimeo player
 410             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 411         )
 412         for embed_re in PLAIN_EMBED_RE:
 413             for mobj in re.finditer(embed_re, webpage):
 414                 urls.append(mobj.group('url'))
 415         return urls
 416
 417     @staticmethod
 418     def _extract_url(url, webpage):
 419         urls = VimeoIE._extract_urls(url, webpage)
 420         return urls[0] if urls else None
 421
 422     def _verify_player_video_password(self, url, video_id):
 423         password = self._downloader.params.get('videopassword')
 424         if password is None:
 425             raise ExtractorError('This video is protected by a password, use the --video-password option')
 426         data = urlencode_postdata({'password': password})
 427         pass_url = url + '/check-password'
 428         password_request = sanitized_Request(pass_url, data)
 429         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 430         password_request.add_header('Referer', url)
 431         return self._download_json(
 432             password_request, video_id,
 433             'Verifying the password', 'Wrong password')
 434
 435     def _real_initialize(self):
 436         self._login()
 437
 438     def _real_extract(self, url):
 439         url, data = unsmuggle_url(url, {})
 440         headers = std_headers.copy()
 441         if 'http_headers' in data:
 442             headers.update(data['http_headers'])
 443         if 'Referer' not in headers:
 444             headers['Referer'] = url
 445
 446         # Extract ID from URL
 447         mobj = re.match(self._VALID_URL, url)
 448         video_id = mobj.group('id')
 449         orig_url = url
 450         if mobj.group('pro') or mobj.group('player'):
 451             url = 'https://player.vimeo.com/video/' + video_id
 452         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 453             url = 'https://vimeo.com/' + video_id
 454
 455         # Retrieve video webpage to extract further information
 456         request = sanitized_Request(url, headers=headers)
 457         try:
 458             webpage, urlh = self._download_webpage_handle(request, video_id)
 459             # Some URLs redirect to ondemand can't be extracted with
 460             # this extractor right away thus should be passed through
 461             # ondemand extractor (e.g. https://vimeo.com/73445910)
 462             if VimeoOndemandIE.suitable(urlh.geturl()):
 463                 return self.url_result(urlh.geturl(), VimeoOndemandIE.ie_key())
 464         except ExtractorError as ee:
 465             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 466                 errmsg = ee.cause.read()
 467                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 468                     raise ExtractorError(
 469                         'Cannot download embed-only video without embedding '
 470                         'URL. Please call youtube-dl with the URL of the page '
 471                         'that embeds this video.',
 472                         expected=True)
 473             raise
 474
 475         # Now we begin extracting as much information as we can from what we
 476         # retrieved. First we extract the information common to all extractors,
 477         # and latter we extract those that are Vimeo specific.
 478         self.report_extraction(video_id)
 479
 480         vimeo_config = self._search_regex(
 481             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 482             'vimeo config', default=None)
 483         if vimeo_config:
 484             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 485             if seed_status.get('state') == 'failed':
 486                 raise ExtractorError(
 487                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 488                     expected=True)
 489
 490         cc_license = None
 491
 492         # Extract the config JSON
 493         try:
 494             try:
 495                 config_url = self._html_search_regex(
 496                     r' data-config-url="(.+?)"', webpage,
 497                     'config URL', default=None)
 498                 if not config_url:
 499                     # Sometimes new react-based page is served instead of old one that require
 500                     # different config URL extraction approach (see
 501                     # https://github.com/rg3/youtube-dl/pull/7209)
 502                     vimeo_clip_page_config = self._search_regex(
 503                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 504                         'vimeo clip page config')
 505                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 506                     config_url = page_config['player']['config_url']
 507                     cc_license = page_config.get('cc_license')
 508                 config_json = self._download_webpage(config_url, video_id)
 509                 config = json.loads(config_json)
 510             except RegexNotFoundError:
 511                 # For pro videos or player.vimeo.com urls
 512                 # We try to find out to which variable is assigned the config dic
 513                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 514                 if m_variable_name is not None:
 515                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 516                 else:
 517                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 518                 config = self._search_regex(config_re, webpage, 'info section',
 519                                             flags=re.DOTALL)
 520                 config = json.loads(config)
 521         except Exception as e:
 522             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 523                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 524
 525             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 526                 if '_video_password_verified' in data:
 527                     raise ExtractorError('video password verification failed!')
 528                 self._verify_video_password(url, video_id, webpage)
 529                 return self._real_extract(
 530                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 531             else:
 532                 raise ExtractorError('Unable to extract info section',
 533                                      cause=e)
 534         else:
 535             if config.get('view') == 4:
 536                 config = self._verify_player_video_password(url, video_id)
 537
 538         def is_rented():
 539             if '>You rented this title.<' in webpage:
 540                 return True
 541             if config.get('user', {}).get('purchased'):
 542                 return True
 543             label = try_get(
 544                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 545             if label and label.startswith('You rented this'):
 546                 return True
 547             return False
 548
 549         if is_rented():
 550             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 551             if feature_id and not data.get('force_feature_id', False):
 552                 return self.url_result(smuggle_url(
 553                     'https://player.vimeo.com/player/%s' % feature_id,
 554                     {'force_feature_id': True}), 'Vimeo')
 555
 556         # Extract video description
 557
 558         video_description = self._html_search_regex(
 559             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 560             webpage, 'description', default=None)
 561         if not video_description:
 562             video_description = self._html_search_meta(
 563                 'description', webpage, default=None)
 564         if not video_description and mobj.group('pro'):
 565             orig_webpage = self._download_webpage(
 566                 orig_url, video_id,
 567                 note='Downloading webpage for description',
 568                 fatal=False)
 569             if orig_webpage:
 570                 video_description = self._html_search_meta(
 571                     'description', orig_webpage, default=None)
 572         if not video_description and not mobj.group('player'):
 573             self._downloader.report_warning('Cannot find video description')
 574
 575         # Extract upload date
 576         video_upload_date = None
 577         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 578         if mobj is not None:
 579             video_upload_date = unified_strdate(mobj.group(1))
 580
 581         try:
 582             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 583             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 584             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 585         except RegexNotFoundError:
 586             # This info is only available in vimeo.com/{id} urls
 587             view_count = None
 588             like_count = None
 589             comment_count = None
 590
 591         formats = []
 592         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 593             'X-Requested-With': 'XMLHttpRequest'})
 594         download_data = self._download_json(download_request, video_id, fatal=False)
 595         if download_data:
 596             source_file = download_data.get('source_file')
 597             if isinstance(source_file, dict):
 598                 download_url = source_file.get('download_url')
 599                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 600                     source_name = source_file.get('public_name', 'Original')
 601                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 602                         ext = source_file.get('extension', determine_ext(download_url)).lower()
 603                         formats.append({
 604                             'url': download_url,
 605                             'ext': ext,
 606                             'width': int_or_none(source_file.get('width')),
 607                             'height': int_or_none(source_file.get('height')),
 608                             'filesize': parse_filesize(source_file.get('size')),
 609                             'format_id': source_name,
 610                             'preference': 1,
 611                         })
 612
 613         info_dict = self._parse_config(config, video_id)
 614         formats.extend(info_dict['formats'])
 615         self._vimeo_sort_formats(formats)
 616
 617         if not cc_license:
 618             cc_license = self._search_regex(
 619                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 620                 webpage, 'license', default=None, group='license')
 621
 622         info_dict.update({
 623             'id': video_id,
 624             'formats': formats,
 625             'upload_date': video_upload_date,
 626             'description': video_description,
 627             'webpage_url': url,
 628             'view_count': view_count,
 629             'like_count': like_count,
 630             'comment_count': comment_count,
 631             'license': cc_license,
 632         })
 633
 634         return info_dict
 635
 636
 637 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 638     IE_NAME = 'vimeo:ondemand'
 639     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 640     _TESTS = [{
 641         # ondemand video not available via https://vimeo.com/id
 642         'url': 'https://vimeo.com/ondemand/20704',
 643         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 644         'info_dict': {
 645             'id': '105442900',
 646             'ext': 'mp4',
 647             'title': 'המעבדה - במאי יותם פלדמן',
 648             'uploader': 'גם סרטים',
 649             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 650             'uploader_id': 'gumfilms',
 651         },
 652         'params': {
 653             'format': 'best[protocol=https]',
 654         },
 655     }, {
 656         # requires Referer to be passed along with og:video:url
 657         'url': 'https://vimeo.com/ondemand/36938/126682985',
 658         'info_dict': {
 659             'id': '126682985',
 660             'ext': 'mp4',
 661             'title': 'Rävlock, rätt läte på rätt plats',
 662             'uploader': 'Lindroth & Norin',
 663             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 664             'uploader_id': 'user14430847',
 665         },
 666         'params': {
 667             'skip_download': True,
 668         },
 669     }, {
 670         'url': 'https://vimeo.com/ondemand/nazmaalik',
 671         'only_matching': True,
 672     }, {
 673         'url': 'https://vimeo.com/ondemand/141692381',
 674         'only_matching': True,
 675     }, {
 676         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 677         'only_matching': True,
 678     }]
 679
 680     def _real_extract(self, url):
 681         video_id = self._match_id(url)
 682         webpage = self._download_webpage(url, video_id)
 683         return self.url_result(
 684             # Some videos require Referer to be passed along with og:video:url
 685             # similarly to generic vimeo embeds (e.g.
 686             # https://vimeo.com/ondemand/36938/126682985).
 687             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 688             VimeoIE.ie_key())
 689
 690
 691 class VimeoChannelIE(VimeoBaseInfoExtractor):
 692     IE_NAME = 'vimeo:channel'
 693     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 694     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 695     _TITLE = None
 696     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 697     _TESTS = [{
 698         'url': 'https://vimeo.com/channels/tributes',
 699         'info_dict': {
 700             'id': 'tributes',
 701             'title': 'Vimeo Tributes',
 702         },
 703         'playlist_mincount': 25,
 704     }]
 705
 706     def _page_url(self, base_url, pagenum):
 707         return '%s/videos/page:%d/' % (base_url, pagenum)
 708
 709     def _extract_list_title(self, webpage):
 710         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 711
 712     def _login_list_password(self, page_url, list_id, webpage):
 713         login_form = self._search_regex(
 714             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 715             webpage, 'login form', default=None)
 716         if not login_form:
 717             return webpage
 718
 719         password = self._downloader.params.get('videopassword')
 720         if password is None:
 721             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 722         fields = self._hidden_inputs(login_form)
 723         token, vuid = self._extract_xsrft_and_vuid(webpage)
 724         fields['token'] = token
 725         fields['password'] = password
 726         post = urlencode_postdata(fields)
 727         password_path = self._search_regex(
 728             r'action="([^"]+)"', login_form, 'password URL')
 729         password_url = compat_urlparse.urljoin(page_url, password_path)
 730         password_request = sanitized_Request(password_url, post)
 731         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 732         self._set_vimeo_cookie('vuid', vuid)
 733         self._set_vimeo_cookie('xsrft', token)
 734
 735         return self._download_webpage(
 736             password_request, list_id,
 737             'Verifying the password', 'Wrong password')
 738
 739     def _title_and_entries(self, list_id, base_url):
 740         for pagenum in itertools.count(1):
 741             page_url = self._page_url(base_url, pagenum)
 742             webpage = self._download_webpage(
 743                 page_url, list_id,
 744                 'Downloading page %s' % pagenum)
 745
 746             if pagenum == 1:
 747                 webpage = self._login_list_password(page_url, list_id, webpage)
 748                 yield self._extract_list_title(webpage)
 749
 750             # Try extracting href first since not all videos are available via
 751             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 752             clips = re.findall(
 753                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 754             if clips:
 755                 for video_id, video_url, video_title in clips:
 756                     yield self.url_result(
 757                         compat_urlparse.urljoin(base_url, video_url),
 758                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 759             # More relaxed fallback
 760             else:
 761                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 762                     yield self.url_result(
 763                         'https://vimeo.com/%s' % video_id,
 764                         VimeoIE.ie_key(), video_id=video_id)
 765
 766             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 767                 break
 768
 769     def _extract_videos(self, list_id, base_url):
 770         title_and_entries = self._title_and_entries(list_id, base_url)
 771         list_title = next(title_and_entries)
 772         return self.playlist_result(title_and_entries, list_id, list_title)
 773
 774     def _real_extract(self, url):
 775         mobj = re.match(self._VALID_URL, url)
 776         channel_id = mobj.group('id')
 777         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 778
 779
 780 class VimeoUserIE(VimeoChannelIE):
 781     IE_NAME = 'vimeo:user'
 782     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 783     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 784     _TESTS = [{
 785         'url': 'https://vimeo.com/nkistudio/videos',
 786         'info_dict': {
 787             'title': 'Nki',
 788             'id': 'nkistudio',
 789         },
 790         'playlist_mincount': 66,
 791     }]
 792
 793     def _real_extract(self, url):
 794         mobj = re.match(self._VALID_URL, url)
 795         name = mobj.group('name')
 796         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 797
 798
 799 class VimeoAlbumIE(VimeoChannelIE):
 800     IE_NAME = 'vimeo:album'
 801     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 802     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 803     _TESTS = [{
 804         'url': 'https://vimeo.com/album/2632481',
 805         'info_dict': {
 806             'id': '2632481',
 807             'title': 'Staff Favorites: November 2013',
 808         },
 809         'playlist_mincount': 13,
 810     }, {
 811         'note': 'Password-protected album',
 812         'url': 'https://vimeo.com/album/3253534',
 813         'info_dict': {
 814             'title': 'test',
 815             'id': '3253534',
 816         },
 817         'playlist_count': 1,
 818         'params': {
 819             'videopassword': 'youtube-dl',
 820         }
 821     }, {
 822         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 823         'only_matching': True,
 824     }, {
 825         # TODO: respect page number
 826         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 827         'only_matching': True,
 828     }]
 829
 830     def _page_url(self, base_url, pagenum):
 831         return '%s/page:%d/' % (base_url, pagenum)
 832
 833     def _real_extract(self, url):
 834         album_id = self._match_id(url)
 835         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 836
 837
 838 class VimeoGroupsIE(VimeoAlbumIE):
 839     IE_NAME = 'vimeo:group'
 840     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 841     _TESTS = [{
 842         'url': 'https://vimeo.com/groups/rolexawards',
 843         'info_dict': {
 844             'id': 'rolexawards',
 845             'title': 'Rolex Awards for Enterprise',
 846         },
 847         'playlist_mincount': 73,
 848     }]
 849
 850     def _extract_list_title(self, webpage):
 851         return self._og_search_title(webpage)
 852
 853     def _real_extract(self, url):
 854         mobj = re.match(self._VALID_URL, url)
 855         name = mobj.group('name')
 856         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 857
 858
 859 class VimeoReviewIE(VimeoBaseInfoExtractor):
 860     IE_NAME = 'vimeo:review'
 861     IE_DESC = 'Review pages on vimeo'
 862     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 863     _TESTS = [{
 864         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 865         'md5': 'c507a72f780cacc12b2248bb4006d253',
 866         'info_dict': {
 867             'id': '75524534',
 868             'ext': 'mp4',
 869             'title': "DICK HARDWICK 'Comedian'",
 870             'uploader': 'Richard Hardwick',
 871             'uploader_id': 'user21297594',
 872         }
 873     }, {
 874         'note': 'video player needs Referer',
 875         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 876         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 877         'info_dict': {
 878             'id': '91613211',
 879             'ext': 'mp4',
 880             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 881             'uploader': 'DevWeek Events',
 882             'duration': 2773,
 883             'thumbnail': r're:^https?://.*\.jpg$',
 884             'uploader_id': 'user22258446',
 885         }
 886     }, {
 887         'note': 'Password protected',
 888         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 889         'info_dict': {
 890             'id': '138823582',
 891             'ext': 'mp4',
 892             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 893             'uploader': 'TMB',
 894             'uploader_id': 'user37284429',
 895         },
 896         'params': {
 897             'videopassword': 'holygrail',
 898         },
 899         'skip': 'video gone',
 900     }]
 901
 902     def _real_initialize(self):
 903         self._login()
 904
 905     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 906         webpage = self._download_webpage(webpage_url, video_id)
 907         config_url = self._html_search_regex(
 908             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 909             'config URL', default=None, group='url')
 910         if not config_url:
 911             data = self._parse_json(self._search_regex(
 912                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 913                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 914             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 915         if config_url is None:
 916             self._verify_video_password(webpage_url, video_id, webpage)
 917             config_url = self._get_config_url(
 918                 webpage_url, video_id, video_password_verified=True)
 919         return config_url
 920
 921     def _real_extract(self, url):
 922         video_id = self._match_id(url)
 923         config_url = self._get_config_url(url, video_id)
 924         config = self._download_json(config_url, video_id)
 925         info_dict = self._parse_config(config, video_id)
 926         self._vimeo_sort_formats(info_dict['formats'])
 927         info_dict['id'] = video_id
 928         return info_dict
 929
 930
 931 class VimeoWatchLaterIE(VimeoChannelIE):
 932     IE_NAME = 'vimeo:watchlater'
 933     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 934     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 935     _TITLE = 'Watch Later'
 936     _LOGIN_REQUIRED = True
 937     _TESTS = [{
 938         'url': 'https://vimeo.com/watchlater',
 939         'only_matching': True,
 940     }]
 941
 942     def _real_initialize(self):
 943         self._login()
 944
 945     def _page_url(self, base_url, pagenum):
 946         url = '%s/page:%d/' % (base_url, pagenum)
 947         request = sanitized_Request(url)
 948         # Set the header to get a partial html page with the ids,
 949         # the normal page doesn't contain them.
 950         request.add_header('X-Requested-With', 'XMLHttpRequest')
 951         return request
 952
 953     def _real_extract(self, url):
 954         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 955
 956
 957 class VimeoLikesIE(InfoExtractor):
 958     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 959     IE_NAME = 'vimeo:likes'
 960     IE_DESC = 'Vimeo user likes'
 961     _TEST = {
 962         'url': 'https://vimeo.com/user755559/likes/',
 963         'playlist_mincount': 293,
 964         'info_dict': {
 965             'id': 'user755559_likes',
 966             'description': 'See all the videos urza likes',
 967             'title': 'Videos urza likes',
 968         },
 969     }
 970
 971     def _real_extract(self, url):
 972         user_id = self._match_id(url)
 973         webpage = self._download_webpage(url, user_id)
 974         page_count = self._int(
 975             self._search_regex(
 976                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 977                     .*?</a></li>\s*<li\s+class="pagination_next">
 978                 ''', webpage, 'page count'),
 979             'page count', fatal=True)
 980         PAGE_SIZE = 12
 981         title = self._html_search_regex(
 982             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 983         description = self._html_search_meta('description', webpage)
 984
 985         def _get_page(idx):
 986             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 987                 user_id, idx + 1)
 988             webpage = self._download_webpage(
 989                 page_url, user_id,
 990                 note='Downloading page %d/%d' % (idx + 1, page_count))
 991             video_list = self._search_regex(
 992                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 993                 webpage, 'video content')
 994             paths = re.findall(
 995                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 996             for path in paths:
 997                 yield {
 998                     '_type': 'url',
 999                     'url': compat_urlparse.urljoin(page_url, path),
1000                 }
1001
1002         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1003
1004         return {
1005             '_type': 'playlist',
1006             'id': 'user%s_likes' % user_id,
1007             'title': title,
1008             'description': description,
1009             'entries': pl,
1010         }