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