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     merge_dicts,
  20     NO_DEFAULT,
  21     RegexNotFoundError,
  22     sanitized_Request,
  23     smuggle_url,
  24     std_headers,
  25     try_get,
  26     unified_timestamp,
  27     unsmuggle_url,
  28     urlencode_postdata,
  29     unescapeHTML,
  30     parse_filesize,
  31 )
  32
  33
  34 class VimeoBaseInfoExtractor(InfoExtractor):
  35     _NETRC_MACHINE = 'vimeo'
  36     _LOGIN_REQUIRED = False
  37     _LOGIN_URL = 'https://vimeo.com/log_in'
  38
  39     def _login(self):
  40         username, password = self._get_login_info()
  41         if username is None:
  42             if self._LOGIN_REQUIRED:
  43                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  44             return
  45         webpage = self._download_webpage(
  46             self._LOGIN_URL, None, 'Downloading login page')
  47         token, vuid = self._extract_xsrft_and_vuid(webpage)
  48         data = {
  49             'action': 'login',
  50             'email': username,
  51             'password': password,
  52             'service': 'vimeo',
  53             'token': token,
  54         }
  55         self._set_vimeo_cookie('vuid', vuid)
  56         try:
  57             self._download_webpage(
  58                 self._LOGIN_URL, None, 'Logging in',
  59                 data=urlencode_postdata(data), headers={
  60                     'Content-Type': 'application/x-www-form-urlencoded',
  61                     'Referer': self._LOGIN_URL,
  62                 })
  63         except ExtractorError as e:
  64             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 418:
  65                 raise ExtractorError(
  66                     'Unable to log in: bad username or password',
  67                     expected=True)
  68             raise ExtractorError('Unable to log in')
  69
  70     def _verify_video_password(self, url, video_id, webpage):
  71         password = self._downloader.params.get('videopassword')
  72         if password is None:
  73             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  74         token, vuid = self._extract_xsrft_and_vuid(webpage)
  75         data = urlencode_postdata({
  76             'password': password,
  77             'token': token,
  78         })
  79         if url.startswith('http://'):
  80             # vimeo only supports https now, but the user can give an http url
  81             url = url.replace('http://', 'https://')
  82         password_request = sanitized_Request(url + '/password', data)
  83         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  84         password_request.add_header('Referer', url)
  85         self._set_vimeo_cookie('vuid', vuid)
  86         return self._download_webpage(
  87             password_request, video_id,
  88             'Verifying the password', 'Wrong password')
  89
  90     def _extract_xsrft_and_vuid(self, webpage):
  91         xsrft = self._search_regex(
  92             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  93             webpage, 'login token', group='xsrft')
  94         vuid = self._search_regex(
  95             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  96             webpage, 'vuid', group='vuid')
  97         return xsrft, vuid
  98
  99     def _set_vimeo_cookie(self, name, value):
 100         self._set_cookie('vimeo.com', name, value)
 101
 102     def _vimeo_sort_formats(self, formats):
 103         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
 104         # at the same time without actual units specified. This lead to wrong sorting.
 105         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
 106
 107     def _parse_config(self, config, video_id):
 108         video_data = config['video']
 109         # Extract title
 110         video_title = video_data['title']
 111
 112         # Extract uploader, uploader_url and uploader_id
 113         video_uploader = video_data.get('owner', {}).get('name')
 114         video_uploader_url = video_data.get('owner', {}).get('url')
 115         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 116
 117         # Extract video thumbnail
 118         video_thumbnail = video_data.get('thumbnail')
 119         if video_thumbnail is None:
 120             video_thumbs = video_data.get('thumbs')
 121             if video_thumbs and isinstance(video_thumbs, dict):
 122                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 123
 124         # Extract video duration
 125         video_duration = int_or_none(video_data.get('duration'))
 126
 127         formats = []
 128         config_files = video_data.get('files') or config['request'].get('files', {})
 129         for f in config_files.get('progressive', []):
 130             video_url = f.get('url')
 131             if not video_url:
 132                 continue
 133             formats.append({
 134                 'url': video_url,
 135                 'format_id': 'http-%s' % f.get('quality'),
 136                 'width': int_or_none(f.get('width')),
 137                 'height': int_or_none(f.get('height')),
 138                 'fps': int_or_none(f.get('fps')),
 139                 'tbr': int_or_none(f.get('bitrate')),
 140             })
 141
 142         for files_type in ('hls', 'dash'):
 143             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 144                 manifest_url = cdn_data.get('url')
 145                 if not manifest_url:
 146                     continue
 147                 format_id = '%s-%s' % (files_type, cdn_name)
 148                 if files_type == 'hls':
 149                     formats.extend(self._extract_m3u8_formats(
 150                         manifest_url, video_id, 'mp4',
 151                         'm3u8_native', m3u8_id=format_id,
 152                         note='Downloading %s m3u8 information' % cdn_name,
 153                         fatal=False))
 154                 elif files_type == 'dash':
 155                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 156                     mpd_manifest_urls = []
 157                     if re.search(mpd_pattern, manifest_url):
 158                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 159                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 160                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 161                     else:
 162                         mpd_manifest_urls = [(format_id, manifest_url)]
 163                     for f_id, m_url in mpd_manifest_urls:
 164                         mpd_formats = self._extract_mpd_formats(
 165                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 166                             'Downloading %s MPD information' % cdn_name,
 167                             fatal=False)
 168                         for f in mpd_formats:
 169                             if f.get('vcodec') == 'none':
 170                                 f['preference'] = -50
 171                             elif f.get('acodec') == 'none':
 172                                 f['preference'] = -40
 173                         formats.extend(mpd_formats)
 174
 175         subtitles = {}
 176         text_tracks = config['request'].get('text_tracks')
 177         if text_tracks:
 178             for tt in text_tracks:
 179                 subtitles[tt['lang']] = [{
 180                     'ext': 'vtt',
 181                     'url': 'https://vimeo.com' + tt['url'],
 182                 }]
 183
 184         return {
 185             'title': video_title,
 186             'uploader': video_uploader,
 187             'uploader_id': video_uploader_id,
 188             'uploader_url': video_uploader_url,
 189             'thumbnail': video_thumbnail,
 190             'duration': video_duration,
 191             'formats': formats,
 192             'subtitles': subtitles,
 193         }
 194
 195
 196 class VimeoIE(VimeoBaseInfoExtractor):
 197     """Information extractor for vimeo.com."""
 198
 199     # _VALID_URL matches Vimeo URLs
 200     _VALID_URL = r'''(?x)
 201                     https?://
 202                         (?:
 203                             (?:
 204                                 www|
 205                                 (?P<player>player)
 206                             )
 207                             \.
 208                         )?
 209                         vimeo(?P<pro>pro)?\.com/
 210                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 211                         (?:.*?/)?
 212                         (?:
 213                             (?:
 214                                 play_redirect_hls|
 215                                 moogaloop\.swf)\?clip_id=
 216                             )?
 217                         (?:videos?/)?
 218                         (?P<id>[0-9]+)
 219                         (?:/[\da-f]+)?
 220                         /?(?:[?&].*)?(?:[#].*)?$
 221                     '''
 222     IE_NAME = 'vimeo'
 223     _TESTS = [
 224         {
 225             'url': 'http://vimeo.com/56015672#at=0',
 226             'md5': '8879b6cc097e987f02484baf890129e5',
 227             'info_dict': {
 228                 'id': '56015672',
 229                 'ext': 'mp4',
 230                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 231                 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479',
 232                 'timestamp': 1355990239,
 233                 'upload_date': '20121220',
 234                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 235                 'uploader_id': 'user7108434',
 236                 'uploader': 'Filippo Valsorda',
 237                 'duration': 10,
 238                 'license': 'by-sa',
 239             },
 240         },
 241         {
 242             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 243             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 244             'note': 'Vimeo Pro video (#1197)',
 245             'info_dict': {
 246                 'id': '68093876',
 247                 'ext': 'mp4',
 248                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 249                 'uploader_id': 'openstreetmapus',
 250                 'uploader': 'OpenStreetMap US',
 251                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 252                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 253                 'duration': 1595,
 254             },
 255         },
 256         {
 257             'url': 'http://player.vimeo.com/video/54469442',
 258             'md5': '619b811a4417aa4abe78dc653becf511',
 259             'note': 'Videos that embed the url in the player page',
 260             'info_dict': {
 261                 'id': '54469442',
 262                 'ext': 'mp4',
 263                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 264                 'uploader': 'The BLN & Business of Software',
 265                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 266                 'uploader_id': 'theblnbusinessofsoftware',
 267                 'duration': 3610,
 268                 'description': None,
 269             },
 270         },
 271         {
 272             'url': 'http://vimeo.com/68375962',
 273             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 274             'note': 'Video protected with password',
 275             'info_dict': {
 276                 'id': '68375962',
 277                 'ext': 'mp4',
 278                 'title': 'youtube-dl password protected test video',
 279                 'timestamp': 1371200155,
 280                 'upload_date': '20130614',
 281                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 282                 'uploader_id': 'user18948128',
 283                 'uploader': 'Jaime Marquínez Ferrándiz',
 284                 'duration': 10,
 285                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 286             },
 287             'params': {
 288                 'videopassword': 'youtube-dl',
 289             },
 290         },
 291         {
 292             'url': 'http://vimeo.com/channels/keypeele/75629013',
 293             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 294             'info_dict': {
 295                 'id': '75629013',
 296                 'ext': 'mp4',
 297                 'title': 'Key & Peele: Terrorist Interrogation',
 298                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 299                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 300                 'uploader_id': 'atencio',
 301                 'uploader': 'Peter Atencio',
 302                 'channel_id': 'keypeele',
 303                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/keypeele',
 304                 'timestamp': 1380339469,
 305                 'upload_date': '20130928',
 306                 'duration': 187,
 307             },
 308             'expected_warnings': ['Unable to download JSON metadata'],
 309         },
 310         {
 311             'url': 'http://vimeo.com/76979871',
 312             'note': 'Video with subtitles',
 313             'info_dict': {
 314                 'id': '76979871',
 315                 'ext': 'mp4',
 316                 'title': 'The New Vimeo Player (You Know, For Videos)',
 317                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 318                 'timestamp': 1381846109,
 319                 'upload_date': '20131015',
 320                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 321                 'uploader_id': 'staff',
 322                 'uploader': 'Vimeo Staff',
 323                 'duration': 62,
 324             }
 325         },
 326         {
 327             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 328             'url': 'https://player.vimeo.com/video/98044508',
 329             'note': 'The js code contains assignments to the same variable as the config',
 330             'info_dict': {
 331                 'id': '98044508',
 332                 'ext': 'mp4',
 333                 'title': 'Pier Solar OUYA Official Trailer',
 334                 'uploader': 'Tulio Gonçalves',
 335                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 336                 'uploader_id': 'user28849593',
 337             },
 338         },
 339         {
 340             # contains original format
 341             'url': 'https://vimeo.com/33951933',
 342             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 343             'info_dict': {
 344                 'id': '33951933',
 345                 'ext': 'mp4',
 346                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 347                 'uploader': 'The DMCI',
 348                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 349                 'uploader_id': 'dmci',
 350                 'timestamp': 1324343742,
 351                 'upload_date': '20111220',
 352                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 353             },
 354         },
 355         {
 356             # only available via https://vimeo.com/channels/tributes/6213729 and
 357             # not via https://vimeo.com/6213729
 358             'url': 'https://vimeo.com/channels/tributes/6213729',
 359             'info_dict': {
 360                 'id': '6213729',
 361                 'ext': 'mp4',
 362                 'title': 'Vimeo Tribute: The Shining',
 363                 'uploader': 'Casey Donahue',
 364                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 365                 'uploader_id': 'caseydonahue',
 366                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/tributes',
 367                 'channel_id': 'tributes',
 368                 'timestamp': 1250886430,
 369                 'upload_date': '20090821',
 370                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 371             },
 372             'params': {
 373                 'skip_download': True,
 374             },
 375             'expected_warnings': ['Unable to download JSON metadata'],
 376         },
 377         {
 378             # redirects to ondemand extractor and should be passed through it
 379             # for successful extraction
 380             'url': 'https://vimeo.com/73445910',
 381             'info_dict': {
 382                 'id': '73445910',
 383                 'ext': 'mp4',
 384                 'title': 'The Reluctant Revolutionary',
 385                 'uploader': '10Ft Films',
 386                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 387                 'uploader_id': 'tenfootfilms',
 388             },
 389             'params': {
 390                 'skip_download': True,
 391             },
 392         },
 393         {
 394             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 395             'only_matching': True,
 396         },
 397         {
 398             'url': 'https://vimeo.com/109815029',
 399             'note': 'Video not completely processed, "failed" seed status',
 400             'only_matching': True,
 401         },
 402         {
 403             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 404             'only_matching': True,
 405         },
 406         {
 407             'url': 'https://vimeo.com/album/2632481/video/79010983',
 408             'only_matching': True,
 409         },
 410         {
 411             # source file returns 403: Forbidden
 412             'url': 'https://vimeo.com/7809605',
 413             'only_matching': True,
 414         },
 415         {
 416             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 417             'only_matching': True,
 418         }
 419     ]
 420
 421     @staticmethod
 422     def _smuggle_referrer(url, referrer_url):
 423         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 424
 425     @staticmethod
 426     def _extract_urls(url, webpage):
 427         urls = []
 428         # Look for embedded (iframe) Vimeo player
 429         for mobj in re.finditer(
 430                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1',
 431                 webpage):
 432             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 433         PLAIN_EMBED_RE = (
 434             # Look for embedded (swf embed) Vimeo player
 435             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 436             # Look more for non-standard embedded Vimeo player
 437             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 438         )
 439         for embed_re in PLAIN_EMBED_RE:
 440             for mobj in re.finditer(embed_re, webpage):
 441                 urls.append(mobj.group('url'))
 442         return urls
 443
 444     @staticmethod
 445     def _extract_url(url, webpage):
 446         urls = VimeoIE._extract_urls(url, webpage)
 447         return urls[0] if urls else None
 448
 449     def _verify_player_video_password(self, url, video_id):
 450         password = self._downloader.params.get('videopassword')
 451         if password is None:
 452             raise ExtractorError('This video is protected by a password, use the --video-password option')
 453         data = urlencode_postdata({'password': password})
 454         pass_url = url + '/check-password'
 455         password_request = sanitized_Request(pass_url, data)
 456         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 457         password_request.add_header('Referer', url)
 458         return self._download_json(
 459             password_request, video_id,
 460             'Verifying the password', 'Wrong password')
 461
 462     def _real_initialize(self):
 463         self._login()
 464
 465     def _real_extract(self, url):
 466         url, data = unsmuggle_url(url, {})
 467         headers = std_headers.copy()
 468         if 'http_headers' in data:
 469             headers.update(data['http_headers'])
 470         if 'Referer' not in headers:
 471             headers['Referer'] = url
 472
 473         channel_id = self._search_regex(
 474             r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None)
 475
 476         # Extract ID from URL
 477         mobj = re.match(self._VALID_URL, url)
 478         video_id = mobj.group('id')
 479         orig_url = url
 480         if mobj.group('pro') or mobj.group('player'):
 481             url = 'https://player.vimeo.com/video/' + video_id
 482         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 483             url = 'https://vimeo.com/' + video_id
 484
 485         # Retrieve video webpage to extract further information
 486         request = sanitized_Request(url, headers=headers)
 487         try:
 488             webpage, urlh = self._download_webpage_handle(request, video_id)
 489             redirect_url = compat_str(urlh.geturl())
 490             # Some URLs redirect to ondemand can't be extracted with
 491             # this extractor right away thus should be passed through
 492             # ondemand extractor (e.g. https://vimeo.com/73445910)
 493             if VimeoOndemandIE.suitable(redirect_url):
 494                 return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
 495         except ExtractorError as ee:
 496             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 497                 errmsg = ee.cause.read()
 498                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 499                     raise ExtractorError(
 500                         'Cannot download embed-only video without embedding '
 501                         'URL. Please call youtube-dl with the URL of the page '
 502                         'that embeds this video.',
 503                         expected=True)
 504             raise
 505
 506         # Now we begin extracting as much information as we can from what we
 507         # retrieved. First we extract the information common to all extractors,
 508         # and latter we extract those that are Vimeo specific.
 509         self.report_extraction(video_id)
 510
 511         vimeo_config = self._search_regex(
 512             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 513             'vimeo config', default=None)
 514         if vimeo_config:
 515             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 516             if seed_status.get('state') == 'failed':
 517                 raise ExtractorError(
 518                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 519                     expected=True)
 520
 521         cc_license = None
 522         timestamp = None
 523
 524         # Extract the config JSON
 525         try:
 526             try:
 527                 config_url = self._html_search_regex(
 528                     r' data-config-url="(.+?)"', webpage,
 529                     'config URL', default=None)
 530                 if not config_url:
 531                     # Sometimes new react-based page is served instead of old one that require
 532                     # different config URL extraction approach (see
 533                     # https://github.com/rg3/youtube-dl/pull/7209)
 534                     vimeo_clip_page_config = self._search_regex(
 535                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 536                         'vimeo clip page config')
 537                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 538                     config_url = page_config['player']['config_url']
 539                     cc_license = page_config.get('cc_license')
 540                     timestamp = try_get(
 541                         page_config, lambda x: x['clip']['uploaded_on'],
 542                         compat_str)
 543                 config_json = self._download_webpage(config_url, video_id)
 544                 config = json.loads(config_json)
 545             except RegexNotFoundError:
 546                 # For pro videos or player.vimeo.com urls
 547                 # We try to find out to which variable is assigned the config dic
 548                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 549                 if m_variable_name is not None:
 550                     config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))]
 551                 else:
 552                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 553                 config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;')
 554                 config = self._search_regex(config_re, webpage, 'info section',
 555                                             flags=re.DOTALL)
 556                 config = json.loads(config)
 557         except Exception as e:
 558             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 559                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 560
 561             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 562                 if '_video_password_verified' in data:
 563                     raise ExtractorError('video password verification failed!')
 564                 self._verify_video_password(redirect_url, video_id, webpage)
 565                 return self._real_extract(
 566                     smuggle_url(redirect_url, {'_video_password_verified': 'verified'}))
 567             else:
 568                 raise ExtractorError('Unable to extract info section',
 569                                      cause=e)
 570         else:
 571             if config.get('view') == 4:
 572                 config = self._verify_player_video_password(redirect_url, video_id)
 573
 574         def is_rented():
 575             if '>You rented this title.<' in webpage:
 576                 return True
 577             if config.get('user', {}).get('purchased'):
 578                 return True
 579             label = try_get(
 580                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 581             if label and label.startswith('You rented this'):
 582                 return True
 583             return False
 584
 585         if is_rented():
 586             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 587             if feature_id and not data.get('force_feature_id', False):
 588                 return self.url_result(smuggle_url(
 589                     'https://player.vimeo.com/player/%s' % feature_id,
 590                     {'force_feature_id': True}), 'Vimeo')
 591
 592         # Extract video description
 593
 594         video_description = self._html_search_regex(
 595             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 596             webpage, 'description', default=None)
 597         if not video_description:
 598             video_description = self._html_search_meta(
 599                 'description', webpage, default=None)
 600         if not video_description and mobj.group('pro'):
 601             orig_webpage = self._download_webpage(
 602                 orig_url, video_id,
 603                 note='Downloading webpage for description',
 604                 fatal=False)
 605             if orig_webpage:
 606                 video_description = self._html_search_meta(
 607                     'description', orig_webpage, default=None)
 608         if not video_description and not mobj.group('player'):
 609             self._downloader.report_warning('Cannot find video description')
 610
 611         # Extract upload date
 612         if not timestamp:
 613             timestamp = self._search_regex(
 614                 r'<time[^>]+datetime="([^"]+)"', webpage,
 615                 'timestamp', default=None)
 616
 617         try:
 618             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 619             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 620             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 621         except RegexNotFoundError:
 622             # This info is only available in vimeo.com/{id} urls
 623             view_count = None
 624             like_count = None
 625             comment_count = None
 626
 627         formats = []
 628         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 629             'X-Requested-With': 'XMLHttpRequest'})
 630         download_data = self._download_json(download_request, video_id, fatal=False)
 631         if download_data:
 632             source_file = download_data.get('source_file')
 633             if isinstance(source_file, dict):
 634                 download_url = source_file.get('download_url')
 635                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 636                     source_name = source_file.get('public_name', 'Original')
 637                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 638                         ext = (try_get(
 639                             source_file, lambda x: x['extension'],
 640                             compat_str) or determine_ext(
 641                             download_url, None) or 'mp4').lower()
 642                         formats.append({
 643                             'url': download_url,
 644                             'ext': ext,
 645                             'width': int_or_none(source_file.get('width')),
 646                             'height': int_or_none(source_file.get('height')),
 647                             'filesize': parse_filesize(source_file.get('size')),
 648                             'format_id': source_name,
 649                             'preference': 1,
 650                         })
 651
 652         info_dict_config = self._parse_config(config, video_id)
 653         formats.extend(info_dict_config['formats'])
 654         self._vimeo_sort_formats(formats)
 655
 656         json_ld = self._search_json_ld(webpage, video_id, default={})
 657
 658         if not cc_license:
 659             cc_license = self._search_regex(
 660                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 661                 webpage, 'license', default=None, group='license')
 662
 663         channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None
 664
 665         info_dict = {
 666             'id': video_id,
 667             'formats': formats,
 668             'timestamp': unified_timestamp(timestamp),
 669             'description': video_description,
 670             'webpage_url': url,
 671             'view_count': view_count,
 672             'like_count': like_count,
 673             'comment_count': comment_count,
 674             'license': cc_license,
 675             'channel_id': channel_id,
 676             'channel_url': channel_url,
 677         }
 678
 679         info_dict = merge_dicts(info_dict, info_dict_config, json_ld)
 680
 681         return info_dict
 682
 683
 684 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 685     IE_NAME = 'vimeo:ondemand'
 686     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 687     _TESTS = [{
 688         # ondemand video not available via https://vimeo.com/id
 689         'url': 'https://vimeo.com/ondemand/20704',
 690         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 691         'info_dict': {
 692             'id': '105442900',
 693             'ext': 'mp4',
 694             'title': 'המעבדה - במאי יותם פלדמן',
 695             'uploader': 'גם סרטים',
 696             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 697             'uploader_id': 'gumfilms',
 698         },
 699         'params': {
 700             'format': 'best[protocol=https]',
 701         },
 702     }, {
 703         # requires Referer to be passed along with og:video:url
 704         'url': 'https://vimeo.com/ondemand/36938/126682985',
 705         'info_dict': {
 706             'id': '126682985',
 707             'ext': 'mp4',
 708             'title': 'Rävlock, rätt läte på rätt plats',
 709             'uploader': 'Lindroth & Norin',
 710             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 711             'uploader_id': 'user14430847',
 712         },
 713         'params': {
 714             'skip_download': True,
 715         },
 716     }, {
 717         'url': 'https://vimeo.com/ondemand/nazmaalik',
 718         'only_matching': True,
 719     }, {
 720         'url': 'https://vimeo.com/ondemand/141692381',
 721         'only_matching': True,
 722     }, {
 723         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 724         'only_matching': True,
 725     }]
 726
 727     def _real_extract(self, url):
 728         video_id = self._match_id(url)
 729         webpage = self._download_webpage(url, video_id)
 730         return self.url_result(
 731             # Some videos require Referer to be passed along with og:video:url
 732             # similarly to generic vimeo embeds (e.g.
 733             # https://vimeo.com/ondemand/36938/126682985).
 734             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 735             VimeoIE.ie_key())
 736
 737
 738 class VimeoChannelIE(VimeoBaseInfoExtractor):
 739     IE_NAME = 'vimeo:channel'
 740     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 741     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 742     _TITLE = None
 743     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 744     _TESTS = [{
 745         'url': 'https://vimeo.com/channels/tributes',
 746         'info_dict': {
 747             'id': 'tributes',
 748             'title': 'Vimeo Tributes',
 749         },
 750         'playlist_mincount': 25,
 751     }]
 752
 753     def _page_url(self, base_url, pagenum):
 754         return '%s/videos/page:%d/' % (base_url, pagenum)
 755
 756     def _extract_list_title(self, webpage):
 757         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 758
 759     def _login_list_password(self, page_url, list_id, webpage):
 760         login_form = self._search_regex(
 761             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 762             webpage, 'login form', default=None)
 763         if not login_form:
 764             return webpage
 765
 766         password = self._downloader.params.get('videopassword')
 767         if password is None:
 768             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 769         fields = self._hidden_inputs(login_form)
 770         token, vuid = self._extract_xsrft_and_vuid(webpage)
 771         fields['token'] = token
 772         fields['password'] = password
 773         post = urlencode_postdata(fields)
 774         password_path = self._search_regex(
 775             r'action="([^"]+)"', login_form, 'password URL')
 776         password_url = compat_urlparse.urljoin(page_url, password_path)
 777         password_request = sanitized_Request(password_url, post)
 778         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 779         self._set_vimeo_cookie('vuid', vuid)
 780         self._set_vimeo_cookie('xsrft', token)
 781
 782         return self._download_webpage(
 783             password_request, list_id,
 784             'Verifying the password', 'Wrong password')
 785
 786     def _title_and_entries(self, list_id, base_url):
 787         for pagenum in itertools.count(1):
 788             page_url = self._page_url(base_url, pagenum)
 789             webpage = self._download_webpage(
 790                 page_url, list_id,
 791                 'Downloading page %s' % pagenum)
 792
 793             if pagenum == 1:
 794                 webpage = self._login_list_password(page_url, list_id, webpage)
 795                 yield self._extract_list_title(webpage)
 796
 797             # Try extracting href first since not all videos are available via
 798             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 799             clips = re.findall(
 800                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 801             if clips:
 802                 for video_id, video_url, video_title in clips:
 803                     yield self.url_result(
 804                         compat_urlparse.urljoin(base_url, video_url),
 805                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 806             # More relaxed fallback
 807             else:
 808                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 809                     yield self.url_result(
 810                         'https://vimeo.com/%s' % video_id,
 811                         VimeoIE.ie_key(), video_id=video_id)
 812
 813             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 814                 break
 815
 816     def _extract_videos(self, list_id, base_url):
 817         title_and_entries = self._title_and_entries(list_id, base_url)
 818         list_title = next(title_and_entries)
 819         return self.playlist_result(title_and_entries, list_id, list_title)
 820
 821     def _real_extract(self, url):
 822         mobj = re.match(self._VALID_URL, url)
 823         channel_id = mobj.group('id')
 824         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 825
 826
 827 class VimeoUserIE(VimeoChannelIE):
 828     IE_NAME = 'vimeo:user'
 829     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 830     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 831     _TESTS = [{
 832         'url': 'https://vimeo.com/nkistudio/videos',
 833         'info_dict': {
 834             'title': 'Nki',
 835             'id': 'nkistudio',
 836         },
 837         'playlist_mincount': 66,
 838     }]
 839
 840     def _real_extract(self, url):
 841         mobj = re.match(self._VALID_URL, url)
 842         name = mobj.group('name')
 843         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 844
 845
 846 class VimeoAlbumIE(VimeoChannelIE):
 847     IE_NAME = 'vimeo:album'
 848     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 849     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 850     _TESTS = [{
 851         'url': 'https://vimeo.com/album/2632481',
 852         'info_dict': {
 853             'id': '2632481',
 854             'title': 'Staff Favorites: November 2013',
 855         },
 856         'playlist_mincount': 13,
 857     }, {
 858         'note': 'Password-protected album',
 859         'url': 'https://vimeo.com/album/3253534',
 860         'info_dict': {
 861             'title': 'test',
 862             'id': '3253534',
 863         },
 864         'playlist_count': 1,
 865         'params': {
 866             'videopassword': 'youtube-dl',
 867         }
 868     }, {
 869         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 870         'only_matching': True,
 871     }, {
 872         # TODO: respect page number
 873         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 874         'only_matching': True,
 875     }]
 876
 877     def _page_url(self, base_url, pagenum):
 878         return '%s/page:%d/' % (base_url, pagenum)
 879
 880     def _real_extract(self, url):
 881         album_id = self._match_id(url)
 882         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 883
 884
 885 class VimeoGroupsIE(VimeoAlbumIE):
 886     IE_NAME = 'vimeo:group'
 887     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 888     _TESTS = [{
 889         'url': 'https://vimeo.com/groups/rolexawards',
 890         'info_dict': {
 891             'id': 'rolexawards',
 892             'title': 'Rolex Awards for Enterprise',
 893         },
 894         'playlist_mincount': 73,
 895     }]
 896
 897     def _extract_list_title(self, webpage):
 898         return self._og_search_title(webpage)
 899
 900     def _real_extract(self, url):
 901         mobj = re.match(self._VALID_URL, url)
 902         name = mobj.group('name')
 903         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 904
 905
 906 class VimeoReviewIE(VimeoBaseInfoExtractor):
 907     IE_NAME = 'vimeo:review'
 908     IE_DESC = 'Review pages on vimeo'
 909     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 910     _TESTS = [{
 911         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 912         'md5': 'c507a72f780cacc12b2248bb4006d253',
 913         'info_dict': {
 914             'id': '75524534',
 915             'ext': 'mp4',
 916             'title': "DICK HARDWICK 'Comedian'",
 917             'uploader': 'Richard Hardwick',
 918             'uploader_id': 'user21297594',
 919         }
 920     }, {
 921         'note': 'video player needs Referer',
 922         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 923         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 924         'info_dict': {
 925             'id': '91613211',
 926             'ext': 'mp4',
 927             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 928             'uploader': 'DevWeek Events',
 929             'duration': 2773,
 930             'thumbnail': r're:^https?://.*\.jpg$',
 931             'uploader_id': 'user22258446',
 932         }
 933     }, {
 934         'note': 'Password protected',
 935         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 936         'info_dict': {
 937             'id': '138823582',
 938             'ext': 'mp4',
 939             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 940             'uploader': 'TMB',
 941             'uploader_id': 'user37284429',
 942         },
 943         'params': {
 944             'videopassword': 'holygrail',
 945         },
 946         'skip': 'video gone',
 947     }]
 948
 949     def _real_initialize(self):
 950         self._login()
 951
 952     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 953         webpage = self._download_webpage(webpage_url, video_id)
 954         config_url = self._html_search_regex(
 955             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 956             'config URL', default=None, group='url')
 957         if not config_url:
 958             data = self._parse_json(self._search_regex(
 959                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 960                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 961             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 962         if config_url is None:
 963             self._verify_video_password(webpage_url, video_id, webpage)
 964             config_url = self._get_config_url(
 965                 webpage_url, video_id, video_password_verified=True)
 966         return config_url
 967
 968     def _real_extract(self, url):
 969         video_id = self._match_id(url)
 970         config_url = self._get_config_url(url, video_id)
 971         config = self._download_json(config_url, video_id)
 972         info_dict = self._parse_config(config, video_id)
 973         self._vimeo_sort_formats(info_dict['formats'])
 974         info_dict['id'] = video_id
 975         return info_dict
 976
 977
 978 class VimeoWatchLaterIE(VimeoChannelIE):
 979     IE_NAME = 'vimeo:watchlater'
 980     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 981     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 982     _TITLE = 'Watch Later'
 983     _LOGIN_REQUIRED = True
 984     _TESTS = [{
 985         'url': 'https://vimeo.com/watchlater',
 986         'only_matching': True,
 987     }]
 988
 989     def _real_initialize(self):
 990         self._login()
 991
 992     def _page_url(self, base_url, pagenum):
 993         url = '%s/page:%d/' % (base_url, pagenum)
 994         request = sanitized_Request(url)
 995         # Set the header to get a partial html page with the ids,
 996         # the normal page doesn't contain them.
 997         request.add_header('X-Requested-With', 'XMLHttpRequest')
 998         return request
 999
1000     def _real_extract(self, url):
1001         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
1002
1003
1004 class VimeoLikesIE(InfoExtractor):
1005     _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)'
1006     IE_NAME = 'vimeo:likes'
1007     IE_DESC = 'Vimeo user likes'
1008     _TESTS = [{
1009         'url': 'https://vimeo.com/user755559/likes/',
1010         'playlist_mincount': 293,
1011         'info_dict': {
1012             'id': 'user755559_likes',
1013             'description': 'See all the videos urza likes',
1014             'title': 'Videos urza likes',
1015         },
1016     }, {
1017         'url': 'https://vimeo.com/stormlapse/likes',
1018         'only_matching': True,
1019     }]
1020
1021     def _real_extract(self, url):
1022         user_id = self._match_id(url)
1023         webpage = self._download_webpage(url, user_id)
1024         page_count = self._int(
1025             self._search_regex(
1026                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
1027                     .*?</a></li>\s*<li\s+class="pagination_next">
1028                 ''', webpage, 'page count', default=1),
1029             'page count', fatal=True)
1030         PAGE_SIZE = 12
1031         title = self._html_search_regex(
1032             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
1033         description = self._html_search_meta('description', webpage)
1034
1035         def _get_page(idx):
1036             page_url = 'https://vimeo.com/%s/likes/page:%d/sort:date' % (
1037                 user_id, idx + 1)
1038             webpage = self._download_webpage(
1039                 page_url, user_id,
1040                 note='Downloading page %d/%d' % (idx + 1, page_count))
1041             video_list = self._search_regex(
1042                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
1043                 webpage, 'video content')
1044             paths = re.findall(
1045                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
1046             for path in paths:
1047                 yield {
1048                     '_type': 'url',
1049                     'url': compat_urlparse.urljoin(page_url, path),
1050                 }
1051
1052         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1053
1054         return {
1055             '_type': 'playlist',
1056             'id': '%s_likes' % user_id,
1057             'title': title,
1058             'description': description,
1059             'entries': pl,
1060         }