youtube_dl/extractor/vimeo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import functools
   6 import json
   7 import re
   8 import itertools
   9
  10 from .common import InfoExtractor
  11 from ..compat import (
  12     compat_kwargs,
  13     compat_HTTPError,
  14     compat_str,
  15     compat_urlparse,
  16 )
  17 from ..utils import (
  18     determine_ext,
  19     ExtractorError,
  20     js_to_json,
  21     int_or_none,
  22     merge_dicts,
  23     NO_DEFAULT,
  24     OnDemandPagedList,
  25     parse_filesize,
  26     RegexNotFoundError,
  27     sanitized_Request,
  28     smuggle_url,
  29     std_headers,
  30     try_get,
  31     unified_timestamp,
  32     unsmuggle_url,
  33     urlencode_postdata,
  34     unescapeHTML,
  35 )
  36
  37
  38 class VimeoBaseInfoExtractor(InfoExtractor):
  39     _NETRC_MACHINE = 'vimeo'
  40     _LOGIN_REQUIRED = False
  41     _LOGIN_URL = 'https://vimeo.com/log_in'
  42
  43     def _login(self):
  44         username, password = self._get_login_info()
  45         if username is None:
  46             if self._LOGIN_REQUIRED:
  47                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  48             return
  49         webpage = self._download_webpage(
  50             self._LOGIN_URL, None, 'Downloading login page')
  51         token, vuid = self._extract_xsrft_and_vuid(webpage)
  52         data = {
  53             'action': 'login',
  54             'email': username,
  55             'password': password,
  56             'service': 'vimeo',
  57             'token': token,
  58         }
  59         self._set_vimeo_cookie('vuid', vuid)
  60         try:
  61             self._download_webpage(
  62                 self._LOGIN_URL, None, 'Logging in',
  63                 data=urlencode_postdata(data), headers={
  64                     'Content-Type': 'application/x-www-form-urlencoded',
  65                     'Referer': self._LOGIN_URL,
  66                 })
  67         except ExtractorError as e:
  68             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 418:
  69                 raise ExtractorError(
  70                     'Unable to log in: bad username or password',
  71                     expected=True)
  72             raise ExtractorError('Unable to log in')
  73
  74     def _verify_video_password(self, url, video_id, webpage):
  75         password = self._downloader.params.get('videopassword')
  76         if password is None:
  77             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  78         token, vuid = self._extract_xsrft_and_vuid(webpage)
  79         data = urlencode_postdata({
  80             'password': password,
  81             'token': token,
  82         })
  83         if url.startswith('http://'):
  84             # vimeo only supports https now, but the user can give an http url
  85             url = url.replace('http://', 'https://')
  86         password_request = sanitized_Request(url + '/password', data)
  87         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  88         password_request.add_header('Referer', url)
  89         self._set_vimeo_cookie('vuid', vuid)
  90         return self._download_webpage(
  91             password_request, video_id,
  92             'Verifying the password', 'Wrong password')
  93
  94     def _extract_xsrft_and_vuid(self, webpage):
  95         xsrft = self._search_regex(
  96             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  97             webpage, 'login token', group='xsrft')
  98         vuid = self._search_regex(
  99             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
 100             webpage, 'vuid', group='vuid')
 101         return xsrft, vuid
 102
 103     def _extract_vimeo_config(self, webpage, video_id, *args, **kwargs):
 104         vimeo_config = self._search_regex(
 105             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));',
 106             webpage, 'vimeo config', *args, **compat_kwargs(kwargs))
 107         if vimeo_config:
 108             return self._parse_json(vimeo_config, video_id)
 109
 110     def _set_vimeo_cookie(self, name, value):
 111         self._set_cookie('vimeo.com', name, value)
 112
 113     def _vimeo_sort_formats(self, formats):
 114         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
 115         # at the same time without actual units specified. This lead to wrong sorting.
 116         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
 117
 118     def _parse_config(self, config, video_id):
 119         video_data = config['video']
 120         video_title = video_data['title']
 121         live_event = video_data.get('live_event') or {}
 122         is_live = live_event.get('status') == 'started'
 123
 124         formats = []
 125         config_files = video_data.get('files') or config['request'].get('files', {})
 126         for f in config_files.get('progressive', []):
 127             video_url = f.get('url')
 128             if not video_url:
 129                 continue
 130             formats.append({
 131                 'url': video_url,
 132                 'format_id': 'http-%s' % f.get('quality'),
 133                 'width': int_or_none(f.get('width')),
 134                 'height': int_or_none(f.get('height')),
 135                 'fps': int_or_none(f.get('fps')),
 136                 'tbr': int_or_none(f.get('bitrate')),
 137             })
 138
 139         # TODO: fix handling of 308 status code returned for live archive manifest requests
 140         for files_type in ('hls', 'dash'):
 141             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 142                 manifest_url = cdn_data.get('url')
 143                 if not manifest_url:
 144                     continue
 145                 format_id = '%s-%s' % (files_type, cdn_name)
 146                 if files_type == 'hls':
 147                     formats.extend(self._extract_m3u8_formats(
 148                         manifest_url, video_id, 'mp4',
 149                         'm3u8' if is_live else 'm3u8_native', m3u8_id=format_id,
 150                         note='Downloading %s m3u8 information' % cdn_name,
 151                         fatal=False))
 152                 elif files_type == 'dash':
 153                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 154                     mpd_manifest_urls = []
 155                     if re.search(mpd_pattern, manifest_url):
 156                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 157                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 158                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 159                     else:
 160                         mpd_manifest_urls = [(format_id, manifest_url)]
 161                     for f_id, m_url in mpd_manifest_urls:
 162                         if 'json=1' in m_url:
 163                             real_m_url = (self._download_json(m_url, video_id, fatal=False) or {}).get('url')
 164                             if real_m_url:
 165                                 m_url = real_m_url
 166                         mpd_formats = self._extract_mpd_formats(
 167                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 168                             'Downloading %s MPD information' % cdn_name,
 169                             fatal=False)
 170                         for f in mpd_formats:
 171                             if f.get('vcodec') == 'none':
 172                                 f['preference'] = -50
 173                             elif f.get('acodec') == 'none':
 174                                 f['preference'] = -40
 175                         formats.extend(mpd_formats)
 176
 177         live_archive = live_event.get('archive') or {}
 178         live_archive_source_url = live_archive.get('source_url')
 179         if live_archive_source_url and live_archive.get('status') == 'done':
 180             formats.append({
 181                 'format_id': 'live-archive-source',
 182                 'url': live_archive_source_url,
 183                 'preference': 1,
 184             })
 185
 186         subtitles = {}
 187         text_tracks = config['request'].get('text_tracks')
 188         if text_tracks:
 189             for tt in text_tracks:
 190                 subtitles[tt['lang']] = [{
 191                     'ext': 'vtt',
 192                     'url': 'https://vimeo.com' + tt['url'],
 193                 }]
 194
 195         thumbnails = []
 196         if not is_live:
 197             for key, thumb in video_data.get('thumbs', {}).items():
 198                 thumbnails.append({
 199                     'id': key,
 200                     'width': int_or_none(key),
 201                     'url': thumb,
 202                 })
 203             thumbnail = video_data.get('thumbnail')
 204             if thumbnail:
 205                 thumbnails.append({
 206                     'url': thumbnail,
 207                 })
 208
 209         owner = video_data.get('owner') or {}
 210         video_uploader_url = owner.get('url')
 211
 212         return {
 213             'id': video_id,
 214             'title': self._live_title(video_title) if is_live else video_title,
 215             'uploader': owner.get('name'),
 216             'uploader_id': video_uploader_url.split('/')[-1] if video_uploader_url else None,
 217             'uploader_url': video_uploader_url,
 218             'thumbnails': thumbnails,
 219             'duration': int_or_none(video_data.get('duration')),
 220             'formats': formats,
 221             'subtitles': subtitles,
 222             'is_live': is_live,
 223         }
 224
 225     def _extract_original_format(self, url, video_id):
 226         download_data = self._download_json(
 227             url, video_id, fatal=False,
 228             query={'action': 'load_download_config'},
 229             headers={'X-Requested-With': 'XMLHttpRequest'})
 230         if download_data:
 231             source_file = download_data.get('source_file')
 232             if isinstance(source_file, dict):
 233                 download_url = source_file.get('download_url')
 234                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 235                     source_name = source_file.get('public_name', 'Original')
 236                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 237                         ext = (try_get(
 238                             source_file, lambda x: x['extension'],
 239                             compat_str) or determine_ext(
 240                             download_url, None) or 'mp4').lower()
 241                         return {
 242                             'url': download_url,
 243                             'ext': ext,
 244                             'width': int_or_none(source_file.get('width')),
 245                             'height': int_or_none(source_file.get('height')),
 246                             'filesize': parse_filesize(source_file.get('size')),
 247                             'format_id': source_name,
 248                             'preference': 1,
 249                         }
 250
 251
 252 class VimeoIE(VimeoBaseInfoExtractor):
 253     """Information extractor for vimeo.com."""
 254
 255     # _VALID_URL matches Vimeo URLs
 256     _VALID_URL = r'''(?x)
 257                     https?://
 258                         (?:
 259                             (?:
 260                                 www|
 261                                 (?P<player>player)
 262                             )
 263                             \.
 264                         )?
 265                         vimeo(?P<pro>pro)?\.com/
 266                         (?!(?:channels|album|showcase)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 267                         (?:.*?/)?
 268                         (?:
 269                             (?:
 270                                 play_redirect_hls|
 271                                 moogaloop\.swf)\?clip_id=
 272                             )?
 273                         (?:videos?/)?
 274                         (?P<id>[0-9]+)
 275                         (?:/[\da-f]+)?
 276                         /?(?:[?&].*)?(?:[#].*)?$
 277                     '''
 278     IE_NAME = 'vimeo'
 279     _TESTS = [
 280         {
 281             'url': 'http://vimeo.com/56015672#at=0',
 282             'md5': '8879b6cc097e987f02484baf890129e5',
 283             'info_dict': {
 284                 'id': '56015672',
 285                 'ext': 'mp4',
 286                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 287                 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479',
 288                 'timestamp': 1355990239,
 289                 'upload_date': '20121220',
 290                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 291                 'uploader_id': 'user7108434',
 292                 'uploader': 'Filippo Valsorda',
 293                 'duration': 10,
 294                 'license': 'by-sa',
 295             },
 296         },
 297         {
 298             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 299             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 300             'note': 'Vimeo Pro video (#1197)',
 301             'info_dict': {
 302                 'id': '68093876',
 303                 'ext': 'mp4',
 304                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 305                 'uploader_id': 'openstreetmapus',
 306                 'uploader': 'OpenStreetMap US',
 307                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 308                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 309                 'duration': 1595,
 310             },
 311         },
 312         {
 313             'url': 'http://player.vimeo.com/video/54469442',
 314             'md5': '619b811a4417aa4abe78dc653becf511',
 315             'note': 'Videos that embed the url in the player page',
 316             'info_dict': {
 317                 'id': '54469442',
 318                 'ext': 'mp4',
 319                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 320                 'uploader': 'The BLN & Business of Software',
 321                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 322                 'uploader_id': 'theblnbusinessofsoftware',
 323                 'duration': 3610,
 324                 'description': None,
 325             },
 326         },
 327         {
 328             'url': 'http://vimeo.com/68375962',
 329             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 330             'note': 'Video protected with password',
 331             'info_dict': {
 332                 'id': '68375962',
 333                 'ext': 'mp4',
 334                 'title': 'youtube-dl password protected test video',
 335                 'timestamp': 1371200155,
 336                 'upload_date': '20130614',
 337                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 338                 'uploader_id': 'user18948128',
 339                 'uploader': 'Jaime Marquínez Ferrándiz',
 340                 'duration': 10,
 341                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 342             },
 343             'params': {
 344                 'videopassword': 'youtube-dl',
 345             },
 346         },
 347         {
 348             'url': 'http://vimeo.com/channels/keypeele/75629013',
 349             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 350             'info_dict': {
 351                 'id': '75629013',
 352                 'ext': 'mp4',
 353                 'title': 'Key & Peele: Terrorist Interrogation',
 354                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 355                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 356                 'uploader_id': 'atencio',
 357                 'uploader': 'Peter Atencio',
 358                 'channel_id': 'keypeele',
 359                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/keypeele',
 360                 'timestamp': 1380339469,
 361                 'upload_date': '20130928',
 362                 'duration': 187,
 363             },
 364             'expected_warnings': ['Unable to download JSON metadata'],
 365         },
 366         {
 367             'url': 'http://vimeo.com/76979871',
 368             'note': 'Video with subtitles',
 369             'info_dict': {
 370                 'id': '76979871',
 371                 'ext': 'mp4',
 372                 'title': 'The New Vimeo Player (You Know, For Videos)',
 373                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 374                 'timestamp': 1381846109,
 375                 'upload_date': '20131015',
 376                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 377                 'uploader_id': 'staff',
 378                 'uploader': 'Vimeo Staff',
 379                 'duration': 62,
 380             }
 381         },
 382         {
 383             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 384             'url': 'https://player.vimeo.com/video/98044508',
 385             'note': 'The js code contains assignments to the same variable as the config',
 386             'info_dict': {
 387                 'id': '98044508',
 388                 'ext': 'mp4',
 389                 'title': 'Pier Solar OUYA Official Trailer',
 390                 'uploader': 'Tulio Gonçalves',
 391                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 392                 'uploader_id': 'user28849593',
 393             },
 394         },
 395         {
 396             # contains original format
 397             'url': 'https://vimeo.com/33951933',
 398             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 399             'info_dict': {
 400                 'id': '33951933',
 401                 'ext': 'mp4',
 402                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 403                 'uploader': 'The DMCI',
 404                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 405                 'uploader_id': 'dmci',
 406                 'timestamp': 1324343742,
 407                 'upload_date': '20111220',
 408                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 409             },
 410         },
 411         {
 412             # only available via https://vimeo.com/channels/tributes/6213729 and
 413             # not via https://vimeo.com/6213729
 414             'url': 'https://vimeo.com/channels/tributes/6213729',
 415             'info_dict': {
 416                 'id': '6213729',
 417                 'ext': 'mp4',
 418                 'title': 'Vimeo Tribute: The Shining',
 419                 'uploader': 'Casey Donahue',
 420                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 421                 'uploader_id': 'caseydonahue',
 422                 'channel_url': r're:https?://(?:www\.)?vimeo\.com/channels/tributes',
 423                 'channel_id': 'tributes',
 424                 'timestamp': 1250886430,
 425                 'upload_date': '20090821',
 426                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 427             },
 428             'params': {
 429                 'skip_download': True,
 430             },
 431             'expected_warnings': ['Unable to download JSON metadata'],
 432         },
 433         {
 434             # redirects to ondemand extractor and should be passed through it
 435             # for successful extraction
 436             'url': 'https://vimeo.com/73445910',
 437             'info_dict': {
 438                 'id': '73445910',
 439                 'ext': 'mp4',
 440                 'title': 'The Reluctant Revolutionary',
 441                 'uploader': '10Ft Films',
 442                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 443                 'uploader_id': 'tenfootfilms',
 444             },
 445             'params': {
 446                 'skip_download': True,
 447             },
 448         },
 449         {
 450             'url': 'http://player.vimeo.com/video/68375962',
 451             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 452             'info_dict': {
 453                 'id': '68375962',
 454                 'ext': 'mp4',
 455                 'title': 'youtube-dl password protected test video',
 456                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 457                 'uploader_id': 'user18948128',
 458                 'uploader': 'Jaime Marquínez Ferrándiz',
 459                 'duration': 10,
 460             },
 461             'params': {
 462                 'videopassword': 'youtube-dl',
 463             },
 464         },
 465         {
 466             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 467             'only_matching': True,
 468         },
 469         {
 470             'url': 'https://vimeo.com/109815029',
 471             'note': 'Video not completely processed, "failed" seed status',
 472             'only_matching': True,
 473         },
 474         {
 475             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 476             'only_matching': True,
 477         },
 478         {
 479             'url': 'https://vimeo.com/album/2632481/video/79010983',
 480             'only_matching': True,
 481         },
 482         {
 483             # source file returns 403: Forbidden
 484             'url': 'https://vimeo.com/7809605',
 485             'only_matching': True,
 486         },
 487         {
 488             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 489             'only_matching': True,
 490         }
 491         # https://gettingthingsdone.com/workflowmap/
 492         # vimeo embed with check-password page protected by Referer header
 493     ]
 494
 495     @staticmethod
 496     def _smuggle_referrer(url, referrer_url):
 497         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 498
 499     @staticmethod
 500     def _extract_urls(url, webpage):
 501         urls = []
 502         # Look for embedded (iframe) Vimeo player
 503         for mobj in re.finditer(
 504                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1',
 505                 webpage):
 506             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 507         PLAIN_EMBED_RE = (
 508             # Look for embedded (swf embed) Vimeo player
 509             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 510             # Look more for non-standard embedded Vimeo player
 511             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 512         )
 513         for embed_re in PLAIN_EMBED_RE:
 514             for mobj in re.finditer(embed_re, webpage):
 515                 urls.append(mobj.group('url'))
 516         return urls
 517
 518     @staticmethod
 519     def _extract_url(url, webpage):
 520         urls = VimeoIE._extract_urls(url, webpage)
 521         return urls[0] if urls else None
 522
 523     def _verify_player_video_password(self, url, video_id, headers):
 524         password = self._downloader.params.get('videopassword')
 525         if password is None:
 526             raise ExtractorError('This video is protected by a password, use the --video-password option')
 527         data = urlencode_postdata({
 528             'password': base64.b64encode(password.encode()),
 529         })
 530         headers = merge_dicts(headers, {
 531             'Content-Type': 'application/x-www-form-urlencoded',
 532         })
 533         checked = self._download_json(
 534             url + '/check-password', video_id,
 535             'Verifying the password', data=data, headers=headers)
 536         if checked is False:
 537             raise ExtractorError('Wrong video password', expected=True)
 538         return checked
 539
 540     def _real_initialize(self):
 541         self._login()
 542
 543     def _real_extract(self, url):
 544         url, data = unsmuggle_url(url, {})
 545         headers = std_headers.copy()
 546         if 'http_headers' in data:
 547             headers.update(data['http_headers'])
 548         if 'Referer' not in headers:
 549             headers['Referer'] = url
 550
 551         channel_id = self._search_regex(
 552             r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None)
 553
 554         # Extract ID from URL
 555         mobj = re.match(self._VALID_URL, url)
 556         video_id = mobj.group('id')
 557         orig_url = url
 558         if mobj.group('pro'):
 559             # some videos require portfolio_id to be present in player url
 560             # https://github.com/ytdl-org/youtube-dl/issues/20070
 561             url = self._extract_url(url, self._download_webpage(url, video_id))
 562         elif mobj.group('player'):
 563             url = 'https://player.vimeo.com/video/' + video_id
 564         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 565             url = 'https://vimeo.com/' + video_id
 566
 567         # Retrieve video webpage to extract further information
 568         request = sanitized_Request(url, headers=headers)
 569         try:
 570             webpage, urlh = self._download_webpage_handle(request, video_id)
 571             redirect_url = compat_str(urlh.geturl())
 572             # Some URLs redirect to ondemand can't be extracted with
 573             # this extractor right away thus should be passed through
 574             # ondemand extractor (e.g. https://vimeo.com/73445910)
 575             if VimeoOndemandIE.suitable(redirect_url):
 576                 return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
 577         except ExtractorError as ee:
 578             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 579                 errmsg = ee.cause.read()
 580                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 581                     raise ExtractorError(
 582                         'Cannot download embed-only video without embedding '
 583                         'URL. Please call youtube-dl with the URL of the page '
 584                         'that embeds this video.',
 585                         expected=True)
 586             raise
 587
 588         # Now we begin extracting as much information as we can from what we
 589         # retrieved. First we extract the information common to all extractors,
 590         # and latter we extract those that are Vimeo specific.
 591         self.report_extraction(video_id)
 592
 593         vimeo_config = self._extract_vimeo_config(webpage, video_id, default=None)
 594         if vimeo_config:
 595             seed_status = vimeo_config.get('seed_status', {})
 596             if seed_status.get('state') == 'failed':
 597                 raise ExtractorError(
 598                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 599                     expected=True)
 600
 601         cc_license = None
 602         timestamp = None
 603
 604         # Extract the config JSON
 605         try:
 606             try:
 607                 config_url = self._html_search_regex(
 608                     r' data-config-url="(.+?)"', webpage,
 609                     'config URL', default=None)
 610                 if not config_url:
 611                     # Sometimes new react-based page is served instead of old one that require
 612                     # different config URL extraction approach (see
 613                     # https://github.com/ytdl-org/youtube-dl/pull/7209)
 614                     vimeo_clip_page_config = self._search_regex(
 615                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 616                         'vimeo clip page config')
 617                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 618                     config_url = page_config['player']['config_url']
 619                     cc_license = page_config.get('cc_license')
 620                     timestamp = try_get(
 621                         page_config, lambda x: x['clip']['uploaded_on'],
 622                         compat_str)
 623                 config_json = self._download_webpage(config_url, video_id)
 624                 config = json.loads(config_json)
 625             except RegexNotFoundError:
 626                 # For pro videos or player.vimeo.com urls
 627                 # We try to find out to which variable is assigned the config dic
 628                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 629                 if m_variable_name is not None:
 630                     config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))]
 631                 else:
 632                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 633                 config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;')
 634                 config_re.append(r'\bconfig\s*=\s*({.+?})\s*;')
 635                 config = self._search_regex(config_re, webpage, 'info section',
 636                                             flags=re.DOTALL)
 637                 config = json.loads(config)
 638         except Exception as e:
 639             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 640                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 641
 642             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 643                 if '_video_password_verified' in data:
 644                     raise ExtractorError('video password verification failed!')
 645                 self._verify_video_password(redirect_url, video_id, webpage)
 646                 return self._real_extract(
 647                     smuggle_url(redirect_url, {'_video_password_verified': 'verified'}))
 648             else:
 649                 raise ExtractorError('Unable to extract info section',
 650                                      cause=e)
 651         else:
 652             if config.get('view') == 4:
 653                 config = self._verify_player_video_password(redirect_url, video_id, headers)
 654
 655         vod = config.get('video', {}).get('vod', {})
 656
 657         def is_rented():
 658             if '>You rented this title.<' in webpage:
 659                 return True
 660             if config.get('user', {}).get('purchased'):
 661                 return True
 662             for purchase_option in vod.get('purchase_options', []):
 663                 if purchase_option.get('purchased'):
 664                     return True
 665                 label = purchase_option.get('label_string')
 666                 if label and (label.startswith('You rented this') or label.endswith(' remaining')):
 667                     return True
 668             return False
 669
 670         if is_rented() and vod.get('is_trailer'):
 671             feature_id = vod.get('feature_id')
 672             if feature_id and not data.get('force_feature_id', False):
 673                 return self.url_result(smuggle_url(
 674                     'https://player.vimeo.com/player/%s' % feature_id,
 675                     {'force_feature_id': True}), 'Vimeo')
 676
 677         # Extract video description
 678
 679         video_description = self._html_search_regex(
 680             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 681             webpage, 'description', default=None)
 682         if not video_description:
 683             video_description = self._html_search_meta(
 684                 'description', webpage, default=None)
 685         if not video_description and mobj.group('pro'):
 686             orig_webpage = self._download_webpage(
 687                 orig_url, video_id,
 688                 note='Downloading webpage for description',
 689                 fatal=False)
 690             if orig_webpage:
 691                 video_description = self._html_search_meta(
 692                     'description', orig_webpage, default=None)
 693         if not video_description and not mobj.group('player'):
 694             self._downloader.report_warning('Cannot find video description')
 695
 696         # Extract upload date
 697         if not timestamp:
 698             timestamp = self._search_regex(
 699                 r'<time[^>]+datetime="([^"]+)"', webpage,
 700                 'timestamp', default=None)
 701
 702         try:
 703             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 704             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 705             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 706         except RegexNotFoundError:
 707             # This info is only available in vimeo.com/{id} urls
 708             view_count = None
 709             like_count = None
 710             comment_count = None
 711
 712         formats = []
 713
 714         source_format = self._extract_original_format(
 715             'https://vimeo.com/' + video_id, video_id)
 716         if source_format:
 717             formats.append(source_format)
 718
 719         info_dict_config = self._parse_config(config, video_id)
 720         formats.extend(info_dict_config['formats'])
 721         self._vimeo_sort_formats(formats)
 722
 723         json_ld = self._search_json_ld(webpage, video_id, default={})
 724
 725         if not cc_license:
 726             cc_license = self._search_regex(
 727                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 728                 webpage, 'license', default=None, group='license')
 729
 730         channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None
 731
 732         info_dict = {
 733             'formats': formats,
 734             'timestamp': unified_timestamp(timestamp),
 735             'description': video_description,
 736             'webpage_url': url,
 737             'view_count': view_count,
 738             'like_count': like_count,
 739             'comment_count': comment_count,
 740             'license': cc_license,
 741             'channel_id': channel_id,
 742             'channel_url': channel_url,
 743         }
 744
 745         info_dict = merge_dicts(info_dict, info_dict_config, json_ld)
 746
 747         return info_dict
 748
 749
 750 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 751     IE_NAME = 'vimeo:ondemand'
 752     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 753     _TESTS = [{
 754         # ondemand video not available via https://vimeo.com/id
 755         'url': 'https://vimeo.com/ondemand/20704',
 756         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 757         'info_dict': {
 758             'id': '105442900',
 759             'ext': 'mp4',
 760             'title': 'המעבדה - במאי יותם פלדמן',
 761             'uploader': 'גם סרטים',
 762             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 763             'uploader_id': 'gumfilms',
 764         },
 765         'params': {
 766             'format': 'best[protocol=https]',
 767         },
 768     }, {
 769         # requires Referer to be passed along with og:video:url
 770         'url': 'https://vimeo.com/ondemand/36938/126682985',
 771         'info_dict': {
 772             'id': '126682985',
 773             'ext': 'mp4',
 774             'title': 'Rävlock, rätt läte på rätt plats',
 775             'uploader': 'Lindroth & Norin',
 776             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 777             'uploader_id': 'user14430847',
 778         },
 779         'params': {
 780             'skip_download': True,
 781         },
 782     }, {
 783         'url': 'https://vimeo.com/ondemand/nazmaalik',
 784         'only_matching': True,
 785     }, {
 786         'url': 'https://vimeo.com/ondemand/141692381',
 787         'only_matching': True,
 788     }, {
 789         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 790         'only_matching': True,
 791     }]
 792
 793     def _real_extract(self, url):
 794         video_id = self._match_id(url)
 795         webpage = self._download_webpage(url, video_id)
 796         return self.url_result(
 797             # Some videos require Referer to be passed along with og:video:url
 798             # similarly to generic vimeo embeds (e.g.
 799             # https://vimeo.com/ondemand/36938/126682985).
 800             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 801             VimeoIE.ie_key())
 802
 803
 804 class VimeoChannelIE(VimeoBaseInfoExtractor):
 805     IE_NAME = 'vimeo:channel'
 806     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 807     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 808     _TITLE = None
 809     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 810     _TESTS = [{
 811         'url': 'https://vimeo.com/channels/tributes',
 812         'info_dict': {
 813             'id': 'tributes',
 814             'title': 'Vimeo Tributes',
 815         },
 816         'playlist_mincount': 25,
 817     }]
 818
 819     def _page_url(self, base_url, pagenum):
 820         return '%s/videos/page:%d/' % (base_url, pagenum)
 821
 822     def _extract_list_title(self, webpage):
 823         return self._TITLE or self._html_search_regex(
 824             self._TITLE_RE, webpage, 'list title', fatal=False)
 825
 826     def _login_list_password(self, page_url, list_id, webpage):
 827         login_form = self._search_regex(
 828             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 829             webpage, 'login form', default=None)
 830         if not login_form:
 831             return webpage
 832
 833         password = self._downloader.params.get('videopassword')
 834         if password is None:
 835             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 836         fields = self._hidden_inputs(login_form)
 837         token, vuid = self._extract_xsrft_and_vuid(webpage)
 838         fields['token'] = token
 839         fields['password'] = password
 840         post = urlencode_postdata(fields)
 841         password_path = self._search_regex(
 842             r'action="([^"]+)"', login_form, 'password URL')
 843         password_url = compat_urlparse.urljoin(page_url, password_path)
 844         password_request = sanitized_Request(password_url, post)
 845         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 846         self._set_vimeo_cookie('vuid', vuid)
 847         self._set_vimeo_cookie('xsrft', token)
 848
 849         return self._download_webpage(
 850             password_request, list_id,
 851             'Verifying the password', 'Wrong password')
 852
 853     def _title_and_entries(self, list_id, base_url):
 854         for pagenum in itertools.count(1):
 855             page_url = self._page_url(base_url, pagenum)
 856             webpage = self._download_webpage(
 857                 page_url, list_id,
 858                 'Downloading page %s' % pagenum)
 859
 860             if pagenum == 1:
 861                 webpage = self._login_list_password(page_url, list_id, webpage)
 862                 yield self._extract_list_title(webpage)
 863
 864             # Try extracting href first since not all videos are available via
 865             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 866             clips = re.findall(
 867                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 868             if clips:
 869                 for video_id, video_url, video_title in clips:
 870                     yield self.url_result(
 871                         compat_urlparse.urljoin(base_url, video_url),
 872                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 873             # More relaxed fallback
 874             else:
 875                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 876                     yield self.url_result(
 877                         'https://vimeo.com/%s' % video_id,
 878                         VimeoIE.ie_key(), video_id=video_id)
 879
 880             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 881                 break
 882
 883     def _extract_videos(self, list_id, base_url):
 884         title_and_entries = self._title_and_entries(list_id, base_url)
 885         list_title = next(title_and_entries)
 886         return self.playlist_result(title_and_entries, list_id, list_title)
 887
 888     def _real_extract(self, url):
 889         mobj = re.match(self._VALID_URL, url)
 890         channel_id = mobj.group('id')
 891         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 892
 893
 894 class VimeoUserIE(VimeoChannelIE):
 895     IE_NAME = 'vimeo:user'
 896     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 897     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 898     _TESTS = [{
 899         'url': 'https://vimeo.com/nkistudio/videos',
 900         'info_dict': {
 901             'title': 'Nki',
 902             'id': 'nkistudio',
 903         },
 904         'playlist_mincount': 66,
 905     }]
 906
 907     def _real_extract(self, url):
 908         mobj = re.match(self._VALID_URL, url)
 909         name = mobj.group('name')
 910         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 911
 912
 913 class VimeoAlbumIE(VimeoChannelIE):
 914     IE_NAME = 'vimeo:album'
 915     _VALID_URL = r'https://vimeo\.com/(?:album|showcase)/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 916     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 917     _TESTS = [{
 918         'url': 'https://vimeo.com/album/2632481',
 919         'info_dict': {
 920             'id': '2632481',
 921             'title': 'Staff Favorites: November 2013',
 922         },
 923         'playlist_mincount': 13,
 924     }, {
 925         'note': 'Password-protected album',
 926         'url': 'https://vimeo.com/album/3253534',
 927         'info_dict': {
 928             'title': 'test',
 929             'id': '3253534',
 930         },
 931         'playlist_count': 1,
 932         'params': {
 933             'videopassword': 'youtube-dl',
 934         }
 935     }]
 936     _PAGE_SIZE = 100
 937
 938     def _fetch_page(self, album_id, authorizaion, hashed_pass, page):
 939         api_page = page + 1
 940         query = {
 941             'fields': 'link',
 942             'page': api_page,
 943             'per_page': self._PAGE_SIZE,
 944         }
 945         if hashed_pass:
 946             query['_hashed_pass'] = hashed_pass
 947         videos = self._download_json(
 948             'https://api.vimeo.com/albums/%s/videos' % album_id,
 949             album_id, 'Downloading page %d' % api_page, query=query, headers={
 950                 'Authorization': 'jwt ' + authorizaion,
 951             })['data']
 952         for video in videos:
 953             link = video.get('link')
 954             if not link:
 955                 continue
 956             yield self.url_result(link, VimeoIE.ie_key(), VimeoIE._match_id(link))
 957
 958     def _real_extract(self, url):
 959         album_id = self._match_id(url)
 960         webpage = self._download_webpage(url, album_id)
 961         webpage = self._login_list_password(url, album_id, webpage)
 962         api_config = self._extract_vimeo_config(webpage, album_id)['api']
 963         entries = OnDemandPagedList(functools.partial(
 964             self._fetch_page, album_id, api_config['jwt'],
 965             api_config.get('hashed_pass')), self._PAGE_SIZE)
 966         return self.playlist_result(entries, album_id, self._html_search_regex(
 967             r'<title>\s*(.+?)(?:\s+on Vimeo)?</title>', webpage, 'title', fatal=False))
 968
 969
 970 class VimeoGroupsIE(VimeoAlbumIE):
 971     IE_NAME = 'vimeo:group'
 972     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 973     _TESTS = [{
 974         'url': 'https://vimeo.com/groups/rolexawards',
 975         'info_dict': {
 976             'id': 'rolexawards',
 977             'title': 'Rolex Awards for Enterprise',
 978         },
 979         'playlist_mincount': 73,
 980     }]
 981
 982     def _extract_list_title(self, webpage):
 983         return self._og_search_title(webpage, fatal=False)
 984
 985     def _real_extract(self, url):
 986         mobj = re.match(self._VALID_URL, url)
 987         name = mobj.group('name')
 988         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 989
 990
 991 class VimeoReviewIE(VimeoBaseInfoExtractor):
 992     IE_NAME = 'vimeo:review'
 993     IE_DESC = 'Review pages on vimeo'
 994     _VALID_URL = r'(?P<url>https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)/[0-9a-f]{10})'
 995     _TESTS = [{
 996         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 997         'md5': 'c507a72f780cacc12b2248bb4006d253',
 998         'info_dict': {
 999             'id': '75524534',
1000             'ext': 'mp4',
1001             'title': "DICK HARDWICK 'Comedian'",
1002             'uploader': 'Richard Hardwick',
1003             'uploader_id': 'user21297594',
1004         }
1005     }, {
1006         'note': 'video player needs Referer',
1007         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
1008         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
1009         'info_dict': {
1010             'id': '91613211',
1011             'ext': 'mp4',
1012             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
1013             'uploader': 'DevWeek Events',
1014             'duration': 2773,
1015             'thumbnail': r're:^https?://.*\.jpg$',
1016             'uploader_id': 'user22258446',
1017         }
1018     }, {
1019         'note': 'Password protected',
1020         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
1021         'info_dict': {
1022             'id': '138823582',
1023             'ext': 'mp4',
1024             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
1025             'uploader': 'TMB',
1026             'uploader_id': 'user37284429',
1027         },
1028         'params': {
1029             'videopassword': 'holygrail',
1030         },
1031         'skip': 'video gone',
1032     }]
1033
1034     def _real_initialize(self):
1035         self._login()
1036
1037     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
1038         webpage = self._download_webpage(webpage_url, video_id)
1039         config_url = self._html_search_regex(
1040             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
1041             'config URL', default=None, group='url')
1042         if not config_url:
1043             data = self._parse_json(self._search_regex(
1044                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
1045                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
1046             config = data.get('vimeo_esi', {}).get('config', {})
1047             config_url = config.get('configUrl') or try_get(config, lambda x: x['clipData']['configUrl'])
1048         if config_url is None:
1049             self._verify_video_password(webpage_url, video_id, webpage)
1050             config_url = self._get_config_url(
1051                 webpage_url, video_id, video_password_verified=True)
1052         return config_url
1053
1054     def _real_extract(self, url):
1055         page_url, video_id = re.match(self._VALID_URL, url).groups()
1056         config_url = self._get_config_url(url, video_id)
1057         config = self._download_json(config_url, video_id)
1058         info_dict = self._parse_config(config, video_id)
1059         source_format = self._extract_original_format(page_url, video_id)
1060         if source_format:
1061             info_dict['formats'].append(source_format)
1062         self._vimeo_sort_formats(info_dict['formats'])
1063         return info_dict
1064
1065
1066 class VimeoWatchLaterIE(VimeoChannelIE):
1067     IE_NAME = 'vimeo:watchlater'
1068     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
1069     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
1070     _TITLE = 'Watch Later'
1071     _LOGIN_REQUIRED = True
1072     _TESTS = [{
1073         'url': 'https://vimeo.com/watchlater',
1074         'only_matching': True,
1075     }]
1076
1077     def _real_initialize(self):
1078         self._login()
1079
1080     def _page_url(self, base_url, pagenum):
1081         url = '%s/page:%d/' % (base_url, pagenum)
1082         request = sanitized_Request(url)
1083         # Set the header to get a partial html page with the ids,
1084         # the normal page doesn't contain them.
1085         request.add_header('X-Requested-With', 'XMLHttpRequest')
1086         return request
1087
1088     def _real_extract(self, url):
1089         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
1090
1091
1092 class VimeoLikesIE(VimeoChannelIE):
1093     _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)'
1094     IE_NAME = 'vimeo:likes'
1095     IE_DESC = 'Vimeo user likes'
1096     _TESTS = [{
1097         'url': 'https://vimeo.com/user755559/likes/',
1098         'playlist_mincount': 293,
1099         'info_dict': {
1100             'id': 'user755559',
1101             'title': 'urza’s Likes',
1102         },
1103     }, {
1104         'url': 'https://vimeo.com/stormlapse/likes',
1105         'only_matching': True,
1106     }]
1107
1108     def _page_url(self, base_url, pagenum):
1109         return '%s/page:%d/' % (base_url, pagenum)
1110
1111     def _real_extract(self, url):
1112         user_id = self._match_id(url)
1113         return self._extract_videos(user_id, 'https://vimeo.com/%s/likes' % user_id)
1114
1115
1116 class VHXEmbedIE(VimeoBaseInfoExtractor):
1117     IE_NAME = 'vhx:embed'
1118     _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P<id>\d+)'
1119
1120     def _real_extract(self, url):
1121         video_id = self._match_id(url)
1122         webpage = self._download_webpage(url, video_id)
1123         config_url = self._parse_json(self._search_regex(
1124             r'window\.OTTData\s*=\s*({.+})', webpage,
1125             'ott data'), video_id, js_to_json)['config_url']
1126         config = self._download_json(config_url, video_id)
1127         info = self._parse_config(config, video_id)
1128         self._vimeo_sort_formats(info['formats'])
1129         return info