youtube_dl/extractor/vimeo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_str,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     InAdvancePagedList,
  18     int_or_none,
  19     NO_DEFAULT,
  20     RegexNotFoundError,
  21     sanitized_Request,
  22     smuggle_url,
  23     std_headers,
  24     unified_strdate,
  25     unsmuggle_url,
  26     urlencode_postdata,
  27     unescapeHTML,
  28     parse_filesize,
  29     try_get,
  30 )
  31
  32
  33 class VimeoBaseInfoExtractor(InfoExtractor):
  34     _NETRC_MACHINE = 'vimeo'
  35     _LOGIN_REQUIRED = False
  36     _LOGIN_URL = 'https://vimeo.com/log_in'
  37
  38     def _login(self):
  39         (username, password) = self._get_login_info()
  40         if username is None:
  41             if self._LOGIN_REQUIRED:
  42                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  43             return
  44         self.report_login()
  45         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  46         token, vuid = self._extract_xsrft_and_vuid(webpage)
  47         data = urlencode_postdata({
  48             'action': 'login',
  49             'email': username,
  50             'password': password,
  51             'service': 'vimeo',
  52             'token': token,
  53         })
  54         login_request = sanitized_Request(self._LOGIN_URL, data)
  55         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  56         login_request.add_header('Referer', self._LOGIN_URL)
  57         self._set_vimeo_cookie('vuid', vuid)
  58         self._download_webpage(login_request, None, False, 'Wrong login info')
  59
  60     def _verify_video_password(self, url, video_id, webpage):
  61         password = self._downloader.params.get('videopassword')
  62         if password is None:
  63             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  64         token, vuid = self._extract_xsrft_and_vuid(webpage)
  65         data = urlencode_postdata({
  66             'password': password,
  67             'token': token,
  68         })
  69         if url.startswith('http://'):
  70             # vimeo only supports https now, but the user can give an http url
  71             url = url.replace('http://', 'https://')
  72         password_request = sanitized_Request(url + '/password', data)
  73         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  74         password_request.add_header('Referer', url)
  75         self._set_vimeo_cookie('vuid', vuid)
  76         return self._download_webpage(
  77             password_request, video_id,
  78             'Verifying the password', 'Wrong password')
  79
  80     def _extract_xsrft_and_vuid(self, webpage):
  81         xsrft = self._search_regex(
  82             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  83             webpage, 'login token', group='xsrft')
  84         vuid = self._search_regex(
  85             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  86             webpage, 'vuid', group='vuid')
  87         return xsrft, vuid
  88
  89     def _set_vimeo_cookie(self, name, value):
  90         self._set_cookie('vimeo.com', name, value)
  91
  92     def _vimeo_sort_formats(self, formats):
  93         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
  94         # at the same time without actual units specified. This lead to wrong sorting.
  95         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
  96
  97     def _parse_config(self, config, video_id):
  98         video_data = config['video']
  99         # Extract title
 100         video_title = video_data['title']
 101
 102         # Extract uploader, uploader_url and uploader_id
 103         video_uploader = video_data.get('owner', {}).get('name')
 104         video_uploader_url = video_data.get('owner', {}).get('url')
 105         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 106
 107         # Extract video thumbnail
 108         video_thumbnail = video_data.get('thumbnail')
 109         if video_thumbnail is None:
 110             video_thumbs = video_data.get('thumbs')
 111             if video_thumbs and isinstance(video_thumbs, dict):
 112                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 113
 114         # Extract video duration
 115         video_duration = int_or_none(video_data.get('duration'))
 116
 117         formats = []
 118         config_files = video_data.get('files') or config['request'].get('files', {})
 119         for f in config_files.get('progressive', []):
 120             video_url = f.get('url')
 121             if not video_url:
 122                 continue
 123             formats.append({
 124                 'url': video_url,
 125                 'format_id': 'http-%s' % f.get('quality'),
 126                 'width': int_or_none(f.get('width')),
 127                 'height': int_or_none(f.get('height')),
 128                 'fps': int_or_none(f.get('fps')),
 129                 'tbr': int_or_none(f.get('bitrate')),
 130             })
 131
 132         for files_type in ('hls', 'dash'):
 133             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 134                 manifest_url = cdn_data.get('url')
 135                 if not manifest_url:
 136                     continue
 137                 format_id = '%s-%s' % (files_type, cdn_name)
 138                 if files_type == 'hls':
 139                     formats.extend(self._extract_m3u8_formats(
 140                         manifest_url, video_id, 'mp4',
 141                         'm3u8_native', m3u8_id=format_id,
 142                         note='Downloading %s m3u8 information' % cdn_name,
 143                         fatal=False))
 144                 elif files_type == 'dash':
 145                     formats.extend(self._extract_mpd_formats(
 146                         manifest_url.replace('/master.json', '/master.mpd'), video_id, format_id,
 147                         'Downloading %s MPD information' % cdn_name,
 148                         fatal=False))
 149
 150         subtitles = {}
 151         text_tracks = config['request'].get('text_tracks')
 152         if text_tracks:
 153             for tt in text_tracks:
 154                 subtitles[tt['lang']] = [{
 155                     'ext': 'vtt',
 156                     'url': 'https://vimeo.com' + tt['url'],
 157                 }]
 158
 159         return {
 160             'title': video_title,
 161             'uploader': video_uploader,
 162             'uploader_id': video_uploader_id,
 163             'uploader_url': video_uploader_url,
 164             'thumbnail': video_thumbnail,
 165             'duration': video_duration,
 166             'formats': formats,
 167             'subtitles': subtitles,
 168         }
 169
 170
 171 class VimeoIE(VimeoBaseInfoExtractor):
 172     """Information extractor for vimeo.com."""
 173
 174     # _VALID_URL matches Vimeo URLs
 175     _VALID_URL = r'''(?x)
 176                     https?://
 177                         (?:
 178                             (?:
 179                                 www|
 180                                 (?P<player>player)
 181                             )
 182                             \.
 183                         )?
 184                         vimeo(?P<pro>pro)?\.com/
 185                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 186                         (?:.*?/)?
 187                         (?:
 188                             (?:
 189                                 play_redirect_hls|
 190                                 moogaloop\.swf)\?clip_id=
 191                             )?
 192                         (?:videos?/)?
 193                         (?P<id>[0-9]+)
 194                         (?:/[\da-f]+)?
 195                         /?(?:[?&].*)?(?:[#].*)?$
 196                     '''
 197     IE_NAME = 'vimeo'
 198     _TESTS = [
 199         {
 200             'url': 'http://vimeo.com/56015672#at=0',
 201             'md5': '8879b6cc097e987f02484baf890129e5',
 202             'info_dict': {
 203                 'id': '56015672',
 204                 'ext': 'mp4',
 205                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 206                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 207                 'upload_date': '20121220',
 208                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user7108434',
 209                 'uploader_id': 'user7108434',
 210                 'uploader': 'Filippo Valsorda',
 211                 'duration': 10,
 212             },
 213         },
 214         {
 215             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 216             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 217             'note': 'Vimeo Pro video (#1197)',
 218             'info_dict': {
 219                 'id': '68093876',
 220                 'ext': 'mp4',
 221                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 222                 'uploader_id': 'openstreetmapus',
 223                 'uploader': 'OpenStreetMap US',
 224                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 225                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 226                 'duration': 1595,
 227             },
 228         },
 229         {
 230             'url': 'http://player.vimeo.com/video/54469442',
 231             'md5': '619b811a4417aa4abe78dc653becf511',
 232             'note': 'Videos that embed the url in the player page',
 233             'info_dict': {
 234                 'id': '54469442',
 235                 'ext': 'mp4',
 236                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 237                 'uploader': 'The BLN & Business of Software',
 238                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 239                 'uploader_id': 'theblnbusinessofsoftware',
 240                 'duration': 3610,
 241                 'description': None,
 242             },
 243         },
 244         {
 245             'url': 'http://vimeo.com/68375962',
 246             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 247             'note': 'Video protected with password',
 248             'info_dict': {
 249                 'id': '68375962',
 250                 'ext': 'mp4',
 251                 'title': 'youtube-dl password protected test video',
 252                 'upload_date': '20130614',
 253                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user18948128',
 254                 'uploader_id': 'user18948128',
 255                 'uploader': 'Jaime Marquínez Ferrándiz',
 256                 'duration': 10,
 257                 'description': 'This is "youtube-dl password protected test video" by  on Vimeo, the home for high quality videos and the people who love them.',
 258             },
 259             'params': {
 260                 'videopassword': 'youtube-dl',
 261             },
 262         },
 263         {
 264             'url': 'http://vimeo.com/channels/keypeele/75629013',
 265             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 266             'info_dict': {
 267                 'id': '75629013',
 268                 'ext': 'mp4',
 269                 'title': 'Key & Peele: Terrorist Interrogation',
 270                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 271                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/atencio',
 272                 'uploader_id': 'atencio',
 273                 'uploader': 'Peter Atencio',
 274                 'upload_date': '20130927',
 275                 'duration': 187,
 276             },
 277         },
 278         {
 279             'url': 'http://vimeo.com/76979871',
 280             'note': 'Video with subtitles',
 281             'info_dict': {
 282                 'id': '76979871',
 283                 'ext': 'mp4',
 284                 'title': 'The New Vimeo Player (You Know, For Videos)',
 285                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 286                 'upload_date': '20131015',
 287                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/staff',
 288                 'uploader_id': 'staff',
 289                 'uploader': 'Vimeo Staff',
 290                 'duration': 62,
 291             }
 292         },
 293         {
 294             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 295             'url': 'https://player.vimeo.com/video/98044508',
 296             'note': 'The js code contains assignments to the same variable as the config',
 297             'info_dict': {
 298                 'id': '98044508',
 299                 'ext': 'mp4',
 300                 'title': 'Pier Solar OUYA Official Trailer',
 301                 'uploader': 'Tulio Gonçalves',
 302                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user28849593',
 303                 'uploader_id': 'user28849593',
 304             },
 305         },
 306         {
 307             # contains original format
 308             'url': 'https://vimeo.com/33951933',
 309             'md5': '2d9f5475e0537f013d0073e812ab89e6',
 310             'info_dict': {
 311                 'id': '33951933',
 312                 'ext': 'mp4',
 313                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 314                 'uploader': 'The DMCI',
 315                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/dmci',
 316                 'uploader_id': 'dmci',
 317                 'upload_date': '20111220',
 318                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 319             },
 320         },
 321         {
 322             # only available via https://vimeo.com/channels/tributes/6213729 and
 323             # not via https://vimeo.com/6213729
 324             'url': 'https://vimeo.com/channels/tributes/6213729',
 325             'info_dict': {
 326                 'id': '6213729',
 327                 'ext': 'mp4',
 328                 'title': 'Vimeo Tribute: The Shining',
 329                 'uploader': 'Casey Donahue',
 330                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 331                 'uploader_id': 'caseydonahue',
 332                 'upload_date': '20090821',
 333                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 334             },
 335             'params': {
 336                 'skip_download': True,
 337             },
 338             'expected_warnings': ['Unable to download JSON metadata'],
 339         },
 340         {
 341             # redirects to ondemand extractor and should be passed throught it
 342             # for successful extraction
 343             'url': 'https://vimeo.com/73445910',
 344             'info_dict': {
 345                 'id': '73445910',
 346                 'ext': 'mp4',
 347                 'title': 'The Reluctant Revolutionary',
 348                 'uploader': '10Ft Films',
 349                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 350                 'uploader_id': 'tenfootfilms',
 351             },
 352             'params': {
 353                 'skip_download': True,
 354             },
 355         },
 356         {
 357             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 358             'only_matching': True,
 359         },
 360         {
 361             'url': 'https://vimeo.com/109815029',
 362             'note': 'Video not completely processed, "failed" seed status',
 363             'only_matching': True,
 364         },
 365         {
 366             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 367             'only_matching': True,
 368         },
 369         {
 370             'url': 'https://vimeo.com/album/2632481/video/79010983',
 371             'only_matching': True,
 372         },
 373         {
 374             # source file returns 403: Forbidden
 375             'url': 'https://vimeo.com/7809605',
 376             'only_matching': True,
 377         },
 378         {
 379             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 380             'only_matching': True,
 381         }
 382     ]
 383
 384     @staticmethod
 385     def _smuggle_referrer(url, referrer_url):
 386         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 387
 388     @staticmethod
 389     def _extract_urls(url, webpage):
 390         urls = []
 391         # Look for embedded (iframe) Vimeo player
 392         for mobj in re.finditer(
 393                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1',
 394                 webpage):
 395             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 396         PLAIN_EMBED_RE = (
 397             # Look for embedded (swf embed) Vimeo player
 398             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 399             # Look more for non-standard embedded Vimeo player
 400             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 401         )
 402         for embed_re in PLAIN_EMBED_RE:
 403             for mobj in re.finditer(embed_re, webpage):
 404                 urls.append(mobj.group('url'))
 405         return urls
 406
 407     @staticmethod
 408     def _extract_url(url, webpage):
 409         urls = VimeoIE._extract_urls(url, webpage)
 410         return urls[0] if urls else None
 411
 412     def _verify_player_video_password(self, url, video_id):
 413         password = self._downloader.params.get('videopassword')
 414         if password is None:
 415             raise ExtractorError('This video is protected by a password, use the --video-password option')
 416         data = urlencode_postdata({'password': password})
 417         pass_url = url + '/check-password'
 418         password_request = sanitized_Request(pass_url, data)
 419         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 420         password_request.add_header('Referer', url)
 421         return self._download_json(
 422             password_request, video_id,
 423             'Verifying the password', 'Wrong password')
 424
 425     def _real_initialize(self):
 426         self._login()
 427
 428     def _real_extract(self, url):
 429         url, data = unsmuggle_url(url, {})
 430         headers = std_headers.copy()
 431         if 'http_headers' in data:
 432             headers.update(data['http_headers'])
 433         if 'Referer' not in headers:
 434             headers['Referer'] = url
 435
 436         # Extract ID from URL
 437         mobj = re.match(self._VALID_URL, url)
 438         video_id = mobj.group('id')
 439         orig_url = url
 440         if mobj.group('pro') or mobj.group('player'):
 441             url = 'https://player.vimeo.com/video/' + video_id
 442         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 443             url = 'https://vimeo.com/' + video_id
 444
 445         # Retrieve video webpage to extract further information
 446         request = sanitized_Request(url, headers=headers)
 447         try:
 448             webpage, urlh = self._download_webpage_handle(request, video_id)
 449             # Some URLs redirect to ondemand can't be extracted with
 450             # this extractor right away thus should be passed through
 451             # ondemand extractor (e.g. https://vimeo.com/73445910)
 452             if VimeoOndemandIE.suitable(urlh.geturl()):
 453                 return self.url_result(urlh.geturl(), VimeoOndemandIE.ie_key())
 454         except ExtractorError as ee:
 455             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 456                 errmsg = ee.cause.read()
 457                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 458                     raise ExtractorError(
 459                         'Cannot download embed-only video without embedding '
 460                         'URL. Please call youtube-dl with the URL of the page '
 461                         'that embeds this video.',
 462                         expected=True)
 463             raise
 464
 465         # Now we begin extracting as much information as we can from what we
 466         # retrieved. First we extract the information common to all extractors,
 467         # and latter we extract those that are Vimeo specific.
 468         self.report_extraction(video_id)
 469
 470         vimeo_config = self._search_regex(
 471             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 472             'vimeo config', default=None)
 473         if vimeo_config:
 474             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 475             if seed_status.get('state') == 'failed':
 476                 raise ExtractorError(
 477                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 478                     expected=True)
 479
 480         # Extract the config JSON
 481         try:
 482             try:
 483                 config_url = self._html_search_regex(
 484                     r' data-config-url="(.+?)"', webpage,
 485                     'config URL', default=None)
 486                 if not config_url:
 487                     # Sometimes new react-based page is served instead of old one that require
 488                     # different config URL extraction approach (see
 489                     # https://github.com/rg3/youtube-dl/pull/7209)
 490                     vimeo_clip_page_config = self._search_regex(
 491                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 492                         'vimeo clip page config')
 493                     config_url = self._parse_json(
 494                         vimeo_clip_page_config, video_id)['player']['config_url']
 495                 config_json = self._download_webpage(config_url, video_id)
 496                 config = json.loads(config_json)
 497             except RegexNotFoundError:
 498                 # For pro videos or player.vimeo.com urls
 499                 # We try to find out to which variable is assigned the config dic
 500                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 501                 if m_variable_name is not None:
 502                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 503                 else:
 504                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 505                 config = self._search_regex(config_re, webpage, 'info section',
 506                                             flags=re.DOTALL)
 507                 config = json.loads(config)
 508         except Exception as e:
 509             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 510                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 511
 512             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 513                 if '_video_password_verified' in data:
 514                     raise ExtractorError('video password verification failed!')
 515                 self._verify_video_password(url, video_id, webpage)
 516                 return self._real_extract(
 517                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 518             else:
 519                 raise ExtractorError('Unable to extract info section',
 520                                      cause=e)
 521         else:
 522             if config.get('view') == 4:
 523                 config = self._verify_player_video_password(url, video_id)
 524
 525         def is_rented():
 526             if '>You rented this title.<' in webpage:
 527                 return True
 528             if config.get('user', {}).get('purchased'):
 529                 return True
 530             label = try_get(
 531                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 532             if label and label.startswith('You rented this'):
 533                 return True
 534             return False
 535
 536         if is_rented():
 537             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 538             if feature_id and not data.get('force_feature_id', False):
 539                 return self.url_result(smuggle_url(
 540                     'https://player.vimeo.com/player/%s' % feature_id,
 541                     {'force_feature_id': True}), 'Vimeo')
 542
 543         # Extract video description
 544
 545         video_description = self._html_search_regex(
 546             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 547             webpage, 'description', default=None)
 548         if not video_description:
 549             video_description = self._html_search_meta(
 550                 'description', webpage, default=None)
 551         if not video_description and mobj.group('pro'):
 552             orig_webpage = self._download_webpage(
 553                 orig_url, video_id,
 554                 note='Downloading webpage for description',
 555                 fatal=False)
 556             if orig_webpage:
 557                 video_description = self._html_search_meta(
 558                     'description', orig_webpage, default=None)
 559         if not video_description and not mobj.group('player'):
 560             self._downloader.report_warning('Cannot find video description')
 561
 562         # Extract upload date
 563         video_upload_date = None
 564         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 565         if mobj is not None:
 566             video_upload_date = unified_strdate(mobj.group(1))
 567
 568         try:
 569             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 570             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 571             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 572         except RegexNotFoundError:
 573             # This info is only available in vimeo.com/{id} urls
 574             view_count = None
 575             like_count = None
 576             comment_count = None
 577
 578         formats = []
 579         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 580             'X-Requested-With': 'XMLHttpRequest'})
 581         download_data = self._download_json(download_request, video_id, fatal=False)
 582         if download_data:
 583             source_file = download_data.get('source_file')
 584             if isinstance(source_file, dict):
 585                 download_url = source_file.get('download_url')
 586                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 587                     source_name = source_file.get('public_name', 'Original')
 588                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 589                         ext = source_file.get('extension', determine_ext(download_url)).lower()
 590                         formats.append({
 591                             'url': download_url,
 592                             'ext': ext,
 593                             'width': int_or_none(source_file.get('width')),
 594                             'height': int_or_none(source_file.get('height')),
 595                             'filesize': parse_filesize(source_file.get('size')),
 596                             'format_id': source_name,
 597                             'preference': 1,
 598                         })
 599
 600         info_dict = self._parse_config(config, video_id)
 601         formats.extend(info_dict['formats'])
 602         self._vimeo_sort_formats(formats)
 603         info_dict.update({
 604             'id': video_id,
 605             'formats': formats,
 606             'upload_date': video_upload_date,
 607             'description': video_description,
 608             'webpage_url': url,
 609             'view_count': view_count,
 610             'like_count': like_count,
 611             'comment_count': comment_count,
 612         })
 613
 614         return info_dict
 615
 616
 617 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 618     IE_NAME = 'vimeo:ondemand'
 619     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 620     _TESTS = [{
 621         # ondemand video not available via https://vimeo.com/id
 622         'url': 'https://vimeo.com/ondemand/20704',
 623         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 624         'info_dict': {
 625             'id': '105442900',
 626             'ext': 'mp4',
 627             'title': 'המעבדה - במאי יותם פלדמן',
 628             'uploader': 'גם סרטים',
 629             'uploader_url': 're:https?://(?:www\.)?vimeo\.com/gumfilms',
 630             'uploader_id': 'gumfilms',
 631         },
 632     }, {
 633         # requires Referer to be passed along with og:video:url
 634         'url': 'https://vimeo.com/ondemand/36938/126682985',
 635         'info_dict': {
 636             'id': '126682985',
 637             'ext': 'mp4',
 638             'title': 'Rävlock, rätt läte på rätt plats',
 639             'uploader': 'Lindroth & Norin',
 640             'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user14430847',
 641             'uploader_id': 'user14430847',
 642         },
 643         'params': {
 644             'skip_download': True,
 645         },
 646     }, {
 647         'url': 'https://vimeo.com/ondemand/nazmaalik',
 648         'only_matching': True,
 649     }, {
 650         'url': 'https://vimeo.com/ondemand/141692381',
 651         'only_matching': True,
 652     }, {
 653         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 654         'only_matching': True,
 655     }]
 656
 657     def _real_extract(self, url):
 658         video_id = self._match_id(url)
 659         webpage = self._download_webpage(url, video_id)
 660         return self.url_result(
 661             # Some videos require Referer to be passed along with og:video:url
 662             # similarly to generic vimeo embeds (e.g.
 663             # https://vimeo.com/ondemand/36938/126682985).
 664             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 665             VimeoIE.ie_key())
 666
 667
 668 class VimeoChannelIE(VimeoBaseInfoExtractor):
 669     IE_NAME = 'vimeo:channel'
 670     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 671     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 672     _TITLE = None
 673     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 674     _TESTS = [{
 675         'url': 'https://vimeo.com/channels/tributes',
 676         'info_dict': {
 677             'id': 'tributes',
 678             'title': 'Vimeo Tributes',
 679         },
 680         'playlist_mincount': 25,
 681     }]
 682
 683     def _page_url(self, base_url, pagenum):
 684         return '%s/videos/page:%d/' % (base_url, pagenum)
 685
 686     def _extract_list_title(self, webpage):
 687         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 688
 689     def _login_list_password(self, page_url, list_id, webpage):
 690         login_form = self._search_regex(
 691             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 692             webpage, 'login form', default=None)
 693         if not login_form:
 694             return webpage
 695
 696         password = self._downloader.params.get('videopassword')
 697         if password is None:
 698             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 699         fields = self._hidden_inputs(login_form)
 700         token, vuid = self._extract_xsrft_and_vuid(webpage)
 701         fields['token'] = token
 702         fields['password'] = password
 703         post = urlencode_postdata(fields)
 704         password_path = self._search_regex(
 705             r'action="([^"]+)"', login_form, 'password URL')
 706         password_url = compat_urlparse.urljoin(page_url, password_path)
 707         password_request = sanitized_Request(password_url, post)
 708         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 709         self._set_vimeo_cookie('vuid', vuid)
 710         self._set_vimeo_cookie('xsrft', token)
 711
 712         return self._download_webpage(
 713             password_request, list_id,
 714             'Verifying the password', 'Wrong password')
 715
 716     def _title_and_entries(self, list_id, base_url):
 717         for pagenum in itertools.count(1):
 718             page_url = self._page_url(base_url, pagenum)
 719             webpage = self._download_webpage(
 720                 page_url, list_id,
 721                 'Downloading page %s' % pagenum)
 722
 723             if pagenum == 1:
 724                 webpage = self._login_list_password(page_url, list_id, webpage)
 725                 yield self._extract_list_title(webpage)
 726
 727             # Try extracting href first since not all videos are available via
 728             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 729             clips = re.findall(
 730                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)', webpage)
 731             if clips:
 732                 for video_id, video_url in clips:
 733                     yield self.url_result(
 734                         compat_urlparse.urljoin(base_url, video_url),
 735                         VimeoIE.ie_key(), video_id=video_id)
 736             # More relaxed fallback
 737             else:
 738                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 739                     yield self.url_result(
 740                         'https://vimeo.com/%s' % video_id,
 741                         VimeoIE.ie_key(), video_id=video_id)
 742
 743             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 744                 break
 745
 746     def _extract_videos(self, list_id, base_url):
 747         title_and_entries = self._title_and_entries(list_id, base_url)
 748         list_title = next(title_and_entries)
 749         return self.playlist_result(title_and_entries, list_id, list_title)
 750
 751     def _real_extract(self, url):
 752         mobj = re.match(self._VALID_URL, url)
 753         channel_id = mobj.group('id')
 754         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 755
 756
 757 class VimeoUserIE(VimeoChannelIE):
 758     IE_NAME = 'vimeo:user'
 759     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 760     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 761     _TESTS = [{
 762         'url': 'https://vimeo.com/nkistudio/videos',
 763         'info_dict': {
 764             'title': 'Nki',
 765             'id': 'nkistudio',
 766         },
 767         'playlist_mincount': 66,
 768     }]
 769
 770     def _real_extract(self, url):
 771         mobj = re.match(self._VALID_URL, url)
 772         name = mobj.group('name')
 773         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 774
 775
 776 class VimeoAlbumIE(VimeoChannelIE):
 777     IE_NAME = 'vimeo:album'
 778     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 779     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 780     _TESTS = [{
 781         'url': 'https://vimeo.com/album/2632481',
 782         'info_dict': {
 783             'id': '2632481',
 784             'title': 'Staff Favorites: November 2013',
 785         },
 786         'playlist_mincount': 13,
 787     }, {
 788         'note': 'Password-protected album',
 789         'url': 'https://vimeo.com/album/3253534',
 790         'info_dict': {
 791             'title': 'test',
 792             'id': '3253534',
 793         },
 794         'playlist_count': 1,
 795         'params': {
 796             'videopassword': 'youtube-dl',
 797         }
 798     }, {
 799         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 800         'only_matching': True,
 801     }, {
 802         # TODO: respect page number
 803         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 804         'only_matching': True,
 805     }]
 806
 807     def _page_url(self, base_url, pagenum):
 808         return '%s/page:%d/' % (base_url, pagenum)
 809
 810     def _real_extract(self, url):
 811         album_id = self._match_id(url)
 812         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 813
 814
 815 class VimeoGroupsIE(VimeoAlbumIE):
 816     IE_NAME = 'vimeo:group'
 817     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 818     _TESTS = [{
 819         'url': 'https://vimeo.com/groups/rolexawards',
 820         'info_dict': {
 821             'id': 'rolexawards',
 822             'title': 'Rolex Awards for Enterprise',
 823         },
 824         'playlist_mincount': 73,
 825     }]
 826
 827     def _extract_list_title(self, webpage):
 828         return self._og_search_title(webpage)
 829
 830     def _real_extract(self, url):
 831         mobj = re.match(self._VALID_URL, url)
 832         name = mobj.group('name')
 833         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 834
 835
 836 class VimeoReviewIE(VimeoBaseInfoExtractor):
 837     IE_NAME = 'vimeo:review'
 838     IE_DESC = 'Review pages on vimeo'
 839     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 840     _TESTS = [{
 841         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 842         'md5': 'c507a72f780cacc12b2248bb4006d253',
 843         'info_dict': {
 844             'id': '75524534',
 845             'ext': 'mp4',
 846             'title': "DICK HARDWICK 'Comedian'",
 847             'uploader': 'Richard Hardwick',
 848             'uploader_id': 'user21297594',
 849         }
 850     }, {
 851         'note': 'video player needs Referer',
 852         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 853         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 854         'info_dict': {
 855             'id': '91613211',
 856             'ext': 'mp4',
 857             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 858             'uploader': 'DevWeek Events',
 859             'duration': 2773,
 860             'thumbnail': 're:^https?://.*\.jpg$',
 861             'uploader_id': 'user22258446',
 862         }
 863     }, {
 864         'note': 'Password protected',
 865         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 866         'info_dict': {
 867             'id': '138823582',
 868             'ext': 'mp4',
 869             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 870             'uploader': 'TMB',
 871             'uploader_id': 'user37284429',
 872         },
 873         'params': {
 874             'videopassword': 'holygrail',
 875         },
 876         'skip': 'video gone',
 877     }]
 878
 879     def _real_initialize(self):
 880         self._login()
 881
 882     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 883         webpage = self._download_webpage(webpage_url, video_id)
 884         data = self._parse_json(self._search_regex(
 885             r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 886             default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 887         config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 888         if config_url is None:
 889             self._verify_video_password(webpage_url, video_id, webpage)
 890             config_url = self._get_config_url(
 891                 webpage_url, video_id, video_password_verified=True)
 892         return config_url
 893
 894     def _real_extract(self, url):
 895         video_id = self._match_id(url)
 896         config_url = self._get_config_url(url, video_id)
 897         config = self._download_json(config_url, video_id)
 898         info_dict = self._parse_config(config, video_id)
 899         self._vimeo_sort_formats(info_dict['formats'])
 900         info_dict['id'] = video_id
 901         return info_dict
 902
 903
 904 class VimeoWatchLaterIE(VimeoChannelIE):
 905     IE_NAME = 'vimeo:watchlater'
 906     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 907     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 908     _TITLE = 'Watch Later'
 909     _LOGIN_REQUIRED = True
 910     _TESTS = [{
 911         'url': 'https://vimeo.com/watchlater',
 912         'only_matching': True,
 913     }]
 914
 915     def _real_initialize(self):
 916         self._login()
 917
 918     def _page_url(self, base_url, pagenum):
 919         url = '%s/page:%d/' % (base_url, pagenum)
 920         request = sanitized_Request(url)
 921         # Set the header to get a partial html page with the ids,
 922         # the normal page doesn't contain them.
 923         request.add_header('X-Requested-With', 'XMLHttpRequest')
 924         return request
 925
 926     def _real_extract(self, url):
 927         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 928
 929
 930 class VimeoLikesIE(InfoExtractor):
 931     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 932     IE_NAME = 'vimeo:likes'
 933     IE_DESC = 'Vimeo user likes'
 934     _TEST = {
 935         'url': 'https://vimeo.com/user755559/likes/',
 936         'playlist_mincount': 293,
 937         'info_dict': {
 938             'id': 'user755559_likes',
 939             'description': 'See all the videos urza likes',
 940             'title': 'Videos urza likes',
 941         },
 942     }
 943
 944     def _real_extract(self, url):
 945         user_id = self._match_id(url)
 946         webpage = self._download_webpage(url, user_id)
 947         page_count = self._int(
 948             self._search_regex(
 949                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 950                     .*?</a></li>\s*<li\s+class="pagination_next">
 951                 ''', webpage, 'page count'),
 952             'page count', fatal=True)
 953         PAGE_SIZE = 12
 954         title = self._html_search_regex(
 955             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 956         description = self._html_search_meta('description', webpage)
 957
 958         def _get_page(idx):
 959             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 960                 user_id, idx + 1)
 961             webpage = self._download_webpage(
 962                 page_url, user_id,
 963                 note='Downloading page %d/%d' % (idx + 1, page_count))
 964             video_list = self._search_regex(
 965                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 966                 webpage, 'video content')
 967             paths = re.findall(
 968                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 969             for path in paths:
 970                 yield {
 971                     '_type': 'url',
 972                     'url': compat_urlparse.urljoin(page_url, path),
 973                 }
 974
 975         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 976
 977         return {
 978             '_type': 'playlist',
 979             'id': 'user%s_likes' % user_id,
 980             'title': title,
 981             'description': description,
 982             'entries': pl,
 983         }