Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vimeo.py

   1 # encoding: 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', 'format_id'))
  96
  97     def _parse_config(self, config, video_id):
  98         # Extract title
  99         video_title = config['video']['title']
 100
 101         # Extract uploader, uploader_url and uploader_id
 102         video_uploader = config['video'].get('owner', {}).get('name')
 103         video_uploader_url = config['video'].get('owner', {}).get('url')
 104         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 105
 106         # Extract video thumbnail
 107         video_thumbnail = config['video'].get('thumbnail')
 108         if video_thumbnail is None:
 109             video_thumbs = config['video'].get('thumbs')
 110             if video_thumbs and isinstance(video_thumbs, dict):
 111                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 112
 113         # Extract video duration
 114         video_duration = int_or_none(config['video'].get('duration'))
 115
 116         formats = []
 117         config_files = config['video'].get('files') or config['request'].get('files', {})
 118         for f in config_files.get('progressive', []):
 119             video_url = f.get('url')
 120             if not video_url:
 121                 continue
 122             formats.append({
 123                 'url': video_url,
 124                 'format_id': 'http-%s' % f.get('quality'),
 125                 'width': int_or_none(f.get('width')),
 126                 'height': int_or_none(f.get('height')),
 127                 'fps': int_or_none(f.get('fps')),
 128                 'tbr': int_or_none(f.get('bitrate')),
 129             })
 130         m3u8_url = config_files.get('hls', {}).get('url')
 131         if m3u8_url:
 132             formats.extend(self._extract_m3u8_formats(
 133                 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
 134
 135         subtitles = {}
 136         text_tracks = config['request'].get('text_tracks')
 137         if text_tracks:
 138             for tt in text_tracks:
 139                 subtitles[tt['lang']] = [{
 140                     'ext': 'vtt',
 141                     'url': 'https://vimeo.com' + tt['url'],
 142                 }]
 143
 144         return {
 145             'title': video_title,
 146             'uploader': video_uploader,
 147             'uploader_id': video_uploader_id,
 148             'uploader_url': video_uploader_url,
 149             'thumbnail': video_thumbnail,
 150             'duration': video_duration,
 151             'formats': formats,
 152             'subtitles': subtitles,
 153         }
 154
 155
 156 class VimeoIE(VimeoBaseInfoExtractor):
 157     """Information extractor for vimeo.com."""
 158
 159     # _VALID_URL matches Vimeo URLs
 160     _VALID_URL = r'''(?x)
 161                     https?://
 162                         (?:
 163                             (?:
 164                                 www|
 165                                 (?P<player>player)
 166                             )
 167                             \.
 168                         )?
 169                         vimeo(?P<pro>pro)?\.com/
 170                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 171                         (?:.*?/)?
 172                         (?:
 173                             (?:
 174                                 play_redirect_hls|
 175                                 moogaloop\.swf)\?clip_id=
 176                             )?
 177                         (?:videos?/)?
 178                         (?P<id>[0-9]+)
 179                         (?:/[\da-f]+)?
 180                         /?(?:[?&].*)?(?:[#].*)?$
 181                     '''
 182     IE_NAME = 'vimeo'
 183     _TESTS = [
 184         {
 185             'url': 'http://vimeo.com/56015672#at=0',
 186             'md5': '8879b6cc097e987f02484baf890129e5',
 187             'info_dict': {
 188                 'id': '56015672',
 189                 'ext': 'mp4',
 190                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 191                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 192                 'upload_date': '20121220',
 193                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user7108434',
 194                 'uploader_id': 'user7108434',
 195                 'uploader': 'Filippo Valsorda',
 196                 'duration': 10,
 197             },
 198         },
 199         {
 200             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 201             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 202             'note': 'Vimeo Pro video (#1197)',
 203             'info_dict': {
 204                 'id': '68093876',
 205                 'ext': 'mp4',
 206                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 207                 'uploader_id': 'openstreetmapus',
 208                 'uploader': 'OpenStreetMap US',
 209                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 210                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 211                 'duration': 1595,
 212             },
 213         },
 214         {
 215             'url': 'http://player.vimeo.com/video/54469442',
 216             'md5': '619b811a4417aa4abe78dc653becf511',
 217             'note': 'Videos that embed the url in the player page',
 218             'info_dict': {
 219                 'id': '54469442',
 220                 'ext': 'mp4',
 221                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 222                 'uploader': 'The BLN & Business of Software',
 223                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 224                 'uploader_id': 'theblnbusinessofsoftware',
 225                 'duration': 3610,
 226                 'description': None,
 227             },
 228         },
 229         {
 230             'url': 'http://vimeo.com/68375962',
 231             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 232             'note': 'Video protected with password',
 233             'info_dict': {
 234                 'id': '68375962',
 235                 'ext': 'mp4',
 236                 'title': 'youtube-dl password protected test video',
 237                 'upload_date': '20130614',
 238                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user18948128',
 239                 'uploader_id': 'user18948128',
 240                 'uploader': 'Jaime Marquínez Ferrándiz',
 241                 'duration': 10,
 242                 'description': 'This is "youtube-dl password protected test video" by  on Vimeo, the home for high quality videos and the people who love them.',
 243             },
 244             'params': {
 245                 'videopassword': 'youtube-dl',
 246             },
 247         },
 248         {
 249             'url': 'http://vimeo.com/channels/keypeele/75629013',
 250             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 251             'info_dict': {
 252                 'id': '75629013',
 253                 'ext': 'mp4',
 254                 'title': 'Key & Peele: Terrorist Interrogation',
 255                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 256                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/atencio',
 257                 'uploader_id': 'atencio',
 258                 'uploader': 'Peter Atencio',
 259                 'upload_date': '20130927',
 260                 'duration': 187,
 261             },
 262         },
 263         {
 264             'url': 'http://vimeo.com/76979871',
 265             'note': 'Video with subtitles',
 266             'info_dict': {
 267                 'id': '76979871',
 268                 'ext': 'mp4',
 269                 'title': 'The New Vimeo Player (You Know, For Videos)',
 270                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 271                 'upload_date': '20131015',
 272                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/staff',
 273                 'uploader_id': 'staff',
 274                 'uploader': 'Vimeo Staff',
 275                 'duration': 62,
 276             }
 277         },
 278         {
 279             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 280             'url': 'https://player.vimeo.com/video/98044508',
 281             'note': 'The js code contains assignments to the same variable as the config',
 282             'info_dict': {
 283                 'id': '98044508',
 284                 'ext': 'mp4',
 285                 'title': 'Pier Solar OUYA Official Trailer',
 286                 'uploader': 'Tulio Gonçalves',
 287                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/user28849593',
 288                 'uploader_id': 'user28849593',
 289             },
 290         },
 291         {
 292             # contains original format
 293             'url': 'https://vimeo.com/33951933',
 294             'md5': '2d9f5475e0537f013d0073e812ab89e6',
 295             'info_dict': {
 296                 'id': '33951933',
 297                 'ext': 'mp4',
 298                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 299                 'uploader': 'The DMCI',
 300                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/dmci',
 301                 'uploader_id': 'dmci',
 302                 'upload_date': '20111220',
 303                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 304             },
 305         },
 306         {
 307             # only available via https://vimeo.com/channels/tributes/6213729 and
 308             # not via https://vimeo.com/6213729
 309             'url': 'https://vimeo.com/channels/tributes/6213729',
 310             'info_dict': {
 311                 'id': '6213729',
 312                 'ext': 'mp4',
 313                 'title': 'Vimeo Tribute: The Shining',
 314                 'uploader': 'Casey Donahue',
 315                 'uploader_url': 're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 316                 'uploader_id': 'caseydonahue',
 317                 'upload_date': '20090821',
 318                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 319             },
 320             'params': {
 321                 'skip_download': True,
 322             },
 323             'expected_warnings': ['Unable to download JSON metadata'],
 324         },
 325         {
 326             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 327             'only_matching': True,
 328         },
 329         {
 330             'url': 'https://vimeo.com/109815029',
 331             'note': 'Video not completely processed, "failed" seed status',
 332             'only_matching': True,
 333         },
 334         {
 335             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 336             'only_matching': True,
 337         },
 338         {
 339             'url': 'https://vimeo.com/album/2632481/video/79010983',
 340             'only_matching': True,
 341         },
 342         {
 343             # source file returns 403: Forbidden
 344             'url': 'https://vimeo.com/7809605',
 345             'only_matching': True,
 346         },
 347         {
 348             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 349             'only_matching': True,
 350         }
 351     ]
 352
 353     @staticmethod
 354     def _extract_vimeo_url(url, webpage):
 355         # Look for embedded (iframe) Vimeo player
 356         mobj = re.search(
 357             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 358         if mobj:
 359             player_url = unescapeHTML(mobj.group('url'))
 360             surl = smuggle_url(player_url, {'http_headers': {'Referer': url}})
 361             return surl
 362         # Look for embedded (swf embed) Vimeo player
 363         mobj = re.search(
 364             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 365         if mobj:
 366             return mobj.group(1)
 367
 368     def _verify_player_video_password(self, url, video_id):
 369         password = self._downloader.params.get('videopassword')
 370         if password is None:
 371             raise ExtractorError('This video is protected by a password, use the --video-password option')
 372         data = urlencode_postdata({'password': password})
 373         pass_url = url + '/check-password'
 374         password_request = sanitized_Request(pass_url, data)
 375         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 376         password_request.add_header('Referer', url)
 377         return self._download_json(
 378             password_request, video_id,
 379             'Verifying the password', 'Wrong password')
 380
 381     def _real_initialize(self):
 382         self._login()
 383
 384     def _real_extract(self, url):
 385         url, data = unsmuggle_url(url, {})
 386         headers = std_headers.copy()
 387         if 'http_headers' in data:
 388             headers.update(data['http_headers'])
 389         if 'Referer' not in headers:
 390             headers['Referer'] = url
 391
 392         # Extract ID from URL
 393         mobj = re.match(self._VALID_URL, url)
 394         video_id = mobj.group('id')
 395         orig_url = url
 396         if mobj.group('pro') or mobj.group('player'):
 397             url = 'https://player.vimeo.com/video/' + video_id
 398         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 399             url = 'https://vimeo.com/' + video_id
 400
 401         # Retrieve video webpage to extract further information
 402         request = sanitized_Request(url, headers=headers)
 403         try:
 404             webpage = self._download_webpage(request, video_id)
 405         except ExtractorError as ee:
 406             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 407                 errmsg = ee.cause.read()
 408                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 409                     raise ExtractorError(
 410                         'Cannot download embed-only video without embedding '
 411                         'URL. Please call youtube-dl with the URL of the page '
 412                         'that embeds this video.',
 413                         expected=True)
 414             raise
 415
 416         # Now we begin extracting as much information as we can from what we
 417         # retrieved. First we extract the information common to all extractors,
 418         # and latter we extract those that are Vimeo specific.
 419         self.report_extraction(video_id)
 420
 421         vimeo_config = self._search_regex(
 422             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 423             'vimeo config', default=None)
 424         if vimeo_config:
 425             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 426             if seed_status.get('state') == 'failed':
 427                 raise ExtractorError(
 428                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 429                     expected=True)
 430
 431         # Extract the config JSON
 432         try:
 433             try:
 434                 config_url = self._html_search_regex(
 435                     r' data-config-url="(.+?)"', webpage,
 436                     'config URL', default=None)
 437                 if not config_url:
 438                     # Sometimes new react-based page is served instead of old one that require
 439                     # different config URL extraction approach (see
 440                     # https://github.com/rg3/youtube-dl/pull/7209)
 441                     vimeo_clip_page_config = self._search_regex(
 442                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 443                         'vimeo clip page config')
 444                     config_url = self._parse_json(
 445                         vimeo_clip_page_config, video_id)['player']['config_url']
 446                 config_json = self._download_webpage(config_url, video_id)
 447                 config = json.loads(config_json)
 448             except RegexNotFoundError:
 449                 # For pro videos or player.vimeo.com urls
 450                 # We try to find out to which variable is assigned the config dic
 451                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 452                 if m_variable_name is not None:
 453                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 454                 else:
 455                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 456                 config = self._search_regex(config_re, webpage, 'info section',
 457                                             flags=re.DOTALL)
 458                 config = json.loads(config)
 459         except Exception as e:
 460             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 461                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 462
 463             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 464                 if '_video_password_verified' in data:
 465                     raise ExtractorError('video password verification failed!')
 466                 self._verify_video_password(url, video_id, webpage)
 467                 return self._real_extract(
 468                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 469             else:
 470                 raise ExtractorError('Unable to extract info section',
 471                                      cause=e)
 472         else:
 473             if config.get('view') == 4:
 474                 config = self._verify_player_video_password(url, video_id)
 475
 476         def is_rented():
 477             if '>You rented this title.<' in webpage:
 478                 return True
 479             if config.get('user', {}).get('purchased'):
 480                 return True
 481             label = try_get(
 482                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 483             if label and label.startswith('You rented this'):
 484                 return True
 485             return False
 486
 487         if is_rented():
 488             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 489             if feature_id and not data.get('force_feature_id', False):
 490                 return self.url_result(smuggle_url(
 491                     'https://player.vimeo.com/player/%s' % feature_id,
 492                     {'force_feature_id': True}), 'Vimeo')
 493
 494         # Extract video description
 495
 496         video_description = self._html_search_regex(
 497             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 498             webpage, 'description', default=None)
 499         if not video_description:
 500             video_description = self._html_search_meta(
 501                 'description', webpage, default=None)
 502         if not video_description and mobj.group('pro'):
 503             orig_webpage = self._download_webpage(
 504                 orig_url, video_id,
 505                 note='Downloading webpage for description',
 506                 fatal=False)
 507             if orig_webpage:
 508                 video_description = self._html_search_meta(
 509                     'description', orig_webpage, default=None)
 510         if not video_description and not mobj.group('player'):
 511             self._downloader.report_warning('Cannot find video description')
 512
 513         # Extract upload date
 514         video_upload_date = None
 515         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 516         if mobj is not None:
 517             video_upload_date = unified_strdate(mobj.group(1))
 518
 519         try:
 520             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 521             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 522             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 523         except RegexNotFoundError:
 524             # This info is only available in vimeo.com/{id} urls
 525             view_count = None
 526             like_count = None
 527             comment_count = None
 528
 529         formats = []
 530         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 531             'X-Requested-With': 'XMLHttpRequest'})
 532         download_data = self._download_json(download_request, video_id, fatal=False)
 533         if download_data:
 534             source_file = download_data.get('source_file')
 535             if isinstance(source_file, dict):
 536                 download_url = source_file.get('download_url')
 537                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 538                     source_name = source_file.get('public_name', 'Original')
 539                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 540                         ext = source_file.get('extension', determine_ext(download_url)).lower()
 541                         formats.append({
 542                             'url': download_url,
 543                             'ext': ext,
 544                             'width': int_or_none(source_file.get('width')),
 545                             'height': int_or_none(source_file.get('height')),
 546                             'filesize': parse_filesize(source_file.get('size')),
 547                             'format_id': source_name,
 548                             'preference': 1,
 549                         })
 550
 551         info_dict = self._parse_config(config, video_id)
 552         formats.extend(info_dict['formats'])
 553         self._vimeo_sort_formats(formats)
 554         info_dict.update({
 555             'id': video_id,
 556             'formats': formats,
 557             'upload_date': video_upload_date,
 558             'description': video_description,
 559             'webpage_url': url,
 560             'view_count': view_count,
 561             'like_count': like_count,
 562             'comment_count': comment_count,
 563         })
 564
 565         return info_dict
 566
 567
 568 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 569     IE_NAME = 'vimeo:ondemand'
 570     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 571     _TESTS = [{
 572         # ondemand video not available via https://vimeo.com/id
 573         'url': 'https://vimeo.com/ondemand/20704',
 574         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 575         'info_dict': {
 576             'id': '105442900',
 577             'ext': 'mp4',
 578             'title': 'המעבדה - במאי יותם פלדמן',
 579             'uploader': 'גם סרטים',
 580             'uploader_url': 're:https?://(?:www\.)?vimeo\.com/gumfilms',
 581             'uploader_id': 'gumfilms',
 582         },
 583     }, {
 584         'url': 'https://vimeo.com/ondemand/nazmaalik',
 585         'only_matching': True,
 586     }, {
 587         'url': 'https://vimeo.com/ondemand/141692381',
 588         'only_matching': True,
 589     }, {
 590         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 591         'only_matching': True,
 592     }]
 593
 594     def _real_extract(self, url):
 595         video_id = self._match_id(url)
 596         webpage = self._download_webpage(url, video_id)
 597         return self.url_result(self._og_search_video_url(webpage), VimeoIE.ie_key())
 598
 599
 600 class VimeoChannelIE(VimeoBaseInfoExtractor):
 601     IE_NAME = 'vimeo:channel'
 602     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 603     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 604     _TITLE = None
 605     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 606     _TESTS = [{
 607         'url': 'https://vimeo.com/channels/tributes',
 608         'info_dict': {
 609             'id': 'tributes',
 610             'title': 'Vimeo Tributes',
 611         },
 612         'playlist_mincount': 25,
 613     }]
 614
 615     def _page_url(self, base_url, pagenum):
 616         return '%s/videos/page:%d/' % (base_url, pagenum)
 617
 618     def _extract_list_title(self, webpage):
 619         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 620
 621     def _login_list_password(self, page_url, list_id, webpage):
 622         login_form = self._search_regex(
 623             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 624             webpage, 'login form', default=None)
 625         if not login_form:
 626             return webpage
 627
 628         password = self._downloader.params.get('videopassword')
 629         if password is None:
 630             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 631         fields = self._hidden_inputs(login_form)
 632         token, vuid = self._extract_xsrft_and_vuid(webpage)
 633         fields['token'] = token
 634         fields['password'] = password
 635         post = urlencode_postdata(fields)
 636         password_path = self._search_regex(
 637             r'action="([^"]+)"', login_form, 'password URL')
 638         password_url = compat_urlparse.urljoin(page_url, password_path)
 639         password_request = sanitized_Request(password_url, post)
 640         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 641         self._set_vimeo_cookie('vuid', vuid)
 642         self._set_vimeo_cookie('xsrft', token)
 643
 644         return self._download_webpage(
 645             password_request, list_id,
 646             'Verifying the password', 'Wrong password')
 647
 648     def _title_and_entries(self, list_id, base_url):
 649         for pagenum in itertools.count(1):
 650             page_url = self._page_url(base_url, pagenum)
 651             webpage = self._download_webpage(
 652                 page_url, list_id,
 653                 'Downloading page %s' % pagenum)
 654
 655             if pagenum == 1:
 656                 webpage = self._login_list_password(page_url, list_id, webpage)
 657                 yield self._extract_list_title(webpage)
 658
 659             # Try extracting href first since not all videos are available via
 660             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 661             clips = re.findall(
 662                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)', webpage)
 663             if clips:
 664                 for video_id, video_url in clips:
 665                     yield self.url_result(
 666                         compat_urlparse.urljoin(base_url, video_url),
 667                         VimeoIE.ie_key(), video_id=video_id)
 668             # More relaxed fallback
 669             else:
 670                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 671                     yield self.url_result(
 672                         'https://vimeo.com/%s' % video_id,
 673                         VimeoIE.ie_key(), video_id=video_id)
 674
 675             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 676                 break
 677
 678     def _extract_videos(self, list_id, base_url):
 679         title_and_entries = self._title_and_entries(list_id, base_url)
 680         list_title = next(title_and_entries)
 681         return self.playlist_result(title_and_entries, list_id, list_title)
 682
 683     def _real_extract(self, url):
 684         mobj = re.match(self._VALID_URL, url)
 685         channel_id = mobj.group('id')
 686         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 687
 688
 689 class VimeoUserIE(VimeoChannelIE):
 690     IE_NAME = 'vimeo:user'
 691     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 692     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 693     _TESTS = [{
 694         'url': 'https://vimeo.com/nkistudio/videos',
 695         'info_dict': {
 696             'title': 'Nki',
 697             'id': 'nkistudio',
 698         },
 699         'playlist_mincount': 66,
 700     }]
 701
 702     def _real_extract(self, url):
 703         mobj = re.match(self._VALID_URL, url)
 704         name = mobj.group('name')
 705         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 706
 707
 708 class VimeoAlbumIE(VimeoChannelIE):
 709     IE_NAME = 'vimeo:album'
 710     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 711     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 712     _TESTS = [{
 713         'url': 'https://vimeo.com/album/2632481',
 714         'info_dict': {
 715             'id': '2632481',
 716             'title': 'Staff Favorites: November 2013',
 717         },
 718         'playlist_mincount': 13,
 719     }, {
 720         'note': 'Password-protected album',
 721         'url': 'https://vimeo.com/album/3253534',
 722         'info_dict': {
 723             'title': 'test',
 724             'id': '3253534',
 725         },
 726         'playlist_count': 1,
 727         'params': {
 728             'videopassword': 'youtube-dl',
 729         }
 730     }, {
 731         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 732         'only_matching': True,
 733     }, {
 734         # TODO: respect page number
 735         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 736         'only_matching': True,
 737     }]
 738
 739     def _page_url(self, base_url, pagenum):
 740         return '%s/page:%d/' % (base_url, pagenum)
 741
 742     def _real_extract(self, url):
 743         album_id = self._match_id(url)
 744         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 745
 746
 747 class VimeoGroupsIE(VimeoAlbumIE):
 748     IE_NAME = 'vimeo:group'
 749     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 750     _TESTS = [{
 751         'url': 'https://vimeo.com/groups/rolexawards',
 752         'info_dict': {
 753             'id': 'rolexawards',
 754             'title': 'Rolex Awards for Enterprise',
 755         },
 756         'playlist_mincount': 73,
 757     }]
 758
 759     def _extract_list_title(self, webpage):
 760         return self._og_search_title(webpage)
 761
 762     def _real_extract(self, url):
 763         mobj = re.match(self._VALID_URL, url)
 764         name = mobj.group('name')
 765         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 766
 767
 768 class VimeoReviewIE(VimeoBaseInfoExtractor):
 769     IE_NAME = 'vimeo:review'
 770     IE_DESC = 'Review pages on vimeo'
 771     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 772     _TESTS = [{
 773         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 774         'md5': 'c507a72f780cacc12b2248bb4006d253',
 775         'info_dict': {
 776             'id': '75524534',
 777             'ext': 'mp4',
 778             'title': "DICK HARDWICK 'Comedian'",
 779             'uploader': 'Richard Hardwick',
 780             'uploader_id': 'user21297594',
 781         }
 782     }, {
 783         'note': 'video player needs Referer',
 784         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 785         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 786         'info_dict': {
 787             'id': '91613211',
 788             'ext': 'mp4',
 789             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 790             'uploader': 'DevWeek Events',
 791             'duration': 2773,
 792             'thumbnail': 're:^https?://.*\.jpg$',
 793             'uploader_id': 'user22258446',
 794         }
 795     }, {
 796         'note': 'Password protected',
 797         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 798         'info_dict': {
 799             'id': '138823582',
 800             'ext': 'mp4',
 801             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 802             'uploader': 'TMB',
 803             'uploader_id': 'user37284429',
 804         },
 805         'params': {
 806             'videopassword': 'holygrail',
 807         },
 808     }]
 809
 810     def _real_initialize(self):
 811         self._login()
 812
 813     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 814         webpage = self._download_webpage(webpage_url, video_id)
 815         config_url = self._html_search_regex(
 816             r'data-config-url="([^"]+)"', webpage, 'config URL',
 817             default=NO_DEFAULT if video_password_verified else None)
 818         if config_url is None:
 819             self._verify_video_password(webpage_url, video_id, webpage)
 820             config_url = self._get_config_url(
 821                 webpage_url, video_id, video_password_verified=True)
 822         return config_url
 823
 824     def _real_extract(self, url):
 825         video_id = self._match_id(url)
 826         config_url = self._get_config_url(url, video_id)
 827         config = self._download_json(config_url, video_id)
 828         info_dict = self._parse_config(config, video_id)
 829         self._vimeo_sort_formats(info_dict['formats'])
 830         info_dict['id'] = video_id
 831         return info_dict
 832
 833
 834 class VimeoWatchLaterIE(VimeoChannelIE):
 835     IE_NAME = 'vimeo:watchlater'
 836     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 837     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 838     _TITLE = 'Watch Later'
 839     _LOGIN_REQUIRED = True
 840     _TESTS = [{
 841         'url': 'https://vimeo.com/watchlater',
 842         'only_matching': True,
 843     }]
 844
 845     def _real_initialize(self):
 846         self._login()
 847
 848     def _page_url(self, base_url, pagenum):
 849         url = '%s/page:%d/' % (base_url, pagenum)
 850         request = sanitized_Request(url)
 851         # Set the header to get a partial html page with the ids,
 852         # the normal page doesn't contain them.
 853         request.add_header('X-Requested-With', 'XMLHttpRequest')
 854         return request
 855
 856     def _real_extract(self, url):
 857         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 858
 859
 860 class VimeoLikesIE(InfoExtractor):
 861     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 862     IE_NAME = 'vimeo:likes'
 863     IE_DESC = 'Vimeo user likes'
 864     _TEST = {
 865         'url': 'https://vimeo.com/user755559/likes/',
 866         'playlist_mincount': 293,
 867         'info_dict': {
 868             'id': 'user755559_likes',
 869             'description': 'See all the videos urza likes',
 870             'title': 'Videos urza likes',
 871         },
 872     }
 873
 874     def _real_extract(self, url):
 875         user_id = self._match_id(url)
 876         webpage = self._download_webpage(url, user_id)
 877         page_count = self._int(
 878             self._search_regex(
 879                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 880                     .*?</a></li>\s*<li\s+class="pagination_next">
 881                 ''', webpage, 'page count'),
 882             'page count', fatal=True)
 883         PAGE_SIZE = 12
 884         title = self._html_search_regex(
 885             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 886         description = self._html_search_meta('description', webpage)
 887
 888         def _get_page(idx):
 889             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 890                 user_id, idx + 1)
 891             webpage = self._download_webpage(
 892                 page_url, user_id,
 893                 note='Downloading page %d/%d' % (idx + 1, page_count))
 894             video_list = self._search_regex(
 895                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 896                 webpage, 'video content')
 897             paths = re.findall(
 898                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 899             for path in paths:
 900                 yield {
 901                     '_type': 'url',
 902                     'url': compat_urlparse.urljoin(page_url, path),
 903                 }
 904
 905         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 906
 907         return {
 908             '_type': 'playlist',
 909             'id': 'user%s_likes' % user_id,
 910             'title': title,
 911             'description': description,
 912             'entries': pl,
 913         }