git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/vimeo.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_str,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     InAdvancePagedList,
  18     int_or_none,
  19     merge_dicts,
  20     NO_DEFAULT,
  21     RegexNotFoundError,
  22     sanitized_Request,
  23     smuggle_url,
  24     std_headers,
  25     try_get,
  26     unified_timestamp,
  27     unsmuggle_url,
  28     urlencode_postdata,
  29     unescapeHTML,
  30     parse_filesize,
  31 )
  32
  33
  34 class VimeoBaseInfoExtractor(InfoExtractor):
  35     _NETRC_MACHINE = 'vimeo'
  36     _LOGIN_REQUIRED = False
  37     _LOGIN_URL = 'https://vimeo.com/log_in'
  38
  39     def _login(self):
  40         username, password = self._get_login_info()
  41         if username is None:
  42             if self._LOGIN_REQUIRED:
  43                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  44             return
  45         webpage = self._download_webpage(
  46             self._LOGIN_URL, None, 'Downloading login page')
  47         token, vuid = self._extract_xsrft_and_vuid(webpage)
  48         data = {
  49             'action': 'login',
  50             'email': username,
  51             'password': password,
  52             'service': 'vimeo',
  53             'token': token,
  54         }
  55         self._set_vimeo_cookie('vuid', vuid)
  56         try:
  57             self._download_webpage(
  58                 self._LOGIN_URL, None, 'Logging in',
  59                 data=urlencode_postdata(data), headers={
  60                     'Content-Type': 'application/x-www-form-urlencoded',
  61                     'Referer': self._LOGIN_URL,
  62                 })
  63         except ExtractorError as e:
  64             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 418:
  65                 raise ExtractorError(
  66                     'Unable to log in: bad username or password',
  67                     expected=True)
  68             raise ExtractorError('Unable to log in')
  69
  70     def _verify_video_password(self, url, video_id, webpage):
  71         password = self._downloader.params.get('videopassword')
  72         if password is None:
  73             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  74         token, vuid = self._extract_xsrft_and_vuid(webpage)
  75         data = urlencode_postdata({
  76             'password': password,
  77             'token': token,
  78         })
  79         if url.startswith('http://'):
  80             # vimeo only supports https now, but the user can give an http url
  81             url = url.replace('http://', 'https://')
  82         password_request = sanitized_Request(url + '/password', data)
  83         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  84         password_request.add_header('Referer', url)
  85         self._set_vimeo_cookie('vuid', vuid)
  86         return self._download_webpage(
  87             password_request, video_id,
  88             'Verifying the password', 'Wrong password')
  89
  90     def _extract_xsrft_and_vuid(self, webpage):
  91         xsrft = self._search_regex(
  92             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  93             webpage, 'login token', group='xsrft')
  94         vuid = self._search_regex(
  95             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  96             webpage, 'vuid', group='vuid')
  97         return xsrft, vuid
  98
  99     def _set_vimeo_cookie(self, name, value):
 100         self._set_cookie('vimeo.com', name, value)
 101
 102     def _vimeo_sort_formats(self, formats):
 103         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
 104         # at the same time without actual units specified. This lead to wrong sorting.
 105         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
 106
 107     def _parse_config(self, config, video_id):
 108         video_data = config['video']
 109         # Extract title
 110         video_title = video_data['title']
 111
 112         # Extract uploader, uploader_url and uploader_id
 113         video_uploader = video_data.get('owner', {}).get('name')
 114         video_uploader_url = video_data.get('owner', {}).get('url')
 115         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 116
 117         # Extract video thumbnail
 118         video_thumbnail = video_data.get('thumbnail')
 119         if video_thumbnail is None:
 120             video_thumbs = video_data.get('thumbs')
 121             if video_thumbs and isinstance(video_thumbs, dict):
 122                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 123
 124         # Extract video duration
 125         video_duration = int_or_none(video_data.get('duration'))
 126
 127         formats = []
 128         config_files = video_data.get('files') or config['request'].get('files', {})
 129         for f in config_files.get('progressive', []):
 130             video_url = f.get('url')
 131             if not video_url:
 132                 continue
 133             formats.append({
 134                 'url': video_url,
 135                 'format_id': 'http-%s' % f.get('quality'),
 136                 'width': int_or_none(f.get('width')),
 137                 'height': int_or_none(f.get('height')),
 138                 'fps': int_or_none(f.get('fps')),
 139                 'tbr': int_or_none(f.get('bitrate')),
 140             })
 141
 142         for files_type in ('hls', 'dash'):
 143             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 144                 manifest_url = cdn_data.get('url')
 145                 if not manifest_url:
 146                     continue
 147                 format_id = '%s-%s' % (files_type, cdn_name)
 148                 if files_type == 'hls':
 149                     formats.extend(self._extract_m3u8_formats(
 150                         manifest_url, video_id, 'mp4',
 151                         'm3u8_native', m3u8_id=format_id,
 152                         note='Downloading %s m3u8 information' % cdn_name,
 153                         fatal=False))
 154                 elif files_type == 'dash':
 155                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 156                     mpd_manifest_urls = []
 157                     if re.search(mpd_pattern, manifest_url):
 158                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 159                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 160                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 161                     else:
 162                         mpd_manifest_urls = [(format_id, manifest_url)]
 163                     for f_id, m_url in mpd_manifest_urls:
 164                         mpd_formats = self._extract_mpd_formats(
 165                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 166                             'Downloading %s MPD information' % cdn_name,
 167                             fatal=False)
 168                         for f in mpd_formats:
 169                             if f.get('vcodec') == 'none':
 170                                 f['preference'] = -50
 171                             elif f.get('acodec') == 'none':
 172                                 f['preference'] = -40
 173                         formats.extend(mpd_formats)
 174
 175         subtitles = {}
 176         text_tracks = config['request'].get('text_tracks')
 177         if text_tracks:
 178             for tt in text_tracks:
 179                 subtitles[tt['lang']] = [{
 180                     'ext': 'vtt',
 181                     'url': 'https://vimeo.com' + tt['url'],
 182                 }]
 183
 184         return {
 185             'title': video_title,
 186             'uploader': video_uploader,
 187             'uploader_id': video_uploader_id,
 188             'uploader_url': video_uploader_url,
 189             'thumbnail': video_thumbnail,
 190             'duration': video_duration,
 191             'formats': formats,
 192             'subtitles': subtitles,
 193         }
 194
 195
 196 class VimeoIE(VimeoBaseInfoExtractor):
 197     """Information extractor for vimeo.com."""
 198
 199     # _VALID_URL matches Vimeo URLs
 200     _VALID_URL = r'''(?x)
 201                     https?://
 202                         (?:
 203                             (?:
 204                                 www|
 205                                 (?P<player>player)
 206                             )
 207                             \.
 208                         )?
 209                         vimeo(?P<pro>pro)?\.com/
 210                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 211                         (?:.*?/)?
 212                         (?:
 213                             (?:
 214                                 play_redirect_hls|
 215                                 moogaloop\.swf)\?clip_id=
 216                             )?
 217                         (?:videos?/)?
 218                         (?P<id>[0-9]+)
 219                         (?:/[\da-f]+)?
 220                         /?(?:[?&].*)?(?:[#].*)?$
 221                     '''
 222     IE_NAME = 'vimeo'
 223     _TESTS = [
 224         {
 225             'url': 'http://vimeo.com/56015672#at=0',
 226             'md5': '8879b6cc097e987f02484baf890129e5',
 227             'info_dict': {
 228                 'id': '56015672',
 229                 'ext': 'mp4',
 230                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 231                 'description': 'md5:509a9ad5c9bf97c60faee9203aca4479',
 232                 'timestamp': 1355990239,
 233                 'upload_date': '20121220',
 234                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 235                 'uploader_id': 'user7108434',
 236                 'uploader': 'Filippo Valsorda',
 237                 'duration': 10,
 238                 'license': 'by-sa',
 239             },
 240         },
 241         {
 242             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 243             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 244             'note': 'Vimeo Pro video (#1197)',
 245             'info_dict': {
 246                 'id': '68093876',
 247                 'ext': 'mp4',
 248                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 249                 'uploader_id': 'openstreetmapus',
 250                 'uploader': 'OpenStreetMap US',
 251                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 252                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 253                 'duration': 1595,
 254             },
 255         },
 256         {
 257             'url': 'http://player.vimeo.com/video/54469442',
 258             'md5': '619b811a4417aa4abe78dc653becf511',
 259             'note': 'Videos that embed the url in the player page',
 260             'info_dict': {
 261                 'id': '54469442',
 262                 'ext': 'mp4',
 263                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 264                 'uploader': 'The BLN & Business of Software',
 265                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 266                 'uploader_id': 'theblnbusinessofsoftware',
 267                 'duration': 3610,
 268                 'description': None,
 269             },
 270         },
 271         {
 272             'url': 'http://vimeo.com/68375962',
 273             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 274             'note': 'Video protected with password',
 275             'info_dict': {
 276                 'id': '68375962',
 277                 'ext': 'mp4',
 278                 'title': 'youtube-dl password protected test video',
 279                 'timestamp': 1371200155,
 280                 'upload_date': '20130614',
 281                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 282                 'uploader_id': 'user18948128',
 283                 'uploader': 'Jaime Marquínez Ferrándiz',
 284                 'duration': 10,
 285                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 286             },
 287             'params': {
 288                 'videopassword': 'youtube-dl',
 289             },
 290         },
 291         {
 292             'url': 'http://vimeo.com/channels/keypeele/75629013',
 293             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 294             'info_dict': {
 295                 'id': '75629013',
 296                 'ext': 'mp4',
 297                 'title': 'Key & Peele: Terrorist Interrogation',
 298                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 299                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 300                 'uploader_id': 'atencio',
 301                 'uploader': 'Peter Atencio',
 302                 'timestamp': 1380339469,
 303                 'upload_date': '20130928',
 304                 'duration': 187,
 305             },
 306         },
 307         {
 308             'url': 'http://vimeo.com/76979871',
 309             'note': 'Video with subtitles',
 310             'info_dict': {
 311                 'id': '76979871',
 312                 'ext': 'mp4',
 313                 'title': 'The New Vimeo Player (You Know, For Videos)',
 314                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 315                 'timestamp': 1381846109,
 316                 'upload_date': '20131015',
 317                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 318                 'uploader_id': 'staff',
 319                 'uploader': 'Vimeo Staff',
 320                 'duration': 62,
 321             }
 322         },
 323         {
 324             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 325             'url': 'https://player.vimeo.com/video/98044508',
 326             'note': 'The js code contains assignments to the same variable as the config',
 327             'info_dict': {
 328                 'id': '98044508',
 329                 'ext': 'mp4',
 330                 'title': 'Pier Solar OUYA Official Trailer',
 331                 'uploader': 'Tulio Gonçalves',
 332                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 333                 'uploader_id': 'user28849593',
 334             },
 335         },
 336         {
 337             # contains original format
 338             'url': 'https://vimeo.com/33951933',
 339             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 340             'info_dict': {
 341                 'id': '33951933',
 342                 'ext': 'mp4',
 343                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 344                 'uploader': 'The DMCI',
 345                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 346                 'uploader_id': 'dmci',
 347                 'timestamp': 1324343742,
 348                 'upload_date': '20111220',
 349                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 350             },
 351         },
 352         {
 353             # only available via https://vimeo.com/channels/tributes/6213729 and
 354             # not via https://vimeo.com/6213729
 355             'url': 'https://vimeo.com/channels/tributes/6213729',
 356             'info_dict': {
 357                 'id': '6213729',
 358                 'ext': 'mov',
 359                 'title': 'Vimeo Tribute: The Shining',
 360                 'uploader': 'Casey Donahue',
 361                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 362                 'uploader_id': 'caseydonahue',
 363                 'timestamp': 1250886430,
 364                 'upload_date': '20090821',
 365                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 366             },
 367             'params': {
 368                 'skip_download': True,
 369             },
 370             'expected_warnings': ['Unable to download JSON metadata'],
 371         },
 372         {
 373             # redirects to ondemand extractor and should be passed through it
 374             # for successful extraction
 375             'url': 'https://vimeo.com/73445910',
 376             'info_dict': {
 377                 'id': '73445910',
 378                 'ext': 'mp4',
 379                 'title': 'The Reluctant Revolutionary',
 380                 'uploader': '10Ft Films',
 381                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 382                 'uploader_id': 'tenfootfilms',
 383             },
 384             'params': {
 385                 'skip_download': True,
 386             },
 387         },
 388         {
 389             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 390             'only_matching': True,
 391         },
 392         {
 393             'url': 'https://vimeo.com/109815029',
 394             'note': 'Video not completely processed, "failed" seed status',
 395             'only_matching': True,
 396         },
 397         {
 398             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 399             'only_matching': True,
 400         },
 401         {
 402             'url': 'https://vimeo.com/album/2632481/video/79010983',
 403             'only_matching': True,
 404         },
 405         {
 406             # source file returns 403: Forbidden
 407             'url': 'https://vimeo.com/7809605',
 408             'only_matching': True,
 409         },
 410         {
 411             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 412             'only_matching': True,
 413         }
 414     ]
 415
 416     @staticmethod
 417     def _smuggle_referrer(url, referrer_url):
 418         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 419
 420     @staticmethod
 421     def _extract_urls(url, webpage):
 422         urls = []
 423         # Look for embedded (iframe) Vimeo player
 424         for mobj in re.finditer(
 425                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1',
 426                 webpage):
 427             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 428         PLAIN_EMBED_RE = (
 429             # Look for embedded (swf embed) Vimeo player
 430             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 431             # Look more for non-standard embedded Vimeo player
 432             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 433         )
 434         for embed_re in PLAIN_EMBED_RE:
 435             for mobj in re.finditer(embed_re, webpage):
 436                 urls.append(mobj.group('url'))
 437         return urls
 438
 439     @staticmethod
 440     def _extract_url(url, webpage):
 441         urls = VimeoIE._extract_urls(url, webpage)
 442         return urls[0] if urls else None
 443
 444     def _verify_player_video_password(self, url, video_id):
 445         password = self._downloader.params.get('videopassword')
 446         if password is None:
 447             raise ExtractorError('This video is protected by a password, use the --video-password option')
 448         data = urlencode_postdata({'password': password})
 449         pass_url = url + '/check-password'
 450         password_request = sanitized_Request(pass_url, data)
 451         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 452         password_request.add_header('Referer', url)
 453         return self._download_json(
 454             password_request, video_id,
 455             'Verifying the password', 'Wrong password')
 456
 457     def _real_initialize(self):
 458         self._login()
 459
 460     def _real_extract(self, url):
 461         url, data = unsmuggle_url(url, {})
 462         headers = std_headers.copy()
 463         if 'http_headers' in data:
 464             headers.update(data['http_headers'])
 465         if 'Referer' not in headers:
 466             headers['Referer'] = url
 467
 468         # Extract ID from URL
 469         mobj = re.match(self._VALID_URL, url)
 470         video_id = mobj.group('id')
 471         orig_url = url
 472         if mobj.group('pro') or mobj.group('player'):
 473             url = 'https://player.vimeo.com/video/' + video_id
 474         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 475             url = 'https://vimeo.com/' + video_id
 476
 477         # Retrieve video webpage to extract further information
 478         request = sanitized_Request(url, headers=headers)
 479         try:
 480             webpage, urlh = self._download_webpage_handle(request, video_id)
 481             redirect_url = compat_str(urlh.geturl())
 482             # Some URLs redirect to ondemand can't be extracted with
 483             # this extractor right away thus should be passed through
 484             # ondemand extractor (e.g. https://vimeo.com/73445910)
 485             if VimeoOndemandIE.suitable(redirect_url):
 486                 return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
 487         except ExtractorError as ee:
 488             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 489                 errmsg = ee.cause.read()
 490                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 491                     raise ExtractorError(
 492                         'Cannot download embed-only video without embedding '
 493                         'URL. Please call youtube-dl with the URL of the page '
 494                         'that embeds this video.',
 495                         expected=True)
 496             raise
 497
 498         # Now we begin extracting as much information as we can from what we
 499         # retrieved. First we extract the information common to all extractors,
 500         # and latter we extract those that are Vimeo specific.
 501         self.report_extraction(video_id)
 502
 503         vimeo_config = self._search_regex(
 504             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 505             'vimeo config', default=None)
 506         if vimeo_config:
 507             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 508             if seed_status.get('state') == 'failed':
 509                 raise ExtractorError(
 510                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 511                     expected=True)
 512
 513         cc_license = None
 514         timestamp = None
 515
 516         # Extract the config JSON
 517         try:
 518             try:
 519                 config_url = self._html_search_regex(
 520                     r' data-config-url="(.+?)"', webpage,
 521                     'config URL', default=None)
 522                 if not config_url:
 523                     # Sometimes new react-based page is served instead of old one that require
 524                     # different config URL extraction approach (see
 525                     # https://github.com/rg3/youtube-dl/pull/7209)
 526                     vimeo_clip_page_config = self._search_regex(
 527                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 528                         'vimeo clip page config')
 529                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 530                     config_url = page_config['player']['config_url']
 531                     cc_license = page_config.get('cc_license')
 532                     timestamp = try_get(
 533                         page_config, lambda x: x['clip']['uploaded_on'],
 534                         compat_str)
 535                 config_json = self._download_webpage(config_url, video_id)
 536                 config = json.loads(config_json)
 537             except RegexNotFoundError:
 538                 # For pro videos or player.vimeo.com urls
 539                 # We try to find out to which variable is assigned the config dic
 540                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 541                 if m_variable_name is not None:
 542                     config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))]
 543                 else:
 544                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 545                 config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;')
 546                 config = self._search_regex(config_re, webpage, 'info section',
 547                                             flags=re.DOTALL)
 548                 config = json.loads(config)
 549         except Exception as e:
 550             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 551                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 552
 553             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 554                 if '_video_password_verified' in data:
 555                     raise ExtractorError('video password verification failed!')
 556                 self._verify_video_password(redirect_url, video_id, webpage)
 557                 return self._real_extract(
 558                     smuggle_url(redirect_url, {'_video_password_verified': 'verified'}))
 559             else:
 560                 raise ExtractorError('Unable to extract info section',
 561                                      cause=e)
 562         else:
 563             if config.get('view') == 4:
 564                 config = self._verify_player_video_password(redirect_url, video_id)
 565
 566         def is_rented():
 567             if '>You rented this title.<' in webpage:
 568                 return True
 569             if config.get('user', {}).get('purchased'):
 570                 return True
 571             label = try_get(
 572                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 573             if label and label.startswith('You rented this'):
 574                 return True
 575             return False
 576
 577         if is_rented():
 578             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 579             if feature_id and not data.get('force_feature_id', False):
 580                 return self.url_result(smuggle_url(
 581                     'https://player.vimeo.com/player/%s' % feature_id,
 582                     {'force_feature_id': True}), 'Vimeo')
 583
 584         # Extract video description
 585
 586         video_description = self._html_search_regex(
 587             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 588             webpage, 'description', default=None)
 589         if not video_description:
 590             video_description = self._html_search_meta(
 591                 'description', webpage, default=None)
 592         if not video_description and mobj.group('pro'):
 593             orig_webpage = self._download_webpage(
 594                 orig_url, video_id,
 595                 note='Downloading webpage for description',
 596                 fatal=False)
 597             if orig_webpage:
 598                 video_description = self._html_search_meta(
 599                     'description', orig_webpage, default=None)
 600         if not video_description and not mobj.group('player'):
 601             self._downloader.report_warning('Cannot find video description')
 602
 603         # Extract upload date
 604         if not timestamp:
 605             timestamp = self._search_regex(
 606                 r'<time[^>]+datetime="([^"]+)"', webpage,
 607                 'timestamp', default=None)
 608
 609         try:
 610             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 611             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 612             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 613         except RegexNotFoundError:
 614             # This info is only available in vimeo.com/{id} urls
 615             view_count = None
 616             like_count = None
 617             comment_count = None
 618
 619         formats = []
 620         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 621             'X-Requested-With': 'XMLHttpRequest'})
 622         download_data = self._download_json(download_request, video_id, fatal=False)
 623         if download_data:
 624             source_file = download_data.get('source_file')
 625             if isinstance(source_file, dict):
 626                 download_url = source_file.get('download_url')
 627                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 628                     source_name = source_file.get('public_name', 'Original')
 629                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 630                         ext = (try_get(
 631                             source_file, lambda x: x['extension'],
 632                             compat_str) or determine_ext(
 633                             download_url, None) or 'mp4').lower()
 634                         formats.append({
 635                             'url': download_url,
 636                             'ext': ext,
 637                             'width': int_or_none(source_file.get('width')),
 638                             'height': int_or_none(source_file.get('height')),
 639                             'filesize': parse_filesize(source_file.get('size')),
 640                             'format_id': source_name,
 641                             'preference': 1,
 642                         })
 643
 644         info_dict_config = self._parse_config(config, video_id)
 645         formats.extend(info_dict_config['formats'])
 646         self._vimeo_sort_formats(formats)
 647
 648         json_ld = self._search_json_ld(webpage, video_id, default={})
 649
 650         if not cc_license:
 651             cc_license = self._search_regex(
 652                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 653                 webpage, 'license', default=None, group='license')
 654
 655         info_dict = {
 656             'id': video_id,
 657             'formats': formats,
 658             'timestamp': unified_timestamp(timestamp),
 659             'description': video_description,
 660             'webpage_url': url,
 661             'view_count': view_count,
 662             'like_count': like_count,
 663             'comment_count': comment_count,
 664             'license': cc_license,
 665         }
 666
 667         info_dict = merge_dicts(info_dict, info_dict_config, json_ld)
 668
 669         return info_dict
 670
 671
 672 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 673     IE_NAME = 'vimeo:ondemand'
 674     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 675     _TESTS = [{
 676         # ondemand video not available via https://vimeo.com/id
 677         'url': 'https://vimeo.com/ondemand/20704',
 678         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 679         'info_dict': {
 680             'id': '105442900',
 681             'ext': 'mp4',
 682             'title': 'המעבדה - במאי יותם פלדמן',
 683             'uploader': 'גם סרטים',
 684             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 685             'uploader_id': 'gumfilms',
 686         },
 687         'params': {
 688             'format': 'best[protocol=https]',
 689         },
 690     }, {
 691         # requires Referer to be passed along with og:video:url
 692         'url': 'https://vimeo.com/ondemand/36938/126682985',
 693         'info_dict': {
 694             'id': '126682985',
 695             'ext': 'mp4',
 696             'title': 'Rävlock, rätt läte på rätt plats',
 697             'uploader': 'Lindroth & Norin',
 698             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 699             'uploader_id': 'user14430847',
 700         },
 701         'params': {
 702             'skip_download': True,
 703         },
 704     }, {
 705         'url': 'https://vimeo.com/ondemand/nazmaalik',
 706         'only_matching': True,
 707     }, {
 708         'url': 'https://vimeo.com/ondemand/141692381',
 709         'only_matching': True,
 710     }, {
 711         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 712         'only_matching': True,
 713     }]
 714
 715     def _real_extract(self, url):
 716         video_id = self._match_id(url)
 717         webpage = self._download_webpage(url, video_id)
 718         return self.url_result(
 719             # Some videos require Referer to be passed along with og:video:url
 720             # similarly to generic vimeo embeds (e.g.
 721             # https://vimeo.com/ondemand/36938/126682985).
 722             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 723             VimeoIE.ie_key())
 724
 725
 726 class VimeoChannelIE(VimeoBaseInfoExtractor):
 727     IE_NAME = 'vimeo:channel'
 728     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 729     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 730     _TITLE = None
 731     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 732     _TESTS = [{
 733         'url': 'https://vimeo.com/channels/tributes',
 734         'info_dict': {
 735             'id': 'tributes',
 736             'title': 'Vimeo Tributes',
 737         },
 738         'playlist_mincount': 25,
 739     }]
 740
 741     def _page_url(self, base_url, pagenum):
 742         return '%s/videos/page:%d/' % (base_url, pagenum)
 743
 744     def _extract_list_title(self, webpage):
 745         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 746
 747     def _login_list_password(self, page_url, list_id, webpage):
 748         login_form = self._search_regex(
 749             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 750             webpage, 'login form', default=None)
 751         if not login_form:
 752             return webpage
 753
 754         password = self._downloader.params.get('videopassword')
 755         if password is None:
 756             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 757         fields = self._hidden_inputs(login_form)
 758         token, vuid = self._extract_xsrft_and_vuid(webpage)
 759         fields['token'] = token
 760         fields['password'] = password
 761         post = urlencode_postdata(fields)
 762         password_path = self._search_regex(
 763             r'action="([^"]+)"', login_form, 'password URL')
 764         password_url = compat_urlparse.urljoin(page_url, password_path)
 765         password_request = sanitized_Request(password_url, post)
 766         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 767         self._set_vimeo_cookie('vuid', vuid)
 768         self._set_vimeo_cookie('xsrft', token)
 769
 770         return self._download_webpage(
 771             password_request, list_id,
 772             'Verifying the password', 'Wrong password')
 773
 774     def _title_and_entries(self, list_id, base_url):
 775         for pagenum in itertools.count(1):
 776             page_url = self._page_url(base_url, pagenum)
 777             webpage = self._download_webpage(
 778                 page_url, list_id,
 779                 'Downloading page %s' % pagenum)
 780
 781             if pagenum == 1:
 782                 webpage = self._login_list_password(page_url, list_id, webpage)
 783                 yield self._extract_list_title(webpage)
 784
 785             # Try extracting href first since not all videos are available via
 786             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 787             clips = re.findall(
 788                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 789             if clips:
 790                 for video_id, video_url, video_title in clips:
 791                     yield self.url_result(
 792                         compat_urlparse.urljoin(base_url, video_url),
 793                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 794             # More relaxed fallback
 795             else:
 796                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 797                     yield self.url_result(
 798                         'https://vimeo.com/%s' % video_id,
 799                         VimeoIE.ie_key(), video_id=video_id)
 800
 801             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 802                 break
 803
 804     def _extract_videos(self, list_id, base_url):
 805         title_and_entries = self._title_and_entries(list_id, base_url)
 806         list_title = next(title_and_entries)
 807         return self.playlist_result(title_and_entries, list_id, list_title)
 808
 809     def _real_extract(self, url):
 810         mobj = re.match(self._VALID_URL, url)
 811         channel_id = mobj.group('id')
 812         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 813
 814
 815 class VimeoUserIE(VimeoChannelIE):
 816     IE_NAME = 'vimeo:user'
 817     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 818     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 819     _TESTS = [{
 820         'url': 'https://vimeo.com/nkistudio/videos',
 821         'info_dict': {
 822             'title': 'Nki',
 823             'id': 'nkistudio',
 824         },
 825         'playlist_mincount': 66,
 826     }]
 827
 828     def _real_extract(self, url):
 829         mobj = re.match(self._VALID_URL, url)
 830         name = mobj.group('name')
 831         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 832
 833
 834 class VimeoAlbumIE(VimeoChannelIE):
 835     IE_NAME = 'vimeo:album'
 836     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 837     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 838     _TESTS = [{
 839         'url': 'https://vimeo.com/album/2632481',
 840         'info_dict': {
 841             'id': '2632481',
 842             'title': 'Staff Favorites: November 2013',
 843         },
 844         'playlist_mincount': 13,
 845     }, {
 846         'note': 'Password-protected album',
 847         'url': 'https://vimeo.com/album/3253534',
 848         'info_dict': {
 849             'title': 'test',
 850             'id': '3253534',
 851         },
 852         'playlist_count': 1,
 853         'params': {
 854             'videopassword': 'youtube-dl',
 855         }
 856     }, {
 857         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 858         'only_matching': True,
 859     }, {
 860         # TODO: respect page number
 861         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 862         'only_matching': True,
 863     }]
 864
 865     def _page_url(self, base_url, pagenum):
 866         return '%s/page:%d/' % (base_url, pagenum)
 867
 868     def _real_extract(self, url):
 869         album_id = self._match_id(url)
 870         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 871
 872
 873 class VimeoGroupsIE(VimeoAlbumIE):
 874     IE_NAME = 'vimeo:group'
 875     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 876     _TESTS = [{
 877         'url': 'https://vimeo.com/groups/rolexawards',
 878         'info_dict': {
 879             'id': 'rolexawards',
 880             'title': 'Rolex Awards for Enterprise',
 881         },
 882         'playlist_mincount': 73,
 883     }]
 884
 885     def _extract_list_title(self, webpage):
 886         return self._og_search_title(webpage)
 887
 888     def _real_extract(self, url):
 889         mobj = re.match(self._VALID_URL, url)
 890         name = mobj.group('name')
 891         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 892
 893
 894 class VimeoReviewIE(VimeoBaseInfoExtractor):
 895     IE_NAME = 'vimeo:review'
 896     IE_DESC = 'Review pages on vimeo'
 897     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 898     _TESTS = [{
 899         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 900         'md5': 'c507a72f780cacc12b2248bb4006d253',
 901         'info_dict': {
 902             'id': '75524534',
 903             'ext': 'mp4',
 904             'title': "DICK HARDWICK 'Comedian'",
 905             'uploader': 'Richard Hardwick',
 906             'uploader_id': 'user21297594',
 907         }
 908     }, {
 909         'note': 'video player needs Referer',
 910         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 911         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 912         'info_dict': {
 913             'id': '91613211',
 914             'ext': 'mp4',
 915             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 916             'uploader': 'DevWeek Events',
 917             'duration': 2773,
 918             'thumbnail': r're:^https?://.*\.jpg$',
 919             'uploader_id': 'user22258446',
 920         }
 921     }, {
 922         'note': 'Password protected',
 923         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 924         'info_dict': {
 925             'id': '138823582',
 926             'ext': 'mp4',
 927             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 928             'uploader': 'TMB',
 929             'uploader_id': 'user37284429',
 930         },
 931         'params': {
 932             'videopassword': 'holygrail',
 933         },
 934         'skip': 'video gone',
 935     }]
 936
 937     def _real_initialize(self):
 938         self._login()
 939
 940     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 941         webpage = self._download_webpage(webpage_url, video_id)
 942         config_url = self._html_search_regex(
 943             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 944             'config URL', default=None, group='url')
 945         if not config_url:
 946             data = self._parse_json(self._search_regex(
 947                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 948                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 949             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 950         if config_url is None:
 951             self._verify_video_password(webpage_url, video_id, webpage)
 952             config_url = self._get_config_url(
 953                 webpage_url, video_id, video_password_verified=True)
 954         return config_url
 955
 956     def _real_extract(self, url):
 957         video_id = self._match_id(url)
 958         config_url = self._get_config_url(url, video_id)
 959         config = self._download_json(config_url, video_id)
 960         info_dict = self._parse_config(config, video_id)
 961         self._vimeo_sort_formats(info_dict['formats'])
 962         info_dict['id'] = video_id
 963         return info_dict
 964
 965
 966 class VimeoWatchLaterIE(VimeoChannelIE):
 967     IE_NAME = 'vimeo:watchlater'
 968     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 969     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 970     _TITLE = 'Watch Later'
 971     _LOGIN_REQUIRED = True
 972     _TESTS = [{
 973         'url': 'https://vimeo.com/watchlater',
 974         'only_matching': True,
 975     }]
 976
 977     def _real_initialize(self):
 978         self._login()
 979
 980     def _page_url(self, base_url, pagenum):
 981         url = '%s/page:%d/' % (base_url, pagenum)
 982         request = sanitized_Request(url)
 983         # Set the header to get a partial html page with the ids,
 984         # the normal page doesn't contain them.
 985         request.add_header('X-Requested-With', 'XMLHttpRequest')
 986         return request
 987
 988     def _real_extract(self, url):
 989         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 990
 991
 992 class VimeoLikesIE(InfoExtractor):
 993     _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)'
 994     IE_NAME = 'vimeo:likes'
 995     IE_DESC = 'Vimeo user likes'
 996     _TESTS = [{
 997         'url': 'https://vimeo.com/user755559/likes/',
 998         'playlist_mincount': 293,
 999         'info_dict': {
1000             'id': 'user755559_likes',
1001             'description': 'See all the videos urza likes',
1002             'title': 'Videos urza likes',
1003         },
1004     }, {
1005         'url': 'https://vimeo.com/stormlapse/likes',
1006         'only_matching': True,
1007     }]
1008
1009     def _real_extract(self, url):
1010         user_id = self._match_id(url)
1011         webpage = self._download_webpage(url, user_id)
1012         page_count = self._int(
1013             self._search_regex(
1014                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
1015                     .*?</a></li>\s*<li\s+class="pagination_next">
1016                 ''', webpage, 'page count', default=1),
1017             'page count', fatal=True)
1018         PAGE_SIZE = 12
1019         title = self._html_search_regex(
1020             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
1021         description = self._html_search_meta('description', webpage)
1022
1023         def _get_page(idx):
1024             page_url = 'https://vimeo.com/%s/likes/page:%d/sort:date' % (
1025                 user_id, idx + 1)
1026             webpage = self._download_webpage(
1027                 page_url, user_id,
1028                 note='Downloading page %d/%d' % (idx + 1, page_count))
1029             video_list = self._search_regex(
1030                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
1031                 webpage, 'video content')
1032             paths = re.findall(
1033                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
1034             for path in paths:
1035                 yield {
1036                     '_type': 'url',
1037                     'url': compat_urlparse.urljoin(page_url, path),
1038                 }
1039
1040         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1041
1042         return {
1043             '_type': 'playlist',
1044             'id': '%s_likes' % user_id,
1045             'title': title,
1046             'description': description,
1047             'entries': pl,
1048         }