Raphaël G. Git Repositories - youtubedl/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     NO_DEFAULT,
  20     RegexNotFoundError,
  21     sanitized_Request,
  22     smuggle_url,
  23     std_headers,
  24     try_get,
  25     unified_timestamp,
  26     unsmuggle_url,
  27     urlencode_postdata,
  28     unescapeHTML,
  29     parse_filesize,
  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                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 146                     mpd_manifest_urls = []
 147                     if re.search(mpd_pattern, manifest_url):
 148                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 149                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 150                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 151                     else:
 152                         mpd_manifest_urls = [(format_id, manifest_url)]
 153                     for f_id, m_url in mpd_manifest_urls:
 154                         mpd_formats = self._extract_mpd_formats(
 155                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 156                             'Downloading %s MPD information' % cdn_name,
 157                             fatal=False)
 158                         for f in mpd_formats:
 159                             if f.get('vcodec') == 'none':
 160                                 f['preference'] = -50
 161                             elif f.get('acodec') == 'none':
 162                                 f['preference'] = -40
 163                         formats.extend(mpd_formats)
 164
 165         subtitles = {}
 166         text_tracks = config['request'].get('text_tracks')
 167         if text_tracks:
 168             for tt in text_tracks:
 169                 subtitles[tt['lang']] = [{
 170                     'ext': 'vtt',
 171                     'url': 'https://vimeo.com' + tt['url'],
 172                 }]
 173
 174         return {
 175             'title': video_title,
 176             'uploader': video_uploader,
 177             'uploader_id': video_uploader_id,
 178             'uploader_url': video_uploader_url,
 179             'thumbnail': video_thumbnail,
 180             'duration': video_duration,
 181             'formats': formats,
 182             'subtitles': subtitles,
 183         }
 184
 185
 186 class VimeoIE(VimeoBaseInfoExtractor):
 187     """Information extractor for vimeo.com."""
 188
 189     # _VALID_URL matches Vimeo URLs
 190     _VALID_URL = r'''(?x)
 191                     https?://
 192                         (?:
 193                             (?:
 194                                 www|
 195                                 (?P<player>player)
 196                             )
 197                             \.
 198                         )?
 199                         vimeo(?P<pro>pro)?\.com/
 200                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 201                         (?:.*?/)?
 202                         (?:
 203                             (?:
 204                                 play_redirect_hls|
 205                                 moogaloop\.swf)\?clip_id=
 206                             )?
 207                         (?:videos?/)?
 208                         (?P<id>[0-9]+)
 209                         (?:/[\da-f]+)?
 210                         /?(?:[?&].*)?(?:[#].*)?$
 211                     '''
 212     IE_NAME = 'vimeo'
 213     _TESTS = [
 214         {
 215             'url': 'http://vimeo.com/56015672#at=0',
 216             'md5': '8879b6cc097e987f02484baf890129e5',
 217             'info_dict': {
 218                 'id': '56015672',
 219                 'ext': 'mp4',
 220                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 221                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 222                 'timestamp': 1355990239,
 223                 'upload_date': '20121220',
 224                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 225                 'uploader_id': 'user7108434',
 226                 'uploader': 'Filippo Valsorda',
 227                 'duration': 10,
 228                 'license': 'by-sa',
 229             },
 230         },
 231         {
 232             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 233             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 234             'note': 'Vimeo Pro video (#1197)',
 235             'info_dict': {
 236                 'id': '68093876',
 237                 'ext': 'mp4',
 238                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 239                 'uploader_id': 'openstreetmapus',
 240                 'uploader': 'OpenStreetMap US',
 241                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 242                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 243                 'duration': 1595,
 244             },
 245         },
 246         {
 247             'url': 'http://player.vimeo.com/video/54469442',
 248             'md5': '619b811a4417aa4abe78dc653becf511',
 249             'note': 'Videos that embed the url in the player page',
 250             'info_dict': {
 251                 'id': '54469442',
 252                 'ext': 'mp4',
 253                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 254                 'uploader': 'The BLN & Business of Software',
 255                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 256                 'uploader_id': 'theblnbusinessofsoftware',
 257                 'duration': 3610,
 258                 'description': None,
 259             },
 260         },
 261         {
 262             'url': 'http://vimeo.com/68375962',
 263             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 264             'note': 'Video protected with password',
 265             'info_dict': {
 266                 'id': '68375962',
 267                 'ext': 'mp4',
 268                 'title': 'youtube-dl password protected test video',
 269                 'timestamp': 1371200155,
 270                 'upload_date': '20130614',
 271                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 272                 'uploader_id': 'user18948128',
 273                 'uploader': 'Jaime Marquínez Ferrándiz',
 274                 'duration': 10,
 275                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 276             },
 277             'params': {
 278                 'videopassword': 'youtube-dl',
 279             },
 280         },
 281         {
 282             'url': 'http://vimeo.com/channels/keypeele/75629013',
 283             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 284             'info_dict': {
 285                 'id': '75629013',
 286                 'ext': 'mp4',
 287                 'title': 'Key & Peele: Terrorist Interrogation',
 288                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 289                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 290                 'uploader_id': 'atencio',
 291                 'uploader': 'Peter Atencio',
 292                 'timestamp': 1380339469,
 293                 'upload_date': '20130928',
 294                 'duration': 187,
 295             },
 296         },
 297         {
 298             'url': 'http://vimeo.com/76979871',
 299             'note': 'Video with subtitles',
 300             'info_dict': {
 301                 'id': '76979871',
 302                 'ext': 'mp4',
 303                 'title': 'The New Vimeo Player (You Know, For Videos)',
 304                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 305                 'timestamp': 1381846109,
 306                 'upload_date': '20131015',
 307                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 308                 'uploader_id': 'staff',
 309                 'uploader': 'Vimeo Staff',
 310                 'duration': 62,
 311             }
 312         },
 313         {
 314             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 315             'url': 'https://player.vimeo.com/video/98044508',
 316             'note': 'The js code contains assignments to the same variable as the config',
 317             'info_dict': {
 318                 'id': '98044508',
 319                 'ext': 'mp4',
 320                 'title': 'Pier Solar OUYA Official Trailer',
 321                 'uploader': 'Tulio Gonçalves',
 322                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 323                 'uploader_id': 'user28849593',
 324             },
 325         },
 326         {
 327             # contains original format
 328             'url': 'https://vimeo.com/33951933',
 329             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 330             'info_dict': {
 331                 'id': '33951933',
 332                 'ext': 'mp4',
 333                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 334                 'uploader': 'The DMCI',
 335                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 336                 'uploader_id': 'dmci',
 337                 'timestamp': 1324343742,
 338                 'upload_date': '20111220',
 339                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 340             },
 341         },
 342         {
 343             # only available via https://vimeo.com/channels/tributes/6213729 and
 344             # not via https://vimeo.com/6213729
 345             'url': 'https://vimeo.com/channels/tributes/6213729',
 346             'info_dict': {
 347                 'id': '6213729',
 348                 'ext': 'mov',
 349                 'title': 'Vimeo Tribute: The Shining',
 350                 'uploader': 'Casey Donahue',
 351                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 352                 'uploader_id': 'caseydonahue',
 353                 'timestamp': 1250886430,
 354                 'upload_date': '20090821',
 355                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 356             },
 357             'params': {
 358                 'skip_download': True,
 359             },
 360             'expected_warnings': ['Unable to download JSON metadata'],
 361         },
 362         {
 363             # redirects to ondemand extractor and should be passed through it
 364             # for successful extraction
 365             'url': 'https://vimeo.com/73445910',
 366             'info_dict': {
 367                 'id': '73445910',
 368                 'ext': 'mp4',
 369                 'title': 'The Reluctant Revolutionary',
 370                 'uploader': '10Ft Films',
 371                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 372                 'uploader_id': 'tenfootfilms',
 373             },
 374             'params': {
 375                 'skip_download': True,
 376             },
 377         },
 378         {
 379             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 380             'only_matching': True,
 381         },
 382         {
 383             'url': 'https://vimeo.com/109815029',
 384             'note': 'Video not completely processed, "failed" seed status',
 385             'only_matching': True,
 386         },
 387         {
 388             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 389             'only_matching': True,
 390         },
 391         {
 392             'url': 'https://vimeo.com/album/2632481/video/79010983',
 393             'only_matching': True,
 394         },
 395         {
 396             # source file returns 403: Forbidden
 397             'url': 'https://vimeo.com/7809605',
 398             'only_matching': True,
 399         },
 400         {
 401             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 402             'only_matching': True,
 403         }
 404     ]
 405
 406     @staticmethod
 407     def _smuggle_referrer(url, referrer_url):
 408         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 409
 410     @staticmethod
 411     def _extract_urls(url, webpage):
 412         urls = []
 413         # Look for embedded (iframe) Vimeo player
 414         for mobj in re.finditer(
 415                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/\d+.*?)\1',
 416                 webpage):
 417             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 418         PLAIN_EMBED_RE = (
 419             # Look for embedded (swf embed) Vimeo player
 420             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 421             # Look more for non-standard embedded Vimeo player
 422             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 423         )
 424         for embed_re in PLAIN_EMBED_RE:
 425             for mobj in re.finditer(embed_re, webpage):
 426                 urls.append(mobj.group('url'))
 427         return urls
 428
 429     @staticmethod
 430     def _extract_url(url, webpage):
 431         urls = VimeoIE._extract_urls(url, webpage)
 432         return urls[0] if urls else None
 433
 434     def _verify_player_video_password(self, url, video_id):
 435         password = self._downloader.params.get('videopassword')
 436         if password is None:
 437             raise ExtractorError('This video is protected by a password, use the --video-password option')
 438         data = urlencode_postdata({'password': password})
 439         pass_url = url + '/check-password'
 440         password_request = sanitized_Request(pass_url, data)
 441         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 442         password_request.add_header('Referer', url)
 443         return self._download_json(
 444             password_request, video_id,
 445             'Verifying the password', 'Wrong password')
 446
 447     def _real_initialize(self):
 448         self._login()
 449
 450     def _real_extract(self, url):
 451         url, data = unsmuggle_url(url, {})
 452         headers = std_headers.copy()
 453         if 'http_headers' in data:
 454             headers.update(data['http_headers'])
 455         if 'Referer' not in headers:
 456             headers['Referer'] = url
 457
 458         # Extract ID from URL
 459         mobj = re.match(self._VALID_URL, url)
 460         video_id = mobj.group('id')
 461         orig_url = url
 462         if mobj.group('pro') or mobj.group('player'):
 463             url = 'https://player.vimeo.com/video/' + video_id
 464         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 465             url = 'https://vimeo.com/' + video_id
 466
 467         # Retrieve video webpage to extract further information
 468         request = sanitized_Request(url, headers=headers)
 469         try:
 470             webpage, urlh = self._download_webpage_handle(request, video_id)
 471             redirect_url = compat_str(urlh.geturl())
 472             # Some URLs redirect to ondemand can't be extracted with
 473             # this extractor right away thus should be passed through
 474             # ondemand extractor (e.g. https://vimeo.com/73445910)
 475             if VimeoOndemandIE.suitable(redirect_url):
 476                 return self.url_result(redirect_url, VimeoOndemandIE.ie_key())
 477         except ExtractorError as ee:
 478             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 479                 errmsg = ee.cause.read()
 480                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 481                     raise ExtractorError(
 482                         'Cannot download embed-only video without embedding '
 483                         'URL. Please call youtube-dl with the URL of the page '
 484                         'that embeds this video.',
 485                         expected=True)
 486             raise
 487
 488         # Now we begin extracting as much information as we can from what we
 489         # retrieved. First we extract the information common to all extractors,
 490         # and latter we extract those that are Vimeo specific.
 491         self.report_extraction(video_id)
 492
 493         vimeo_config = self._search_regex(
 494             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 495             'vimeo config', default=None)
 496         if vimeo_config:
 497             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 498             if seed_status.get('state') == 'failed':
 499                 raise ExtractorError(
 500                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 501                     expected=True)
 502
 503         cc_license = None
 504         timestamp = None
 505
 506         # Extract the config JSON
 507         try:
 508             try:
 509                 config_url = self._html_search_regex(
 510                     r' data-config-url="(.+?)"', webpage,
 511                     'config URL', default=None)
 512                 if not config_url:
 513                     # Sometimes new react-based page is served instead of old one that require
 514                     # different config URL extraction approach (see
 515                     # https://github.com/rg3/youtube-dl/pull/7209)
 516                     vimeo_clip_page_config = self._search_regex(
 517                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 518                         'vimeo clip page config')
 519                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 520                     config_url = page_config['player']['config_url']
 521                     cc_license = page_config.get('cc_license')
 522                     timestamp = try_get(
 523                         page_config, lambda x: x['clip']['uploaded_on'],
 524                         compat_str)
 525                 config_json = self._download_webpage(config_url, video_id)
 526                 config = json.loads(config_json)
 527             except RegexNotFoundError:
 528                 # For pro videos or player.vimeo.com urls
 529                 # We try to find out to which variable is assigned the config dic
 530                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 531                 if m_variable_name is not None:
 532                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 533                 else:
 534                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 535                 config = self._search_regex(config_re, webpage, 'info section',
 536                                             flags=re.DOTALL)
 537                 config = json.loads(config)
 538         except Exception as e:
 539             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 540                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 541
 542             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 543                 if '_video_password_verified' in data:
 544                     raise ExtractorError('video password verification failed!')
 545                 self._verify_video_password(redirect_url, video_id, webpage)
 546                 return self._real_extract(
 547                     smuggle_url(redirect_url, {'_video_password_verified': 'verified'}))
 548             else:
 549                 raise ExtractorError('Unable to extract info section',
 550                                      cause=e)
 551         else:
 552             if config.get('view') == 4:
 553                 config = self._verify_player_video_password(redirect_url, video_id)
 554
 555         def is_rented():
 556             if '>You rented this title.<' in webpage:
 557                 return True
 558             if config.get('user', {}).get('purchased'):
 559                 return True
 560             label = try_get(
 561                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 562             if label and label.startswith('You rented this'):
 563                 return True
 564             return False
 565
 566         if is_rented():
 567             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 568             if feature_id and not data.get('force_feature_id', False):
 569                 return self.url_result(smuggle_url(
 570                     'https://player.vimeo.com/player/%s' % feature_id,
 571                     {'force_feature_id': True}), 'Vimeo')
 572
 573         # Extract video description
 574
 575         video_description = self._html_search_regex(
 576             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 577             webpage, 'description', default=None)
 578         if not video_description:
 579             video_description = self._html_search_meta(
 580                 'description', webpage, default=None)
 581         if not video_description and mobj.group('pro'):
 582             orig_webpage = self._download_webpage(
 583                 orig_url, video_id,
 584                 note='Downloading webpage for description',
 585                 fatal=False)
 586             if orig_webpage:
 587                 video_description = self._html_search_meta(
 588                     'description', orig_webpage, default=None)
 589         if not video_description and not mobj.group('player'):
 590             self._downloader.report_warning('Cannot find video description')
 591
 592         # Extract upload date
 593         if not timestamp:
 594             timestamp = self._search_regex(
 595                 r'<time[^>]+datetime="([^"]+)"', webpage,
 596                 'timestamp', default=None)
 597
 598         try:
 599             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 600             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 601             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 602         except RegexNotFoundError:
 603             # This info is only available in vimeo.com/{id} urls
 604             view_count = None
 605             like_count = None
 606             comment_count = None
 607
 608         formats = []
 609         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 610             'X-Requested-With': 'XMLHttpRequest'})
 611         download_data = self._download_json(download_request, video_id, fatal=False)
 612         if download_data:
 613             source_file = download_data.get('source_file')
 614             if isinstance(source_file, dict):
 615                 download_url = source_file.get('download_url')
 616                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 617                     source_name = source_file.get('public_name', 'Original')
 618                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 619                         ext = (try_get(
 620                             source_file, lambda x: x['extension'],
 621                             compat_str) or determine_ext(
 622                             download_url, None) or 'mp4').lower()
 623                         formats.append({
 624                             'url': download_url,
 625                             'ext': ext,
 626                             'width': int_or_none(source_file.get('width')),
 627                             'height': int_or_none(source_file.get('height')),
 628                             'filesize': parse_filesize(source_file.get('size')),
 629                             'format_id': source_name,
 630                             'preference': 1,
 631                         })
 632
 633         info_dict = self._parse_config(config, video_id)
 634         formats.extend(info_dict['formats'])
 635         self._vimeo_sort_formats(formats)
 636
 637         if not cc_license:
 638             cc_license = self._search_regex(
 639                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 640                 webpage, 'license', default=None, group='license')
 641
 642         info_dict.update({
 643             'id': video_id,
 644             'formats': formats,
 645             'timestamp': unified_timestamp(timestamp),
 646             'description': video_description,
 647             'webpage_url': url,
 648             'view_count': view_count,
 649             'like_count': like_count,
 650             'comment_count': comment_count,
 651             'license': cc_license,
 652         })
 653
 654         return info_dict
 655
 656
 657 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 658     IE_NAME = 'vimeo:ondemand'
 659     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 660     _TESTS = [{
 661         # ondemand video not available via https://vimeo.com/id
 662         'url': 'https://vimeo.com/ondemand/20704',
 663         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 664         'info_dict': {
 665             'id': '105442900',
 666             'ext': 'mp4',
 667             'title': 'המעבדה - במאי יותם פלדמן',
 668             'uploader': 'גם סרטים',
 669             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 670             'uploader_id': 'gumfilms',
 671         },
 672         'params': {
 673             'format': 'best[protocol=https]',
 674         },
 675     }, {
 676         # requires Referer to be passed along with og:video:url
 677         'url': 'https://vimeo.com/ondemand/36938/126682985',
 678         'info_dict': {
 679             'id': '126682985',
 680             'ext': 'mp4',
 681             'title': 'Rävlock, rätt läte på rätt plats',
 682             'uploader': 'Lindroth & Norin',
 683             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 684             'uploader_id': 'user14430847',
 685         },
 686         'params': {
 687             'skip_download': True,
 688         },
 689     }, {
 690         'url': 'https://vimeo.com/ondemand/nazmaalik',
 691         'only_matching': True,
 692     }, {
 693         'url': 'https://vimeo.com/ondemand/141692381',
 694         'only_matching': True,
 695     }, {
 696         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 697         'only_matching': True,
 698     }]
 699
 700     def _real_extract(self, url):
 701         video_id = self._match_id(url)
 702         webpage = self._download_webpage(url, video_id)
 703         return self.url_result(
 704             # Some videos require Referer to be passed along with og:video:url
 705             # similarly to generic vimeo embeds (e.g.
 706             # https://vimeo.com/ondemand/36938/126682985).
 707             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 708             VimeoIE.ie_key())
 709
 710
 711 class VimeoChannelIE(VimeoBaseInfoExtractor):
 712     IE_NAME = 'vimeo:channel'
 713     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 714     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 715     _TITLE = None
 716     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 717     _TESTS = [{
 718         'url': 'https://vimeo.com/channels/tributes',
 719         'info_dict': {
 720             'id': 'tributes',
 721             'title': 'Vimeo Tributes',
 722         },
 723         'playlist_mincount': 25,
 724     }]
 725
 726     def _page_url(self, base_url, pagenum):
 727         return '%s/videos/page:%d/' % (base_url, pagenum)
 728
 729     def _extract_list_title(self, webpage):
 730         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 731
 732     def _login_list_password(self, page_url, list_id, webpage):
 733         login_form = self._search_regex(
 734             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 735             webpage, 'login form', default=None)
 736         if not login_form:
 737             return webpage
 738
 739         password = self._downloader.params.get('videopassword')
 740         if password is None:
 741             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 742         fields = self._hidden_inputs(login_form)
 743         token, vuid = self._extract_xsrft_and_vuid(webpage)
 744         fields['token'] = token
 745         fields['password'] = password
 746         post = urlencode_postdata(fields)
 747         password_path = self._search_regex(
 748             r'action="([^"]+)"', login_form, 'password URL')
 749         password_url = compat_urlparse.urljoin(page_url, password_path)
 750         password_request = sanitized_Request(password_url, post)
 751         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 752         self._set_vimeo_cookie('vuid', vuid)
 753         self._set_vimeo_cookie('xsrft', token)
 754
 755         return self._download_webpage(
 756             password_request, list_id,
 757             'Verifying the password', 'Wrong password')
 758
 759     def _title_and_entries(self, list_id, base_url):
 760         for pagenum in itertools.count(1):
 761             page_url = self._page_url(base_url, pagenum)
 762             webpage = self._download_webpage(
 763                 page_url, list_id,
 764                 'Downloading page %s' % pagenum)
 765
 766             if pagenum == 1:
 767                 webpage = self._login_list_password(page_url, list_id, webpage)
 768                 yield self._extract_list_title(webpage)
 769
 770             # Try extracting href first since not all videos are available via
 771             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 772             clips = re.findall(
 773                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 774             if clips:
 775                 for video_id, video_url, video_title in clips:
 776                     yield self.url_result(
 777                         compat_urlparse.urljoin(base_url, video_url),
 778                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 779             # More relaxed fallback
 780             else:
 781                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 782                     yield self.url_result(
 783                         'https://vimeo.com/%s' % video_id,
 784                         VimeoIE.ie_key(), video_id=video_id)
 785
 786             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 787                 break
 788
 789     def _extract_videos(self, list_id, base_url):
 790         title_and_entries = self._title_and_entries(list_id, base_url)
 791         list_title = next(title_and_entries)
 792         return self.playlist_result(title_and_entries, list_id, list_title)
 793
 794     def _real_extract(self, url):
 795         mobj = re.match(self._VALID_URL, url)
 796         channel_id = mobj.group('id')
 797         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 798
 799
 800 class VimeoUserIE(VimeoChannelIE):
 801     IE_NAME = 'vimeo:user'
 802     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 803     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 804     _TESTS = [{
 805         'url': 'https://vimeo.com/nkistudio/videos',
 806         'info_dict': {
 807             'title': 'Nki',
 808             'id': 'nkistudio',
 809         },
 810         'playlist_mincount': 66,
 811     }]
 812
 813     def _real_extract(self, url):
 814         mobj = re.match(self._VALID_URL, url)
 815         name = mobj.group('name')
 816         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 817
 818
 819 class VimeoAlbumIE(VimeoChannelIE):
 820     IE_NAME = 'vimeo:album'
 821     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 822     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 823     _TESTS = [{
 824         'url': 'https://vimeo.com/album/2632481',
 825         'info_dict': {
 826             'id': '2632481',
 827             'title': 'Staff Favorites: November 2013',
 828         },
 829         'playlist_mincount': 13,
 830     }, {
 831         'note': 'Password-protected album',
 832         'url': 'https://vimeo.com/album/3253534',
 833         'info_dict': {
 834             'title': 'test',
 835             'id': '3253534',
 836         },
 837         'playlist_count': 1,
 838         'params': {
 839             'videopassword': 'youtube-dl',
 840         }
 841     }, {
 842         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 843         'only_matching': True,
 844     }, {
 845         # TODO: respect page number
 846         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 847         'only_matching': True,
 848     }]
 849
 850     def _page_url(self, base_url, pagenum):
 851         return '%s/page:%d/' % (base_url, pagenum)
 852
 853     def _real_extract(self, url):
 854         album_id = self._match_id(url)
 855         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 856
 857
 858 class VimeoGroupsIE(VimeoAlbumIE):
 859     IE_NAME = 'vimeo:group'
 860     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 861     _TESTS = [{
 862         'url': 'https://vimeo.com/groups/rolexawards',
 863         'info_dict': {
 864             'id': 'rolexawards',
 865             'title': 'Rolex Awards for Enterprise',
 866         },
 867         'playlist_mincount': 73,
 868     }]
 869
 870     def _extract_list_title(self, webpage):
 871         return self._og_search_title(webpage)
 872
 873     def _real_extract(self, url):
 874         mobj = re.match(self._VALID_URL, url)
 875         name = mobj.group('name')
 876         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 877
 878
 879 class VimeoReviewIE(VimeoBaseInfoExtractor):
 880     IE_NAME = 'vimeo:review'
 881     IE_DESC = 'Review pages on vimeo'
 882     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 883     _TESTS = [{
 884         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 885         'md5': 'c507a72f780cacc12b2248bb4006d253',
 886         'info_dict': {
 887             'id': '75524534',
 888             'ext': 'mp4',
 889             'title': "DICK HARDWICK 'Comedian'",
 890             'uploader': 'Richard Hardwick',
 891             'uploader_id': 'user21297594',
 892         }
 893     }, {
 894         'note': 'video player needs Referer',
 895         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 896         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 897         'info_dict': {
 898             'id': '91613211',
 899             'ext': 'mp4',
 900             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 901             'uploader': 'DevWeek Events',
 902             'duration': 2773,
 903             'thumbnail': r're:^https?://.*\.jpg$',
 904             'uploader_id': 'user22258446',
 905         }
 906     }, {
 907         'note': 'Password protected',
 908         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 909         'info_dict': {
 910             'id': '138823582',
 911             'ext': 'mp4',
 912             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 913             'uploader': 'TMB',
 914             'uploader_id': 'user37284429',
 915         },
 916         'params': {
 917             'videopassword': 'holygrail',
 918         },
 919         'skip': 'video gone',
 920     }]
 921
 922     def _real_initialize(self):
 923         self._login()
 924
 925     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 926         webpage = self._download_webpage(webpage_url, video_id)
 927         config_url = self._html_search_regex(
 928             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 929             'config URL', default=None, group='url')
 930         if not config_url:
 931             data = self._parse_json(self._search_regex(
 932                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 933                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 934             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 935         if config_url is None:
 936             self._verify_video_password(webpage_url, video_id, webpage)
 937             config_url = self._get_config_url(
 938                 webpage_url, video_id, video_password_verified=True)
 939         return config_url
 940
 941     def _real_extract(self, url):
 942         video_id = self._match_id(url)
 943         config_url = self._get_config_url(url, video_id)
 944         config = self._download_json(config_url, video_id)
 945         info_dict = self._parse_config(config, video_id)
 946         self._vimeo_sort_formats(info_dict['formats'])
 947         info_dict['id'] = video_id
 948         return info_dict
 949
 950
 951 class VimeoWatchLaterIE(VimeoChannelIE):
 952     IE_NAME = 'vimeo:watchlater'
 953     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 954     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 955     _TITLE = 'Watch Later'
 956     _LOGIN_REQUIRED = True
 957     _TESTS = [{
 958         'url': 'https://vimeo.com/watchlater',
 959         'only_matching': True,
 960     }]
 961
 962     def _real_initialize(self):
 963         self._login()
 964
 965     def _page_url(self, base_url, pagenum):
 966         url = '%s/page:%d/' % (base_url, pagenum)
 967         request = sanitized_Request(url)
 968         # Set the header to get a partial html page with the ids,
 969         # the normal page doesn't contain them.
 970         request.add_header('X-Requested-With', 'XMLHttpRequest')
 971         return request
 972
 973     def _real_extract(self, url):
 974         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 975
 976
 977 class VimeoLikesIE(InfoExtractor):
 978     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 979     IE_NAME = 'vimeo:likes'
 980     IE_DESC = 'Vimeo user likes'
 981     _TEST = {
 982         'url': 'https://vimeo.com/user755559/likes/',
 983         'playlist_mincount': 293,
 984         'info_dict': {
 985             'id': 'user755559_likes',
 986             'description': 'See all the videos urza likes',
 987             'title': 'Videos urza likes',
 988         },
 989     }
 990
 991     def _real_extract(self, url):
 992         user_id = self._match_id(url)
 993         webpage = self._download_webpage(url, user_id)
 994         page_count = self._int(
 995             self._search_regex(
 996                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 997                     .*?</a></li>\s*<li\s+class="pagination_next">
 998                 ''', webpage, 'page count'),
 999             'page count', fatal=True)
1000         PAGE_SIZE = 12
1001         title = self._html_search_regex(
1002             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
1003         description = self._html_search_meta('description', webpage)
1004
1005         def _get_page(idx):
1006             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
1007                 user_id, idx + 1)
1008             webpage = self._download_webpage(
1009                 page_url, user_id,
1010                 note='Downloading page %d/%d' % (idx + 1, page_count))
1011             video_list = self._search_regex(
1012                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
1013                 webpage, 'video content')
1014             paths = re.findall(
1015                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
1016             for path in paths:
1017                 yield {
1018                     '_type': 'url',
1019                     'url': compat_urlparse.urljoin(page_url, path),
1020                 }
1021
1022         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1023
1024         return {
1025             '_type': 'playlist',
1026             'id': 'user%s_likes' % user_id,
1027             'title': title,
1028             'description': description,
1029             'entries': pl,
1030         }