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