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_strdate,
  26     unified_timestamp,
  27     unsmuggle_url,
  28     urlencode_postdata,
  29     unescapeHTML,
  30     parse_filesize,
  31     try_get,
  32 )
  33
  34
  35 class VimeoBaseInfoExtractor(InfoExtractor):
  36     _NETRC_MACHINE = 'vimeo'
  37     _LOGIN_REQUIRED = False
  38     _LOGIN_URL = 'https://vimeo.com/log_in'
  39
  40     def _login(self):
  41         (username, password) = self._get_login_info()
  42         if username is None:
  43             if self._LOGIN_REQUIRED:
  44                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  45             return
  46         self.report_login()
  47         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  48         token, vuid = self._extract_xsrft_and_vuid(webpage)
  49         data = urlencode_postdata({
  50             'action': 'login',
  51             'email': username,
  52             'password': password,
  53             'service': 'vimeo',
  54             'token': token,
  55         })
  56         login_request = sanitized_Request(self._LOGIN_URL, data)
  57         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  58         login_request.add_header('Referer', self._LOGIN_URL)
  59         self._set_vimeo_cookie('vuid', vuid)
  60         self._download_webpage(login_request, None, False, 'Wrong login info')
  61
  62     def _verify_video_password(self, url, video_id, webpage):
  63         password = self._downloader.params.get('videopassword')
  64         if password is None:
  65             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
  66         token, vuid = self._extract_xsrft_and_vuid(webpage)
  67         data = urlencode_postdata({
  68             'password': password,
  69             'token': token,
  70         })
  71         if url.startswith('http://'):
  72             # vimeo only supports https now, but the user can give an http url
  73             url = url.replace('http://', 'https://')
  74         password_request = sanitized_Request(url + '/password', data)
  75         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  76         password_request.add_header('Referer', url)
  77         self._set_vimeo_cookie('vuid', vuid)
  78         return self._download_webpage(
  79             password_request, video_id,
  80             'Verifying the password', 'Wrong password')
  81
  82     def _extract_xsrft_and_vuid(self, webpage):
  83         xsrft = self._search_regex(
  84             r'(?:(?P<q1>["\'])xsrft(?P=q1)\s*:|xsrft\s*[=:])\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  85             webpage, 'login token', group='xsrft')
  86         vuid = self._search_regex(
  87             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  88             webpage, 'vuid', group='vuid')
  89         return xsrft, vuid
  90
  91     def _set_vimeo_cookie(self, name, value):
  92         self._set_cookie('vimeo.com', name, value)
  93
  94     def _vimeo_sort_formats(self, formats):
  95         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
  96         # at the same time without actual units specified. This lead to wrong sorting.
  97         self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
  98
  99     def _parse_config(self, config, video_id):
 100         video_data = config['video']
 101         # Extract title
 102         video_title = video_data['title']
 103
 104         # Extract uploader, uploader_url and uploader_id
 105         video_uploader = video_data.get('owner', {}).get('name')
 106         video_uploader_url = video_data.get('owner', {}).get('url')
 107         video_uploader_id = video_uploader_url.split('/')[-1] if video_uploader_url else None
 108
 109         # Extract video thumbnail
 110         video_thumbnail = video_data.get('thumbnail')
 111         if video_thumbnail is None:
 112             video_thumbs = video_data.get('thumbs')
 113             if video_thumbs and isinstance(video_thumbs, dict):
 114                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 115
 116         # Extract video duration
 117         video_duration = int_or_none(video_data.get('duration'))
 118
 119         formats = []
 120         config_files = video_data.get('files') or config['request'].get('files', {})
 121         for f in config_files.get('progressive', []):
 122             video_url = f.get('url')
 123             if not video_url:
 124                 continue
 125             formats.append({
 126                 'url': video_url,
 127                 'format_id': 'http-%s' % f.get('quality'),
 128                 'width': int_or_none(f.get('width')),
 129                 'height': int_or_none(f.get('height')),
 130                 'fps': int_or_none(f.get('fps')),
 131                 'tbr': int_or_none(f.get('bitrate')),
 132             })
 133
 134         for files_type in ('hls', 'dash'):
 135             for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items():
 136                 manifest_url = cdn_data.get('url')
 137                 if not manifest_url:
 138                     continue
 139                 format_id = '%s-%s' % (files_type, cdn_name)
 140                 if files_type == 'hls':
 141                     formats.extend(self._extract_m3u8_formats(
 142                         manifest_url, video_id, 'mp4',
 143                         'm3u8_native', m3u8_id=format_id,
 144                         note='Downloading %s m3u8 information' % cdn_name,
 145                         fatal=False))
 146                 elif files_type == 'dash':
 147                     mpd_pattern = r'/%s/(?:sep/)?video/' % video_id
 148                     mpd_manifest_urls = []
 149                     if re.search(mpd_pattern, manifest_url):
 150                         for suffix, repl in (('', 'video'), ('_sep', 'sep/video')):
 151                             mpd_manifest_urls.append((format_id + suffix, re.sub(
 152                                 mpd_pattern, '/%s/%s/' % (video_id, repl), manifest_url)))
 153                     else:
 154                         mpd_manifest_urls = [(format_id, manifest_url)]
 155                     for f_id, m_url in mpd_manifest_urls:
 156                         formats.extend(self._extract_mpd_formats(
 157                             m_url.replace('/master.json', '/master.mpd'), video_id, f_id,
 158                             'Downloading %s MPD information' % cdn_name,
 159                             fatal=False))
 160
 161         subtitles = {}
 162         text_tracks = config['request'].get('text_tracks')
 163         if text_tracks:
 164             for tt in text_tracks:
 165                 subtitles[tt['lang']] = [{
 166                     'ext': 'vtt',
 167                     'url': 'https://vimeo.com' + tt['url'],
 168                 }]
 169
 170         return {
 171             'title': video_title,
 172             'uploader': video_uploader,
 173             'uploader_id': video_uploader_id,
 174             'uploader_url': video_uploader_url,
 175             'thumbnail': video_thumbnail,
 176             'duration': video_duration,
 177             'formats': formats,
 178             'subtitles': subtitles,
 179         }
 180
 181
 182 class VimeoIE(VimeoBaseInfoExtractor):
 183     """Information extractor for vimeo.com."""
 184
 185     # _VALID_URL matches Vimeo URLs
 186     _VALID_URL = r'''(?x)
 187                     https?://
 188                         (?:
 189                             (?:
 190                                 www|
 191                                 (?P<player>player)
 192                             )
 193                             \.
 194                         )?
 195                         vimeo(?P<pro>pro)?\.com/
 196                         (?!(?:channels|album)/[^/?#]+/?(?:$|[?#])|[^/]+/review/|ondemand/)
 197                         (?:.*?/)?
 198                         (?:
 199                             (?:
 200                                 play_redirect_hls|
 201                                 moogaloop\.swf)\?clip_id=
 202                             )?
 203                         (?:videos?/)?
 204                         (?P<id>[0-9]+)
 205                         (?:/[\da-f]+)?
 206                         /?(?:[?&].*)?(?:[#].*)?$
 207                     '''
 208     IE_NAME = 'vimeo'
 209     _TESTS = [
 210         {
 211             'url': 'http://vimeo.com/56015672#at=0',
 212             'md5': '8879b6cc097e987f02484baf890129e5',
 213             'info_dict': {
 214                 'id': '56015672',
 215                 'ext': 'mp4',
 216                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
 217                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
 218                 'timestamp': 1355990239,
 219                 'upload_date': '20121220',
 220                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user7108434',
 221                 'uploader_id': 'user7108434',
 222                 'uploader': 'Filippo Valsorda',
 223                 'duration': 10,
 224                 'license': 'by-sa',
 225             },
 226         },
 227         {
 228             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
 229             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
 230             'note': 'Vimeo Pro video (#1197)',
 231             'info_dict': {
 232                 'id': '68093876',
 233                 'ext': 'mp4',
 234                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/openstreetmapus',
 235                 'uploader_id': 'openstreetmapus',
 236                 'uploader': 'OpenStreetMap US',
 237                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 238                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 239                 'duration': 1595,
 240             },
 241         },
 242         {
 243             'url': 'http://player.vimeo.com/video/54469442',
 244             'md5': '619b811a4417aa4abe78dc653becf511',
 245             'note': 'Videos that embed the url in the player page',
 246             'info_dict': {
 247                 'id': '54469442',
 248                 'ext': 'mp4',
 249                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 250                 'uploader': 'The BLN & Business of Software',
 251                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware',
 252                 'uploader_id': 'theblnbusinessofsoftware',
 253                 'duration': 3610,
 254                 'description': None,
 255             },
 256         },
 257         {
 258             'url': 'http://vimeo.com/68375962',
 259             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 260             'note': 'Video protected with password',
 261             'info_dict': {
 262                 'id': '68375962',
 263                 'ext': 'mp4',
 264                 'title': 'youtube-dl password protected test video',
 265                 'timestamp': 1371200155,
 266                 'upload_date': '20130614',
 267                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user18948128',
 268                 'uploader_id': 'user18948128',
 269                 'uploader': 'Jaime Marquínez Ferrándiz',
 270                 'duration': 10,
 271                 'description': 'md5:dca3ea23adb29ee387127bc4ddfce63f',
 272             },
 273             'params': {
 274                 'videopassword': 'youtube-dl',
 275             },
 276         },
 277         {
 278             'url': 'http://vimeo.com/channels/keypeele/75629013',
 279             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 280             'info_dict': {
 281                 'id': '75629013',
 282                 'ext': 'mp4',
 283                 'title': 'Key & Peele: Terrorist Interrogation',
 284                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 285                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/atencio',
 286                 'uploader_id': 'atencio',
 287                 'uploader': 'Peter Atencio',
 288                 'timestamp': 1380339469,
 289                 'upload_date': '20130928',
 290                 'duration': 187,
 291             },
 292         },
 293         {
 294             'url': 'http://vimeo.com/76979871',
 295             'note': 'Video with subtitles',
 296             'info_dict': {
 297                 'id': '76979871',
 298                 'ext': 'mp4',
 299                 'title': 'The New Vimeo Player (You Know, For Videos)',
 300                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 301                 'timestamp': 1381846109,
 302                 'upload_date': '20131015',
 303                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/staff',
 304                 'uploader_id': 'staff',
 305                 'uploader': 'Vimeo Staff',
 306                 'duration': 62,
 307             }
 308         },
 309         {
 310             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 311             'url': 'https://player.vimeo.com/video/98044508',
 312             'note': 'The js code contains assignments to the same variable as the config',
 313             'info_dict': {
 314                 'id': '98044508',
 315                 'ext': 'mp4',
 316                 'title': 'Pier Solar OUYA Official Trailer',
 317                 'uploader': 'Tulio Gonçalves',
 318                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user28849593',
 319                 'uploader_id': 'user28849593',
 320             },
 321         },
 322         {
 323             # contains original format
 324             'url': 'https://vimeo.com/33951933',
 325             'md5': '53c688fa95a55bf4b7293d37a89c5c53',
 326             'info_dict': {
 327                 'id': '33951933',
 328                 'ext': 'mp4',
 329                 'title': 'FOX CLASSICS - Forever Classic ID - A Full Minute',
 330                 'uploader': 'The DMCI',
 331                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/dmci',
 332                 'uploader_id': 'dmci',
 333                 'timestamp': 1324343742,
 334                 'upload_date': '20111220',
 335                 'description': 'md5:ae23671e82d05415868f7ad1aec21147',
 336             },
 337         },
 338         {
 339             # only available via https://vimeo.com/channels/tributes/6213729 and
 340             # not via https://vimeo.com/6213729
 341             'url': 'https://vimeo.com/channels/tributes/6213729',
 342             'info_dict': {
 343                 'id': '6213729',
 344                 'ext': 'mov',
 345                 'title': 'Vimeo Tribute: The Shining',
 346                 'uploader': 'Casey Donahue',
 347                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/caseydonahue',
 348                 'uploader_id': 'caseydonahue',
 349                 'timestamp': 1250886430,
 350                 'upload_date': '20090821',
 351                 'description': 'md5:bdbf314014e58713e6e5b66eb252f4a6',
 352             },
 353             'params': {
 354                 'skip_download': True,
 355             },
 356             'expected_warnings': ['Unable to download JSON metadata'],
 357         },
 358         {
 359             # redirects to ondemand extractor and should be passed through it
 360             # for successful extraction
 361             'url': 'https://vimeo.com/73445910',
 362             'info_dict': {
 363                 'id': '73445910',
 364                 'ext': 'mp4',
 365                 'title': 'The Reluctant Revolutionary',
 366                 'uploader': '10Ft Films',
 367                 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/tenfootfilms',
 368                 'uploader_id': 'tenfootfilms',
 369             },
 370             'params': {
 371                 'skip_download': True,
 372             },
 373         },
 374         {
 375             'url': 'http://vimeo.com/moogaloop.swf?clip_id=2539741',
 376             'only_matching': True,
 377         },
 378         {
 379             'url': 'https://vimeo.com/109815029',
 380             'note': 'Video not completely processed, "failed" seed status',
 381             'only_matching': True,
 382         },
 383         {
 384             'url': 'https://vimeo.com/groups/travelhd/videos/22439234',
 385             'only_matching': True,
 386         },
 387         {
 388             'url': 'https://vimeo.com/album/2632481/video/79010983',
 389             'only_matching': True,
 390         },
 391         {
 392             # source file returns 403: Forbidden
 393             'url': 'https://vimeo.com/7809605',
 394             'only_matching': True,
 395         },
 396         {
 397             'url': 'https://vimeo.com/160743502/abd0e13fb4',
 398             'only_matching': True,
 399         }
 400     ]
 401
 402     @staticmethod
 403     def _smuggle_referrer(url, referrer_url):
 404         return smuggle_url(url, {'http_headers': {'Referer': referrer_url}})
 405
 406     @staticmethod
 407     def _extract_urls(url, webpage):
 408         urls = []
 409         # Look for embedded (iframe) Vimeo player
 410         for mobj in re.finditer(
 411                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1',
 412                 webpage):
 413             urls.append(VimeoIE._smuggle_referrer(unescapeHTML(mobj.group('url')), url))
 414         PLAIN_EMBED_RE = (
 415             # Look for embedded (swf embed) Vimeo player
 416             r'<embed[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)\1',
 417             # Look more for non-standard embedded Vimeo player
 418             r'<video[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)\1',
 419         )
 420         for embed_re in PLAIN_EMBED_RE:
 421             for mobj in re.finditer(embed_re, webpage):
 422                 urls.append(mobj.group('url'))
 423         return urls
 424
 425     @staticmethod
 426     def _extract_url(url, webpage):
 427         urls = VimeoIE._extract_urls(url, webpage)
 428         return urls[0] if urls else None
 429
 430     def _verify_player_video_password(self, url, video_id):
 431         password = self._downloader.params.get('videopassword')
 432         if password is None:
 433             raise ExtractorError('This video is protected by a password, use the --video-password option')
 434         data = urlencode_postdata({'password': password})
 435         pass_url = url + '/check-password'
 436         password_request = sanitized_Request(pass_url, data)
 437         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 438         password_request.add_header('Referer', url)
 439         return self._download_json(
 440             password_request, video_id,
 441             'Verifying the password', 'Wrong password')
 442
 443     def _real_initialize(self):
 444         self._login()
 445
 446     def _real_extract(self, url):
 447         url, data = unsmuggle_url(url, {})
 448         headers = std_headers.copy()
 449         if 'http_headers' in data:
 450             headers.update(data['http_headers'])
 451         if 'Referer' not in headers:
 452             headers['Referer'] = url
 453
 454         # Extract ID from URL
 455         mobj = re.match(self._VALID_URL, url)
 456         video_id = mobj.group('id')
 457         orig_url = url
 458         if mobj.group('pro') or mobj.group('player'):
 459             url = 'https://player.vimeo.com/video/' + video_id
 460         elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')):
 461             url = 'https://vimeo.com/' + video_id
 462
 463         # Retrieve video webpage to extract further information
 464         request = sanitized_Request(url, headers=headers)
 465         try:
 466             webpage, urlh = self._download_webpage_handle(request, video_id)
 467             # Some URLs redirect to ondemand can't be extracted with
 468             # this extractor right away thus should be passed through
 469             # ondemand extractor (e.g. https://vimeo.com/73445910)
 470             if VimeoOndemandIE.suitable(urlh.geturl()):
 471                 return self.url_result(urlh.geturl(), VimeoOndemandIE.ie_key())
 472         except ExtractorError as ee:
 473             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 474                 errmsg = ee.cause.read()
 475                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 476                     raise ExtractorError(
 477                         'Cannot download embed-only video without embedding '
 478                         'URL. Please call youtube-dl with the URL of the page '
 479                         'that embeds this video.',
 480                         expected=True)
 481             raise
 482
 483         # Now we begin extracting as much information as we can from what we
 484         # retrieved. First we extract the information common to all extractors,
 485         # and latter we extract those that are Vimeo specific.
 486         self.report_extraction(video_id)
 487
 488         vimeo_config = self._search_regex(
 489             r'vimeo\.config\s*=\s*(?:({.+?})|_extend\([^,]+,\s+({.+?})\));', webpage,
 490             'vimeo config', default=None)
 491         if vimeo_config:
 492             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 493             if seed_status.get('state') == 'failed':
 494                 raise ExtractorError(
 495                     '%s said: %s' % (self.IE_NAME, seed_status['title']),
 496                     expected=True)
 497
 498         cc_license = None
 499         timestamp = None
 500
 501         # Extract the config JSON
 502         try:
 503             try:
 504                 config_url = self._html_search_regex(
 505                     r' data-config-url="(.+?)"', webpage,
 506                     'config URL', default=None)
 507                 if not config_url:
 508                     # Sometimes new react-based page is served instead of old one that require
 509                     # different config URL extraction approach (see
 510                     # https://github.com/rg3/youtube-dl/pull/7209)
 511                     vimeo_clip_page_config = self._search_regex(
 512                         r'vimeo\.clip_page_config\s*=\s*({.+?});', webpage,
 513                         'vimeo clip page config')
 514                     page_config = self._parse_json(vimeo_clip_page_config, video_id)
 515                     config_url = page_config['player']['config_url']
 516                     cc_license = page_config.get('cc_license')
 517                     timestamp = try_get(
 518                         page_config, lambda x: x['clip']['uploaded_on'],
 519                         compat_str)
 520                 config_json = self._download_webpage(config_url, video_id)
 521                 config = json.loads(config_json)
 522             except RegexNotFoundError:
 523                 # For pro videos or player.vimeo.com urls
 524                 # We try to find out to which variable is assigned the config dic
 525                 m_variable_name = re.search(r'(\w)\.video\.id', webpage)
 526                 if m_variable_name is not None:
 527                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 528                 else:
 529                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 530                 config = self._search_regex(config_re, webpage, 'info section',
 531                                             flags=re.DOTALL)
 532                 config = json.loads(config)
 533         except Exception as e:
 534             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 535                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 536
 537             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 538                 if '_video_password_verified' in data:
 539                     raise ExtractorError('video password verification failed!')
 540                 self._verify_video_password(url, video_id, webpage)
 541                 return self._real_extract(
 542                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 543             else:
 544                 raise ExtractorError('Unable to extract info section',
 545                                      cause=e)
 546         else:
 547             if config.get('view') == 4:
 548                 config = self._verify_player_video_password(url, video_id)
 549
 550         def is_rented():
 551             if '>You rented this title.<' in webpage:
 552                 return True
 553             if config.get('user', {}).get('purchased'):
 554                 return True
 555             label = try_get(
 556                 config, lambda x: x['video']['vod']['purchase_options'][0]['label_string'], compat_str)
 557             if label and label.startswith('You rented this'):
 558                 return True
 559             return False
 560
 561         if is_rented():
 562             feature_id = config.get('video', {}).get('vod', {}).get('feature_id')
 563             if feature_id and not data.get('force_feature_id', False):
 564                 return self.url_result(smuggle_url(
 565                     'https://player.vimeo.com/player/%s' % feature_id,
 566                     {'force_feature_id': True}), 'Vimeo')
 567
 568         # Extract video description
 569
 570         video_description = self._html_search_regex(
 571             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 572             webpage, 'description', default=None)
 573         if not video_description:
 574             video_description = self._html_search_meta(
 575                 'description', webpage, default=None)
 576         if not video_description and mobj.group('pro'):
 577             orig_webpage = self._download_webpage(
 578                 orig_url, video_id,
 579                 note='Downloading webpage for description',
 580                 fatal=False)
 581             if orig_webpage:
 582                 video_description = self._html_search_meta(
 583                     'description', orig_webpage, default=None)
 584         if not video_description and not mobj.group('player'):
 585             self._downloader.report_warning('Cannot find video description')
 586
 587         # Extract upload date
 588         if not timestamp:
 589             timestamp = self._search_regex(
 590                 r'<time[^>]+datetime="([^"]+)"', webpage,
 591                 'timestamp', default=None)
 592
 593         try:
 594             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 595             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 596             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 597         except RegexNotFoundError:
 598             # This info is only available in vimeo.com/{id} urls
 599             view_count = None
 600             like_count = None
 601             comment_count = None
 602
 603         formats = []
 604         download_request = sanitized_Request('https://vimeo.com/%s?action=load_download_config' % video_id, headers={
 605             'X-Requested-With': 'XMLHttpRequest'})
 606         download_data = self._download_json(download_request, video_id, fatal=False)
 607         if download_data:
 608             source_file = download_data.get('source_file')
 609             if isinstance(source_file, dict):
 610                 download_url = source_file.get('download_url')
 611                 if download_url and not source_file.get('is_cold') and not source_file.get('is_defrosting'):
 612                     source_name = source_file.get('public_name', 'Original')
 613                     if self._is_valid_url(download_url, video_id, '%s video' % source_name):
 614                         ext = source_file.get('extension', determine_ext(download_url)).lower()
 615                         formats.append({
 616                             'url': download_url,
 617                             'ext': ext,
 618                             'width': int_or_none(source_file.get('width')),
 619                             'height': int_or_none(source_file.get('height')),
 620                             'filesize': parse_filesize(source_file.get('size')),
 621                             'format_id': source_name,
 622                             'preference': 1,
 623                         })
 624
 625         info_dict = self._parse_config(config, video_id)
 626         formats.extend(info_dict['formats'])
 627         self._vimeo_sort_formats(formats)
 628
 629         if not cc_license:
 630             cc_license = self._search_regex(
 631                 r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1',
 632                 webpage, 'license', default=None, group='license')
 633
 634         info_dict.update({
 635             'id': video_id,
 636             'formats': formats,
 637             'timestamp': unified_timestamp(timestamp),
 638             'description': video_description,
 639             'webpage_url': url,
 640             'view_count': view_count,
 641             'like_count': like_count,
 642             'comment_count': comment_count,
 643             'license': cc_license,
 644         })
 645
 646         return info_dict
 647
 648
 649 class VimeoOndemandIE(VimeoBaseInfoExtractor):
 650     IE_NAME = 'vimeo:ondemand'
 651     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?P<id>[^/?#&]+)'
 652     _TESTS = [{
 653         # ondemand video not available via https://vimeo.com/id
 654         'url': 'https://vimeo.com/ondemand/20704',
 655         'md5': 'c424deda8c7f73c1dfb3edd7630e2f35',
 656         'info_dict': {
 657             'id': '105442900',
 658             'ext': 'mp4',
 659             'title': 'המעבדה - במאי יותם פלדמן',
 660             'uploader': 'גם סרטים',
 661             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/gumfilms',
 662             'uploader_id': 'gumfilms',
 663         },
 664         'params': {
 665             'format': 'best[protocol=https]',
 666         },
 667     }, {
 668         # requires Referer to be passed along with og:video:url
 669         'url': 'https://vimeo.com/ondemand/36938/126682985',
 670         'info_dict': {
 671             'id': '126682985',
 672             'ext': 'mp4',
 673             'title': 'Rävlock, rätt läte på rätt plats',
 674             'uploader': 'Lindroth & Norin',
 675             'uploader_url': r're:https?://(?:www\.)?vimeo\.com/user14430847',
 676             'uploader_id': 'user14430847',
 677         },
 678         'params': {
 679             'skip_download': True,
 680         },
 681     }, {
 682         'url': 'https://vimeo.com/ondemand/nazmaalik',
 683         'only_matching': True,
 684     }, {
 685         'url': 'https://vimeo.com/ondemand/141692381',
 686         'only_matching': True,
 687     }, {
 688         'url': 'https://vimeo.com/ondemand/thelastcolony/150274832',
 689         'only_matching': True,
 690     }]
 691
 692     def _real_extract(self, url):
 693         video_id = self._match_id(url)
 694         webpage = self._download_webpage(url, video_id)
 695         return self.url_result(
 696             # Some videos require Referer to be passed along with og:video:url
 697             # similarly to generic vimeo embeds (e.g.
 698             # https://vimeo.com/ondemand/36938/126682985).
 699             VimeoIE._smuggle_referrer(self._og_search_video_url(webpage), url),
 700             VimeoIE.ie_key())
 701
 702
 703 class VimeoChannelIE(VimeoBaseInfoExtractor):
 704     IE_NAME = 'vimeo:channel'
 705     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 706     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 707     _TITLE = None
 708     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 709     _TESTS = [{
 710         'url': 'https://vimeo.com/channels/tributes',
 711         'info_dict': {
 712             'id': 'tributes',
 713             'title': 'Vimeo Tributes',
 714         },
 715         'playlist_mincount': 25,
 716     }]
 717
 718     def _page_url(self, base_url, pagenum):
 719         return '%s/videos/page:%d/' % (base_url, pagenum)
 720
 721     def _extract_list_title(self, webpage):
 722         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 723
 724     def _login_list_password(self, page_url, list_id, webpage):
 725         login_form = self._search_regex(
 726             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 727             webpage, 'login form', default=None)
 728         if not login_form:
 729             return webpage
 730
 731         password = self._downloader.params.get('videopassword')
 732         if password is None:
 733             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 734         fields = self._hidden_inputs(login_form)
 735         token, vuid = self._extract_xsrft_and_vuid(webpage)
 736         fields['token'] = token
 737         fields['password'] = password
 738         post = urlencode_postdata(fields)
 739         password_path = self._search_regex(
 740             r'action="([^"]+)"', login_form, 'password URL')
 741         password_url = compat_urlparse.urljoin(page_url, password_path)
 742         password_request = sanitized_Request(password_url, post)
 743         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 744         self._set_vimeo_cookie('vuid', vuid)
 745         self._set_vimeo_cookie('xsrft', token)
 746
 747         return self._download_webpage(
 748             password_request, list_id,
 749             'Verifying the password', 'Wrong password')
 750
 751     def _title_and_entries(self, list_id, base_url):
 752         for pagenum in itertools.count(1):
 753             page_url = self._page_url(base_url, pagenum)
 754             webpage = self._download_webpage(
 755                 page_url, list_id,
 756                 'Downloading page %s' % pagenum)
 757
 758             if pagenum == 1:
 759                 webpage = self._login_list_password(page_url, list_id, webpage)
 760                 yield self._extract_list_title(webpage)
 761
 762             # Try extracting href first since not all videos are available via
 763             # short https://vimeo.com/id URL (e.g. https://vimeo.com/channels/tributes/6213729)
 764             clips = re.findall(
 765                 r'id="clip_(\d+)"[^>]*>\s*<a[^>]+href="(/(?:[^/]+/)*\1)(?:[^>]+\btitle="([^"]+)")?', webpage)
 766             if clips:
 767                 for video_id, video_url, video_title in clips:
 768                     yield self.url_result(
 769                         compat_urlparse.urljoin(base_url, video_url),
 770                         VimeoIE.ie_key(), video_id=video_id, video_title=video_title)
 771             # More relaxed fallback
 772             else:
 773                 for video_id in re.findall(r'id=["\']clip_(\d+)', webpage):
 774                     yield self.url_result(
 775                         'https://vimeo.com/%s' % video_id,
 776                         VimeoIE.ie_key(), video_id=video_id)
 777
 778             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 779                 break
 780
 781     def _extract_videos(self, list_id, base_url):
 782         title_and_entries = self._title_and_entries(list_id, base_url)
 783         list_title = next(title_and_entries)
 784         return self.playlist_result(title_and_entries, list_id, list_title)
 785
 786     def _real_extract(self, url):
 787         mobj = re.match(self._VALID_URL, url)
 788         channel_id = mobj.group('id')
 789         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 790
 791
 792 class VimeoUserIE(VimeoChannelIE):
 793     IE_NAME = 'vimeo:user'
 794     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 795     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 796     _TESTS = [{
 797         'url': 'https://vimeo.com/nkistudio/videos',
 798         'info_dict': {
 799             'title': 'Nki',
 800             'id': 'nkistudio',
 801         },
 802         'playlist_mincount': 66,
 803     }]
 804
 805     def _real_extract(self, url):
 806         mobj = re.match(self._VALID_URL, url)
 807         name = mobj.group('name')
 808         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 809
 810
 811 class VimeoAlbumIE(VimeoChannelIE):
 812     IE_NAME = 'vimeo:album'
 813     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)(?:$|[?#]|/(?!video))'
 814     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 815     _TESTS = [{
 816         'url': 'https://vimeo.com/album/2632481',
 817         'info_dict': {
 818             'id': '2632481',
 819             'title': 'Staff Favorites: November 2013',
 820         },
 821         'playlist_mincount': 13,
 822     }, {
 823         'note': 'Password-protected album',
 824         'url': 'https://vimeo.com/album/3253534',
 825         'info_dict': {
 826             'title': 'test',
 827             'id': '3253534',
 828         },
 829         'playlist_count': 1,
 830         'params': {
 831             'videopassword': 'youtube-dl',
 832         }
 833     }, {
 834         'url': 'https://vimeo.com/album/2632481/sort:plays/format:thumbnail',
 835         'only_matching': True,
 836     }, {
 837         # TODO: respect page number
 838         'url': 'https://vimeo.com/album/2632481/page:2/sort:plays/format:thumbnail',
 839         'only_matching': True,
 840     }]
 841
 842     def _page_url(self, base_url, pagenum):
 843         return '%s/page:%d/' % (base_url, pagenum)
 844
 845     def _real_extract(self, url):
 846         album_id = self._match_id(url)
 847         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 848
 849
 850 class VimeoGroupsIE(VimeoAlbumIE):
 851     IE_NAME = 'vimeo:group'
 852     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)(?:/(?!videos?/\d+)|$)'
 853     _TESTS = [{
 854         'url': 'https://vimeo.com/groups/rolexawards',
 855         'info_dict': {
 856             'id': 'rolexawards',
 857             'title': 'Rolex Awards for Enterprise',
 858         },
 859         'playlist_mincount': 73,
 860     }]
 861
 862     def _extract_list_title(self, webpage):
 863         return self._og_search_title(webpage)
 864
 865     def _real_extract(self, url):
 866         mobj = re.match(self._VALID_URL, url)
 867         name = mobj.group('name')
 868         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 869
 870
 871 class VimeoReviewIE(VimeoBaseInfoExtractor):
 872     IE_NAME = 'vimeo:review'
 873     IE_DESC = 'Review pages on vimeo'
 874     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 875     _TESTS = [{
 876         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 877         'md5': 'c507a72f780cacc12b2248bb4006d253',
 878         'info_dict': {
 879             'id': '75524534',
 880             'ext': 'mp4',
 881             'title': "DICK HARDWICK 'Comedian'",
 882             'uploader': 'Richard Hardwick',
 883             'uploader_id': 'user21297594',
 884         }
 885     }, {
 886         'note': 'video player needs Referer',
 887         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 888         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 889         'info_dict': {
 890             'id': '91613211',
 891             'ext': 'mp4',
 892             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 893             'uploader': 'DevWeek Events',
 894             'duration': 2773,
 895             'thumbnail': r're:^https?://.*\.jpg$',
 896             'uploader_id': 'user22258446',
 897         }
 898     }, {
 899         'note': 'Password protected',
 900         'url': 'https://vimeo.com/user37284429/review/138823582/c4d865efde',
 901         'info_dict': {
 902             'id': '138823582',
 903             'ext': 'mp4',
 904             'title': 'EFFICIENT PICKUP MASTERCLASS MODULE 1',
 905             'uploader': 'TMB',
 906             'uploader_id': 'user37284429',
 907         },
 908         'params': {
 909             'videopassword': 'holygrail',
 910         },
 911         'skip': 'video gone',
 912     }]
 913
 914     def _real_initialize(self):
 915         self._login()
 916
 917     def _get_config_url(self, webpage_url, video_id, video_password_verified=False):
 918         webpage = self._download_webpage(webpage_url, video_id)
 919         config_url = self._html_search_regex(
 920             r'data-config-url=(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 921             'config URL', default=None, group='url')
 922         if not config_url:
 923             data = self._parse_json(self._search_regex(
 924                 r'window\s*=\s*_extend\(window,\s*({.+?})\);', webpage, 'data',
 925                 default=NO_DEFAULT if video_password_verified else '{}'), video_id)
 926             config_url = data.get('vimeo_esi', {}).get('config', {}).get('configUrl')
 927         if config_url is None:
 928             self._verify_video_password(webpage_url, video_id, webpage)
 929             config_url = self._get_config_url(
 930                 webpage_url, video_id, video_password_verified=True)
 931         return config_url
 932
 933     def _real_extract(self, url):
 934         video_id = self._match_id(url)
 935         config_url = self._get_config_url(url, video_id)
 936         config = self._download_json(config_url, video_id)
 937         info_dict = self._parse_config(config, video_id)
 938         self._vimeo_sort_formats(info_dict['formats'])
 939         info_dict['id'] = video_id
 940         return info_dict
 941
 942
 943 class VimeoWatchLaterIE(VimeoChannelIE):
 944     IE_NAME = 'vimeo:watchlater'
 945     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 946     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 947     _TITLE = 'Watch Later'
 948     _LOGIN_REQUIRED = True
 949     _TESTS = [{
 950         'url': 'https://vimeo.com/watchlater',
 951         'only_matching': True,
 952     }]
 953
 954     def _real_initialize(self):
 955         self._login()
 956
 957     def _page_url(self, base_url, pagenum):
 958         url = '%s/page:%d/' % (base_url, pagenum)
 959         request = sanitized_Request(url)
 960         # Set the header to get a partial html page with the ids,
 961         # the normal page doesn't contain them.
 962         request.add_header('X-Requested-With', 'XMLHttpRequest')
 963         return request
 964
 965     def _real_extract(self, url):
 966         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 967
 968
 969 class VimeoLikesIE(InfoExtractor):
 970     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 971     IE_NAME = 'vimeo:likes'
 972     IE_DESC = 'Vimeo user likes'
 973     _TEST = {
 974         'url': 'https://vimeo.com/user755559/likes/',
 975         'playlist_mincount': 293,
 976         'info_dict': {
 977             'id': 'user755559_likes',
 978             'description': 'See all the videos urza likes',
 979             'title': 'Videos urza likes',
 980         },
 981     }
 982
 983     def _real_extract(self, url):
 984         user_id = self._match_id(url)
 985         webpage = self._download_webpage(url, user_id)
 986         page_count = self._int(
 987             self._search_regex(
 988                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 989                     .*?</a></li>\s*<li\s+class="pagination_next">
 990                 ''', webpage, 'page count'),
 991             'page count', fatal=True)
 992         PAGE_SIZE = 12
 993         title = self._html_search_regex(
 994             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 995         description = self._html_search_meta('description', webpage)
 996
 997         def _get_page(idx):
 998             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 999                 user_id, idx + 1)
1000             webpage = self._download_webpage(
1001                 page_url, user_id,
1002                 note='Downloading page %d/%d' % (idx + 1, page_count))
1003             video_list = self._search_regex(
1004                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
1005                 webpage, 'video content')
1006             paths = re.findall(
1007                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
1008             for path in paths:
1009                 yield {
1010                     '_type': 'url',
1011                     'url': compat_urlparse.urljoin(page_url, path),
1012                 }
1013
1014         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
1015
1016         return {
1017             '_type': 'playlist',
1018             'id': 'user%s_likes' % user_id,
1019             'title': title,
1020             'description': description,
1021             'entries': pl,
1022         }