youtube_dl/extractor/vimeo.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_urllib_parse,
  12     compat_urllib_request,
  13     compat_urlparse,
  14 )
  15 from ..utils import (
  16     ExtractorError,
  17     InAdvancePagedList,
  18     int_or_none,
  19     RegexNotFoundError,
  20     std_headers,
  21     unsmuggle_url,
  22     urlencode_postdata,
  23 )
  24
  25
  26 class VimeoBaseInfoExtractor(InfoExtractor):
  27     _NETRC_MACHINE = 'vimeo'
  28     _LOGIN_REQUIRED = False
  29
  30     def _login(self):
  31         (username, password) = self._get_login_info()
  32         if username is None:
  33             if self._LOGIN_REQUIRED:
  34                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  35             return
  36         self.report_login()
  37         login_url = 'https://vimeo.com/log_in'
  38         webpage = self._download_webpage(login_url, None, False)
  39         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
  40         data = urlencode_postdata({
  41             'email': username,
  42             'password': password,
  43             'action': 'login',
  44             'service': 'vimeo',
  45             'token': token,
  46         })
  47         login_request = compat_urllib_request.Request(login_url, data)
  48         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  49         login_request.add_header('Cookie', 'xsrft=%s' % token)
  50         self._download_webpage(login_request, None, False, 'Wrong login info')
  51
  52
  53 class VimeoIE(VimeoBaseInfoExtractor):
  54     """Information extractor for vimeo.com."""
  55
  56     # _VALID_URL matches Vimeo URLs
  57     _VALID_URL = r'''(?x)
  58         https?://
  59         (?:(?:www|(?P<player>player))\.)?
  60         vimeo(?P<pro>pro)?\.com/
  61         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  62         (?:.*?/)?
  63         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  64         (?:videos?/)?
  65         (?P<id>[0-9]+)
  66         /?(?:[?&].*)?(?:[#].*)?$'''
  67     IE_NAME = 'vimeo'
  68     _TESTS = [
  69         {
  70             'url': 'http://vimeo.com/56015672#at=0',
  71             'md5': '8879b6cc097e987f02484baf890129e5',
  72             'info_dict': {
  73                 'id': '56015672',
  74                 'ext': 'mp4',
  75                 "upload_date": "20121220",
  76                 "description": "This is a test case for youtube-dl.\nFor more information, see github.com/rg3/youtube-dl\nTest chars: \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  77                 "uploader_id": "user7108434",
  78                 "uploader": "Filippo Valsorda",
  79                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  80                 "duration": 10,
  81             },
  82         },
  83         {
  84             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  85             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  86             'note': 'Vimeo Pro video (#1197)',
  87             'info_dict': {
  88                 'id': '68093876',
  89                 'ext': 'mp4',
  90                 'uploader_id': 'openstreetmapus',
  91                 'uploader': 'OpenStreetMap US',
  92                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
  93                 'description': 'md5:380943ec71b89736ff4bf27183233d09',
  94                 'duration': 1595,
  95             },
  96         },
  97         {
  98             'url': 'http://player.vimeo.com/video/54469442',
  99             'md5': '619b811a4417aa4abe78dc653becf511',
 100             'note': 'Videos that embed the url in the player page',
 101             'info_dict': {
 102                 'id': '54469442',
 103                 'ext': 'mp4',
 104                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 105                 'uploader': 'The BLN & Business of Software',
 106                 'uploader_id': 'theblnbusinessofsoftware',
 107                 'duration': 3610,
 108                 'description': None,
 109             },
 110         },
 111         {
 112             'url': 'http://vimeo.com/68375962',
 113             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 114             'note': 'Video protected with password',
 115             'info_dict': {
 116                 'id': '68375962',
 117                 'ext': 'mp4',
 118                 'title': 'youtube-dl password protected test video',
 119                 'upload_date': '20130614',
 120                 'uploader_id': 'user18948128',
 121                 'uploader': 'Jaime Marquínez Ferrándiz',
 122                 'duration': 10,
 123                 'description': 'This is "youtube-dl password protected test video" by Jaime Marquínez Ferrándiz on Vimeo, the home for high quality videos and the people who love them.',
 124             },
 125             'params': {
 126                 'videopassword': 'youtube-dl',
 127             },
 128         },
 129         {
 130             'url': 'http://vimeo.com/channels/keypeele/75629013',
 131             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 132             'note': 'Video is freely available via original URL '
 133                     'and protected with password when accessed via http://vimeo.com/75629013',
 134             'info_dict': {
 135                 'id': '75629013',
 136                 'ext': 'mp4',
 137                 'title': 'Key & Peele: Terrorist Interrogation',
 138                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 139                 'uploader_id': 'atencio',
 140                 'uploader': 'Peter Atencio',
 141                 'duration': 187,
 142             },
 143         },
 144         {
 145             'url': 'http://vimeo.com/76979871',
 146             'md5': '3363dd6ffebe3784d56f4132317fd446',
 147             'note': 'Video with subtitles',
 148             'info_dict': {
 149                 'id': '76979871',
 150                 'ext': 'mp4',
 151                 'title': 'The New Vimeo Player (You Know, For Videos)',
 152                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 153                 'upload_date': '20131015',
 154                 'uploader_id': 'staff',
 155                 'uploader': 'Vimeo Staff',
 156                 'duration': 62,
 157             }
 158         },
 159         {
 160             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 161             'url': 'https://player.vimeo.com/video/98044508',
 162             'note': 'The js code contains assignments to the same variable as the config',
 163             'info_dict': {
 164                 'id': '98044508',
 165                 'ext': 'mp4',
 166                 'title': 'Pier Solar OUYA Official Trailer',
 167                 'uploader': 'Tulio Gonçalves',
 168                 'uploader_id': 'user28849593',
 169             },
 170         },
 171     ]
 172
 173     def _verify_video_password(self, url, video_id, webpage):
 174         password = self._downloader.params.get('videopassword', None)
 175         if password is None:
 176             raise ExtractorError('This video is protected by a password, use the --video-password option')
 177         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 178         data = compat_urllib_parse.urlencode({
 179             'password': password,
 180             'token': token,
 181         })
 182         # I didn't manage to use the password with https
 183         if url.startswith('https'):
 184             pass_url = url.replace('https', 'http')
 185         else:
 186             pass_url = url
 187         password_request = compat_urllib_request.Request(pass_url + '/password', data)
 188         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 189         password_request.add_header('Cookie', 'xsrft=%s' % token)
 190         return self._download_webpage(
 191             password_request, video_id,
 192             'Verifying the password', 'Wrong password')
 193
 194     def _verify_player_video_password(self, url, video_id):
 195         password = self._downloader.params.get('videopassword', None)
 196         if password is None:
 197             raise ExtractorError('This video is protected by a password, use the --video-password option')
 198         data = compat_urllib_parse.urlencode({'password': password})
 199         pass_url = url + '/check-password'
 200         password_request = compat_urllib_request.Request(pass_url, data)
 201         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 202         return self._download_json(
 203             password_request, video_id,
 204             'Verifying the password',
 205             'Wrong password')
 206
 207     def _real_initialize(self):
 208         self._login()
 209
 210     def _real_extract(self, url):
 211         url, data = unsmuggle_url(url)
 212         headers = std_headers
 213         if data is not None:
 214             headers = headers.copy()
 215             headers.update(data)
 216         if 'Referer' not in headers:
 217             headers['Referer'] = url
 218
 219         # Extract ID from URL
 220         mobj = re.match(self._VALID_URL, url)
 221         video_id = mobj.group('id')
 222         orig_url = url
 223         if mobj.group('pro') or mobj.group('player'):
 224             url = 'http://player.vimeo.com/video/' + video_id
 225
 226         # Retrieve video webpage to extract further information
 227         request = compat_urllib_request.Request(url, None, headers)
 228         try:
 229             webpage = self._download_webpage(request, video_id)
 230         except ExtractorError as ee:
 231             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 232                 errmsg = ee.cause.read()
 233                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 234                     raise ExtractorError(
 235                         'Cannot download embed-only video without embedding '
 236                         'URL. Please call youtube-dl with the URL of the page '
 237                         'that embeds this video.',
 238                         expected=True)
 239             raise
 240
 241         # Now we begin extracting as much information as we can from what we
 242         # retrieved. First we extract the information common to all extractors,
 243         # and latter we extract those that are Vimeo specific.
 244         self.report_extraction(video_id)
 245
 246         # Extract the config JSON
 247         try:
 248             try:
 249                 config_url = self._html_search_regex(
 250                     r' data-config-url="(.+?)"', webpage, 'config URL')
 251                 config_json = self._download_webpage(config_url, video_id)
 252                 config = json.loads(config_json)
 253             except RegexNotFoundError:
 254                 # For pro videos or player.vimeo.com urls
 255                 # We try to find out to which variable is assigned the config dic
 256                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 257                 if m_variable_name is not None:
 258                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 259                 else:
 260                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 261                 config = self._search_regex(config_re, webpage, 'info section',
 262                                             flags=re.DOTALL)
 263                 config = json.loads(config)
 264         except Exception as e:
 265             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 266                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 267
 268             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 269                 self._verify_video_password(url, video_id, webpage)
 270                 return self._real_extract(url)
 271             else:
 272                 raise ExtractorError('Unable to extract info section',
 273                                      cause=e)
 274         else:
 275             if config.get('view') == 4:
 276                 config = self._verify_player_video_password(url, video_id)
 277
 278         # Extract title
 279         video_title = config["video"]["title"]
 280
 281         # Extract uploader and uploader_id
 282         video_uploader = config["video"]["owner"]["name"]
 283         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 284
 285         # Extract video thumbnail
 286         video_thumbnail = config["video"].get("thumbnail")
 287         if video_thumbnail is None:
 288             video_thumbs = config["video"].get("thumbs")
 289             if video_thumbs and isinstance(video_thumbs, dict):
 290                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 291
 292         # Extract video description
 293
 294         video_description = self._html_search_regex(
 295             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 296             webpage, 'description', default=None)
 297         if not video_description:
 298             video_description = self._html_search_meta(
 299                 'description', webpage, default=None)
 300         if not video_description and mobj.group('pro'):
 301             orig_webpage = self._download_webpage(
 302                 orig_url, video_id,
 303                 note='Downloading webpage for description',
 304                 fatal=False)
 305             if orig_webpage:
 306                 video_description = self._html_search_meta(
 307                     'description', orig_webpage, default=None)
 308         if not video_description and not mobj.group('player'):
 309             self._downloader.report_warning('Cannot find video description')
 310
 311         # Extract video duration
 312         video_duration = int_or_none(config["video"].get("duration"))
 313
 314         # Extract upload date
 315         video_upload_date = None
 316         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 317         if mobj is not None:
 318             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 319
 320         try:
 321             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 322             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 323             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 324         except RegexNotFoundError:
 325             # This info is only available in vimeo.com/{id} urls
 326             view_count = None
 327             like_count = None
 328             comment_count = None
 329
 330         # Vimeo specific: extract request signature and timestamp
 331         sig = config['request']['signature']
 332         timestamp = config['request']['timestamp']
 333
 334         # Vimeo specific: extract video codec and quality information
 335         # First consider quality, then codecs, then take everything
 336         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 337         files = {'hd': [], 'sd': [], 'other': []}
 338         config_files = config["video"].get("files") or config["request"].get("files")
 339         for codec_name, codec_extension in codecs:
 340             for quality in config_files.get(codec_name, []):
 341                 format_id = '-'.join((codec_name, quality)).lower()
 342                 key = quality if quality in files else 'other'
 343                 video_url = None
 344                 if isinstance(config_files[codec_name], dict):
 345                     file_info = config_files[codec_name][quality]
 346                     video_url = file_info.get('url')
 347                 else:
 348                     file_info = {}
 349                 if video_url is None:
 350                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 351                         % (video_id, sig, timestamp, quality, codec_name.upper())
 352
 353                 files[key].append({
 354                     'ext': codec_extension,
 355                     'url': video_url,
 356                     'format_id': format_id,
 357                     'width': file_info.get('width'),
 358                     'height': file_info.get('height'),
 359                 })
 360         formats = []
 361         for key in ('other', 'sd', 'hd'):
 362             formats += files[key]
 363         if len(formats) == 0:
 364             raise ExtractorError('No known codec found')
 365
 366         subtitles = {}
 367         text_tracks = config['request'].get('text_tracks')
 368         if text_tracks:
 369             for tt in text_tracks:
 370                 subtitles[tt['lang']] = [{
 371                     'ext': 'vtt',
 372                     'url': 'http://vimeo.com' + tt['url'],
 373                 }]
 374
 375         return {
 376             'id': video_id,
 377             'uploader': video_uploader,
 378             'uploader_id': video_uploader_id,
 379             'upload_date': video_upload_date,
 380             'title': video_title,
 381             'thumbnail': video_thumbnail,
 382             'description': video_description,
 383             'duration': video_duration,
 384             'formats': formats,
 385             'webpage_url': url,
 386             'view_count': view_count,
 387             'like_count': like_count,
 388             'comment_count': comment_count,
 389             'subtitles': subtitles,
 390         }
 391
 392
 393 class VimeoChannelIE(InfoExtractor):
 394     IE_NAME = 'vimeo:channel'
 395     _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 396     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 397     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 398     _TESTS = [{
 399         'url': 'http://vimeo.com/channels/tributes',
 400         'info_dict': {
 401             'title': 'Vimeo Tributes',
 402         },
 403         'playlist_mincount': 25,
 404     }]
 405
 406     def _page_url(self, base_url, pagenum):
 407         return '%s/videos/page:%d/' % (base_url, pagenum)
 408
 409     def _extract_list_title(self, webpage):
 410         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 411
 412     def _login_list_password(self, page_url, list_id, webpage):
 413         login_form = self._search_regex(
 414             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 415             webpage, 'login form', default=None)
 416         if not login_form:
 417             return webpage
 418
 419         password = self._downloader.params.get('videopassword', None)
 420         if password is None:
 421             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 422         fields = dict(re.findall(r'''(?x)<input\s+
 423             type="hidden"\s+
 424             name="([^"]+)"\s+
 425             value="([^"]*)"
 426             ''', login_form))
 427         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 428         fields['token'] = token
 429         fields['password'] = password
 430         post = compat_urllib_parse.urlencode(fields)
 431         password_path = self._search_regex(
 432             r'action="([^"]+)"', login_form, 'password URL')
 433         password_url = compat_urlparse.urljoin(page_url, password_path)
 434         password_request = compat_urllib_request.Request(password_url, post)
 435         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 436         self._set_cookie('vimeo.com', 'xsrft', token)
 437
 438         return self._download_webpage(
 439             password_request, list_id,
 440             'Verifying the password', 'Wrong password')
 441
 442     def _extract_videos(self, list_id, base_url):
 443         video_ids = []
 444         for pagenum in itertools.count(1):
 445             page_url = self._page_url(base_url, pagenum)
 446             webpage = self._download_webpage(
 447                 page_url, list_id,
 448                 'Downloading page %s' % pagenum)
 449
 450             if pagenum == 1:
 451                 webpage = self._login_list_password(page_url, list_id, webpage)
 452
 453             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 454             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 455                 break
 456
 457         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 458                    for video_id in video_ids]
 459         return {'_type': 'playlist',
 460                 'id': list_id,
 461                 'title': self._extract_list_title(webpage),
 462                 'entries': entries,
 463                 }
 464
 465     def _real_extract(self, url):
 466         mobj = re.match(self._VALID_URL, url)
 467         channel_id = mobj.group('id')
 468         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 469
 470
 471 class VimeoUserIE(VimeoChannelIE):
 472     IE_NAME = 'vimeo:user'
 473     _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 474     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 475     _TESTS = [{
 476         'url': 'http://vimeo.com/nkistudio/videos',
 477         'info_dict': {
 478             'title': 'Nki',
 479         },
 480         'playlist_mincount': 66,
 481     }]
 482
 483     def _real_extract(self, url):
 484         mobj = re.match(self._VALID_URL, url)
 485         name = mobj.group('name')
 486         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 487
 488
 489 class VimeoAlbumIE(VimeoChannelIE):
 490     IE_NAME = 'vimeo:album'
 491     _VALID_URL = r'https?://vimeo\.com/album/(?P<id>\d+)'
 492     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 493     _TESTS = [{
 494         'url': 'http://vimeo.com/album/2632481',
 495         'info_dict': {
 496             'title': 'Staff Favorites: November 2013',
 497         },
 498         'playlist_mincount': 13,
 499     }, {
 500         'note': 'Password-protected album',
 501         'url': 'https://vimeo.com/album/3253534',
 502         'info_dict': {
 503             'title': 'test',
 504             'id': '3253534',
 505         },
 506         'playlist_count': 1,
 507         'params': {
 508             'videopassword': 'youtube-dl',
 509         }
 510     }]
 511
 512     def _page_url(self, base_url, pagenum):
 513         return '%s/page:%d/' % (base_url, pagenum)
 514
 515     def _real_extract(self, url):
 516         album_id = self._match_id(url)
 517         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 518
 519
 520 class VimeoGroupsIE(VimeoAlbumIE):
 521     IE_NAME = 'vimeo:group'
 522     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 523     _TESTS = [{
 524         'url': 'http://vimeo.com/groups/rolexawards',
 525         'info_dict': {
 526             'title': 'Rolex Awards for Enterprise',
 527         },
 528         'playlist_mincount': 73,
 529     }]
 530
 531     def _extract_list_title(self, webpage):
 532         return self._og_search_title(webpage)
 533
 534     def _real_extract(self, url):
 535         mobj = re.match(self._VALID_URL, url)
 536         name = mobj.group('name')
 537         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 538
 539
 540 class VimeoReviewIE(InfoExtractor):
 541     IE_NAME = 'vimeo:review'
 542     IE_DESC = 'Review pages on vimeo'
 543     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 544     _TESTS = [{
 545         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 546         'md5': 'c507a72f780cacc12b2248bb4006d253',
 547         'info_dict': {
 548             'id': '75524534',
 549             'ext': 'mp4',
 550             'title': "DICK HARDWICK 'Comedian'",
 551             'uploader': 'Richard Hardwick',
 552         }
 553     }, {
 554         'note': 'video player needs Referer',
 555         'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
 556         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 557         'info_dict': {
 558             'id': '91613211',
 559             'ext': 'mp4',
 560             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 561             'uploader': 'DevWeek Events',
 562             'duration': 2773,
 563             'thumbnail': 're:^https?://.*\.jpg$',
 564         }
 565     }]
 566
 567     def _real_extract(self, url):
 568         mobj = re.match(self._VALID_URL, url)
 569         video_id = mobj.group('id')
 570         player_url = 'https://player.vimeo.com/player/' + video_id
 571         return self.url_result(player_url, 'Vimeo', video_id)
 572
 573
 574 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 575     IE_NAME = 'vimeo:watchlater'
 576     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 577     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 578     _LOGIN_REQUIRED = True
 579     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 580     _TESTS = [{
 581         'url': 'http://vimeo.com/home/watchlater',
 582         'only_matching': True,
 583     }]
 584
 585     def _real_initialize(self):
 586         self._login()
 587
 588     def _page_url(self, base_url, pagenum):
 589         url = '%s/page:%d/' % (base_url, pagenum)
 590         request = compat_urllib_request.Request(url)
 591         # Set the header to get a partial html page with the ids,
 592         # the normal page doesn't contain them.
 593         request.add_header('X-Requested-With', 'XMLHttpRequest')
 594         return request
 595
 596     def _real_extract(self, url):
 597         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
 598
 599
 600 class VimeoLikesIE(InfoExtractor):
 601     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 602     IE_NAME = 'vimeo:likes'
 603     IE_DESC = 'Vimeo user likes'
 604     _TEST = {
 605         'url': 'https://vimeo.com/user755559/likes/',
 606         'playlist_mincount': 293,
 607         "info_dict": {
 608             "description": "See all the videos urza likes",
 609             "title": 'Videos urza likes',
 610         },
 611     }
 612
 613     def _real_extract(self, url):
 614         user_id = self._match_id(url)
 615         webpage = self._download_webpage(url, user_id)
 616         page_count = self._int(
 617             self._search_regex(
 618                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 619                     .*?</a></li>\s*<li\s+class="pagination_next">
 620                 ''', webpage, 'page count'),
 621             'page count', fatal=True)
 622         PAGE_SIZE = 12
 623         title = self._html_search_regex(
 624             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 625         description = self._html_search_meta('description', webpage)
 626
 627         def _get_page(idx):
 628             page_url = '%s//vimeo.com/user%s/likes/page:%d/sort:date' % (
 629                 self.http_scheme(), user_id, idx + 1)
 630             webpage = self._download_webpage(
 631                 page_url, user_id,
 632                 note='Downloading page %d/%d' % (idx + 1, page_count))
 633             video_list = self._search_regex(
 634                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 635                 webpage, 'video content')
 636             paths = re.findall(
 637                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 638             for path in paths:
 639                 yield {
 640                     '_type': 'url',
 641                     'url': compat_urlparse.urljoin(page_url, path),
 642                 }
 643
 644         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 645
 646         return {
 647             '_type': 'playlist',
 648             'id': 'user%s_likes' % user_id,
 649             'title': title,
 650             'description': description,
 651             'entries': pl,
 652         }