_ Git - youtube-dl/blob - 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     smuggle_url,
  21     std_headers,
  22     unified_strdate,
  23     unsmuggle_url,
  24     urlencode_postdata,
  25     unescapeHTML,
  26 )
  27
  28
  29 class VimeoBaseInfoExtractor(InfoExtractor):
  30     _NETRC_MACHINE = 'vimeo'
  31     _LOGIN_REQUIRED = False
  32     _LOGIN_URL = 'https://vimeo.com/log_in'
  33
  34     def _login(self):
  35         (username, password) = self._get_login_info()
  36         if username is None:
  37             if self._LOGIN_REQUIRED:
  38                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  39             return
  40         self.report_login()
  41         webpage = self._download_webpage(self._LOGIN_URL, None, False)
  42         token = self._extract_xsrft(webpage)
  43         vuid = self._search_regex(
  44             r'["\']vuid["\']\s*:\s*(["\'])(?P<vuid>.+?)\1',
  45             webpage, 'vuid', group='vuid')
  46         data = urlencode_postdata({
  47             'action': 'login',
  48             'email': username,
  49             'password': password,
  50             'service': 'vimeo',
  51             'token': token,
  52         })
  53         login_request = compat_urllib_request.Request(self._LOGIN_URL, data)
  54         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  55         login_request.add_header('Cookie', 'vuid=%s' % vuid)
  56         login_request.add_header('Referer', self._LOGIN_URL)
  57         self._download_webpage(login_request, None, False, 'Wrong login info')
  58
  59     def _extract_xsrft(self, webpage):
  60         return self._search_regex(
  61             r'xsrft\s*[=:]\s*(?P<q>["\'])(?P<xsrft>.+?)(?P=q)',
  62             webpage, 'login token', group='xsrft')
  63
  64
  65 class VimeoIE(VimeoBaseInfoExtractor):
  66     """Information extractor for vimeo.com."""
  67
  68     # _VALID_URL matches Vimeo URLs
  69     _VALID_URL = r'''(?x)
  70         https?://
  71         (?:(?:www|(?P<player>player))\.)?
  72         vimeo(?P<pro>pro)?\.com/
  73         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  74         (?:.*?/)?
  75         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  76         (?:videos?/)?
  77         (?P<id>[0-9]+)
  78         /?(?:[?&].*)?(?:[#].*)?$'''
  79     IE_NAME = 'vimeo'
  80     _TESTS = [
  81         {
  82             'url': 'http://vimeo.com/56015672#at=0',
  83             'md5': '8879b6cc097e987f02484baf890129e5',
  84             'info_dict': {
  85                 'id': '56015672',
  86                 'ext': 'mp4',
  87                 'title': "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  88                 'description': 'md5:2d3305bad981a06ff79f027f19865021',
  89                 'upload_date': '20121220',
  90                 'uploader_id': 'user7108434',
  91                 'uploader': 'Filippo Valsorda',
  92                 'duration': 10,
  93             },
  94         },
  95         {
  96             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  97             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  98             'note': 'Vimeo Pro video (#1197)',
  99             'info_dict': {
 100                 'id': '68093876',
 101                 'ext': 'mp4',
 102                 'uploader_id': 'openstreetmapus',
 103                 'uploader': 'OpenStreetMap US',
 104                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
 105                 'description': 'md5:fd69a7b8d8c34a4e1d2ec2e4afd6ec30',
 106                 'duration': 1595,
 107             },
 108         },
 109         {
 110             'url': 'http://player.vimeo.com/video/54469442',
 111             'md5': '619b811a4417aa4abe78dc653becf511',
 112             'note': 'Videos that embed the url in the player page',
 113             'info_dict': {
 114                 'id': '54469442',
 115                 'ext': 'mp4',
 116                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 117                 'uploader': 'The BLN & Business of Software',
 118                 'uploader_id': 'theblnbusinessofsoftware',
 119                 'duration': 3610,
 120                 'description': None,
 121             },
 122         },
 123         {
 124             'url': 'http://vimeo.com/68375962',
 125             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 126             'note': 'Video protected with password',
 127             'info_dict': {
 128                 'id': '68375962',
 129                 'ext': 'mp4',
 130                 'title': 'youtube-dl password protected test video',
 131                 'upload_date': '20130614',
 132                 'uploader_id': 'user18948128',
 133                 'uploader': 'Jaime Marquínez Ferrándiz',
 134                 'duration': 10,
 135                 '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.',
 136             },
 137             'params': {
 138                 'videopassword': 'youtube-dl',
 139             },
 140         },
 141         {
 142             'url': 'http://vimeo.com/channels/keypeele/75629013',
 143             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 144             'note': 'Video is freely available via original URL '
 145                     'and protected with password when accessed via http://vimeo.com/75629013',
 146             'info_dict': {
 147                 'id': '75629013',
 148                 'ext': 'mp4',
 149                 'title': 'Key & Peele: Terrorist Interrogation',
 150                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 151                 'uploader_id': 'atencio',
 152                 'uploader': 'Peter Atencio',
 153                 'upload_date': '20130927',
 154                 'duration': 187,
 155             },
 156         },
 157         {
 158             'url': 'http://vimeo.com/76979871',
 159             'note': 'Video with subtitles',
 160             'info_dict': {
 161                 'id': '76979871',
 162                 'ext': 'mp4',
 163                 'title': 'The New Vimeo Player (You Know, For Videos)',
 164                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 165                 'upload_date': '20131015',
 166                 'uploader_id': 'staff',
 167                 'uploader': 'Vimeo Staff',
 168                 'duration': 62,
 169             }
 170         },
 171         {
 172             # from https://www.ouya.tv/game/Pier-Solar-and-the-Great-Architects/
 173             'url': 'https://player.vimeo.com/video/98044508',
 174             'note': 'The js code contains assignments to the same variable as the config',
 175             'info_dict': {
 176                 'id': '98044508',
 177                 'ext': 'mp4',
 178                 'title': 'Pier Solar OUYA Official Trailer',
 179                 'uploader': 'Tulio Gonçalves',
 180                 'uploader_id': 'user28849593',
 181             },
 182         },
 183     ]
 184
 185     @staticmethod
 186     def _extract_vimeo_url(url, webpage):
 187         # Look for embedded (iframe) Vimeo player
 188         mobj = re.search(
 189             r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage)
 190         if mobj:
 191             player_url = unescapeHTML(mobj.group('url'))
 192             surl = smuggle_url(player_url, {'Referer': url})
 193             return surl
 194         # Look for embedded (swf embed) Vimeo player
 195         mobj = re.search(
 196             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
 197         if mobj:
 198             return mobj.group(1)
 199
 200     def _verify_video_password(self, url, video_id, webpage):
 201         password = self._downloader.params.get('videopassword', None)
 202         if password is None:
 203             raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
 204         token = self._extract_xsrft(webpage)
 205         data = urlencode_postdata({
 206             'password': password,
 207             'token': token,
 208         })
 209         if url.startswith('http://'):
 210             # vimeo only supports https now, but the user can give an http url
 211             url = url.replace('http://', 'https://')
 212         password_request = compat_urllib_request.Request(url + '/password', data)
 213         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 214         password_request.add_header('Referer', url)
 215         return self._download_webpage(
 216             password_request, video_id,
 217             'Verifying the password', 'Wrong password')
 218
 219     def _verify_player_video_password(self, url, video_id):
 220         password = self._downloader.params.get('videopassword', None)
 221         if password is None:
 222             raise ExtractorError('This video is protected by a password, use the --video-password option')
 223         data = compat_urllib_parse.urlencode({'password': password})
 224         pass_url = url + '/check-password'
 225         password_request = compat_urllib_request.Request(pass_url, data)
 226         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 227         return self._download_json(
 228             password_request, video_id,
 229             'Verifying the password',
 230             'Wrong password')
 231
 232     def _real_initialize(self):
 233         self._login()
 234
 235     def _real_extract(self, url):
 236         url, data = unsmuggle_url(url)
 237         headers = std_headers
 238         if data is not None:
 239             headers = headers.copy()
 240             headers.update(data)
 241         if 'Referer' not in headers:
 242             headers['Referer'] = url
 243
 244         # Extract ID from URL
 245         mobj = re.match(self._VALID_URL, url)
 246         video_id = mobj.group('id')
 247         orig_url = url
 248         if mobj.group('pro') or mobj.group('player'):
 249             url = 'https://player.vimeo.com/video/' + video_id
 250         else:
 251             url = 'https://vimeo.com/' + video_id
 252
 253         # Retrieve video webpage to extract further information
 254         request = compat_urllib_request.Request(url, None, headers)
 255         try:
 256             webpage = self._download_webpage(request, video_id)
 257         except ExtractorError as ee:
 258             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 259                 errmsg = ee.cause.read()
 260                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 261                     raise ExtractorError(
 262                         'Cannot download embed-only video without embedding '
 263                         'URL. Please call youtube-dl with the URL of the page '
 264                         'that embeds this video.',
 265                         expected=True)
 266             raise
 267
 268         # Now we begin extracting as much information as we can from what we
 269         # retrieved. First we extract the information common to all extractors,
 270         # and latter we extract those that are Vimeo specific.
 271         self.report_extraction(video_id)
 272
 273         vimeo_config = self._search_regex(
 274             r'vimeo\.config\s*=\s*({.+?});', webpage,
 275             'vimeo config', default=None)
 276         if vimeo_config:
 277             seed_status = self._parse_json(vimeo_config, video_id).get('seed_status', {})
 278             if seed_status.get('state') == 'failed':
 279                 raise ExtractorError(
 280                     '%s returned error: %s' % (self.IE_NAME, seed_status['title']),
 281                     expected=True)
 282
 283         # Extract the config JSON
 284         try:
 285             try:
 286                 config_url = self._html_search_regex(
 287                     r' data-config-url="(.+?)"', webpage, 'config URL')
 288                 config_json = self._download_webpage(config_url, video_id)
 289                 config = json.loads(config_json)
 290             except RegexNotFoundError:
 291                 # For pro videos or player.vimeo.com urls
 292                 # We try to find out to which variable is assigned the config dic
 293                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 294                 if m_variable_name is not None:
 295                     config_re = r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))
 296                 else:
 297                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 298                 config = self._search_regex(config_re, webpage, 'info section',
 299                                             flags=re.DOTALL)
 300                 config = json.loads(config)
 301         except Exception as e:
 302             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 303                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 304
 305             if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None:
 306                 if data and '_video_password_verified' in data:
 307                     raise ExtractorError('video password verification failed!')
 308                 self._verify_video_password(url, video_id, webpage)
 309                 return self._real_extract(
 310                     smuggle_url(url, {'_video_password_verified': 'verified'}))
 311             else:
 312                 raise ExtractorError('Unable to extract info section',
 313                                      cause=e)
 314         else:
 315             if config.get('view') == 4:
 316                 config = self._verify_player_video_password(url, video_id)
 317
 318         # Extract title
 319         video_title = config["video"]["title"]
 320
 321         # Extract uploader and uploader_id
 322         video_uploader = config["video"]["owner"]["name"]
 323         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 324
 325         # Extract video thumbnail
 326         video_thumbnail = config["video"].get("thumbnail")
 327         if video_thumbnail is None:
 328             video_thumbs = config["video"].get("thumbs")
 329             if video_thumbs and isinstance(video_thumbs, dict):
 330                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 331
 332         # Extract video description
 333
 334         video_description = self._html_search_regex(
 335             r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>',
 336             webpage, 'description', default=None)
 337         if not video_description:
 338             video_description = self._html_search_meta(
 339                 'description', webpage, default=None)
 340         if not video_description and mobj.group('pro'):
 341             orig_webpage = self._download_webpage(
 342                 orig_url, video_id,
 343                 note='Downloading webpage for description',
 344                 fatal=False)
 345             if orig_webpage:
 346                 video_description = self._html_search_meta(
 347                     'description', orig_webpage, default=None)
 348         if not video_description and not mobj.group('player'):
 349             self._downloader.report_warning('Cannot find video description')
 350
 351         # Extract video duration
 352         video_duration = int_or_none(config["video"].get("duration"))
 353
 354         # Extract upload date
 355         video_upload_date = None
 356         mobj = re.search(r'<time[^>]+datetime="([^"]+)"', webpage)
 357         if mobj is not None:
 358             video_upload_date = unified_strdate(mobj.group(1))
 359
 360         try:
 361             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 362             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 363             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 364         except RegexNotFoundError:
 365             # This info is only available in vimeo.com/{id} urls
 366             view_count = None
 367             like_count = None
 368             comment_count = None
 369
 370         # Vimeo specific: extract request signature and timestamp
 371         sig = config['request']['signature']
 372         timestamp = config['request']['timestamp']
 373
 374         # Vimeo specific: extract video codec and quality information
 375         # First consider quality, then codecs, then take everything
 376         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 377         files = {'hd': [], 'sd': [], 'other': []}
 378         config_files = config["video"].get("files") or config["request"].get("files")
 379         for codec_name, codec_extension in codecs:
 380             for quality in config_files.get(codec_name, []):
 381                 format_id = '-'.join((codec_name, quality)).lower()
 382                 key = quality if quality in files else 'other'
 383                 video_url = None
 384                 if isinstance(config_files[codec_name], dict):
 385                     file_info = config_files[codec_name][quality]
 386                     video_url = file_info.get('url')
 387                 else:
 388                     file_info = {}
 389                 if video_url is None:
 390                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 391                         % (video_id, sig, timestamp, quality, codec_name.upper())
 392
 393                 files[key].append({
 394                     'ext': codec_extension,
 395                     'url': video_url,
 396                     'format_id': format_id,
 397                     'width': int_or_none(file_info.get('width')),
 398                     'height': int_or_none(file_info.get('height')),
 399                     'tbr': int_or_none(file_info.get('bitrate')),
 400                 })
 401         formats = []
 402         m3u8_url = config_files.get('hls', {}).get('all')
 403         if m3u8_url:
 404             formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', 'm3u8_native', 0, 'hls', fatal=False)
 405         for key in ('other', 'sd', 'hd'):
 406             formats += files[key]
 407         self._sort_formats(formats)
 408
 409         subtitles = {}
 410         text_tracks = config['request'].get('text_tracks')
 411         if text_tracks:
 412             for tt in text_tracks:
 413                 subtitles[tt['lang']] = [{
 414                     'ext': 'vtt',
 415                     'url': 'https://vimeo.com' + tt['url'],
 416                 }]
 417
 418         return {
 419             'id': video_id,
 420             'uploader': video_uploader,
 421             'uploader_id': video_uploader_id,
 422             'upload_date': video_upload_date,
 423             'title': video_title,
 424             'thumbnail': video_thumbnail,
 425             'description': video_description,
 426             'duration': video_duration,
 427             'formats': formats,
 428             'webpage_url': url,
 429             'view_count': view_count,
 430             'like_count': like_count,
 431             'comment_count': comment_count,
 432             'subtitles': subtitles,
 433         }
 434
 435
 436 class VimeoChannelIE(VimeoBaseInfoExtractor):
 437     IE_NAME = 'vimeo:channel'
 438     _VALID_URL = r'https://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 439     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 440     _TITLE = None
 441     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 442     _TESTS = [{
 443         'url': 'https://vimeo.com/channels/tributes',
 444         'info_dict': {
 445             'id': 'tributes',
 446             'title': 'Vimeo Tributes',
 447         },
 448         'playlist_mincount': 25,
 449     }]
 450
 451     def _page_url(self, base_url, pagenum):
 452         return '%s/videos/page:%d/' % (base_url, pagenum)
 453
 454     def _extract_list_title(self, webpage):
 455         return self._TITLE or self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 456
 457     def _login_list_password(self, page_url, list_id, webpage):
 458         login_form = self._search_regex(
 459             r'(?s)<form[^>]+?id="pw_form"(.*?)</form>',
 460             webpage, 'login form', default=None)
 461         if not login_form:
 462             return webpage
 463
 464         password = self._downloader.params.get('videopassword', None)
 465         if password is None:
 466             raise ExtractorError('This album is protected by a password, use the --video-password option', expected=True)
 467         fields = self._hidden_inputs(login_form)
 468         token = self._extract_xsrft(webpage)
 469         fields['token'] = token
 470         fields['password'] = password
 471         post = urlencode_postdata(fields)
 472         password_path = self._search_regex(
 473             r'action="([^"]+)"', login_form, 'password URL')
 474         password_url = compat_urlparse.urljoin(page_url, password_path)
 475         password_request = compat_urllib_request.Request(password_url, post)
 476         password_request.add_header('Content-type', 'application/x-www-form-urlencoded')
 477         self._set_cookie('vimeo.com', 'xsrft', token)
 478
 479         return self._download_webpage(
 480             password_request, list_id,
 481             'Verifying the password', 'Wrong password')
 482
 483     def _extract_videos(self, list_id, base_url):
 484         video_ids = []
 485         for pagenum in itertools.count(1):
 486             page_url = self._page_url(base_url, pagenum)
 487             webpage = self._download_webpage(
 488                 page_url, list_id,
 489                 'Downloading page %s' % pagenum)
 490
 491             if pagenum == 1:
 492                 webpage = self._login_list_password(page_url, list_id, webpage)
 493
 494             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 495             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 496                 break
 497
 498         entries = [self.url_result('https://vimeo.com/%s' % video_id, 'Vimeo')
 499                    for video_id in video_ids]
 500         return {'_type': 'playlist',
 501                 'id': list_id,
 502                 'title': self._extract_list_title(webpage),
 503                 'entries': entries,
 504                 }
 505
 506     def _real_extract(self, url):
 507         mobj = re.match(self._VALID_URL, url)
 508         channel_id = mobj.group('id')
 509         return self._extract_videos(channel_id, 'https://vimeo.com/channels/%s' % channel_id)
 510
 511
 512 class VimeoUserIE(VimeoChannelIE):
 513     IE_NAME = 'vimeo:user'
 514     _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 515     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 516     _TESTS = [{
 517         'url': 'https://vimeo.com/nkistudio/videos',
 518         'info_dict': {
 519             'title': 'Nki',
 520             'id': 'nkistudio',
 521         },
 522         'playlist_mincount': 66,
 523     }]
 524
 525     def _real_extract(self, url):
 526         mobj = re.match(self._VALID_URL, url)
 527         name = mobj.group('name')
 528         return self._extract_videos(name, 'https://vimeo.com/%s' % name)
 529
 530
 531 class VimeoAlbumIE(VimeoChannelIE):
 532     IE_NAME = 'vimeo:album'
 533     _VALID_URL = r'https://vimeo\.com/album/(?P<id>\d+)'
 534     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 535     _TESTS = [{
 536         'url': 'https://vimeo.com/album/2632481',
 537         'info_dict': {
 538             'id': '2632481',
 539             'title': 'Staff Favorites: November 2013',
 540         },
 541         'playlist_mincount': 13,
 542     }, {
 543         'note': 'Password-protected album',
 544         'url': 'https://vimeo.com/album/3253534',
 545         'info_dict': {
 546             'title': 'test',
 547             'id': '3253534',
 548         },
 549         'playlist_count': 1,
 550         'params': {
 551             'videopassword': 'youtube-dl',
 552         }
 553     }]
 554
 555     def _page_url(self, base_url, pagenum):
 556         return '%s/page:%d/' % (base_url, pagenum)
 557
 558     def _real_extract(self, url):
 559         album_id = self._match_id(url)
 560         return self._extract_videos(album_id, 'https://vimeo.com/album/%s' % album_id)
 561
 562
 563 class VimeoGroupsIE(VimeoAlbumIE):
 564     IE_NAME = 'vimeo:group'
 565     _VALID_URL = r'https://vimeo\.com/groups/(?P<name>[^/]+)'
 566     _TESTS = [{
 567         'url': 'https://vimeo.com/groups/rolexawards',
 568         'info_dict': {
 569             'id': 'rolexawards',
 570             'title': 'Rolex Awards for Enterprise',
 571         },
 572         'playlist_mincount': 73,
 573     }]
 574
 575     def _extract_list_title(self, webpage):
 576         return self._og_search_title(webpage)
 577
 578     def _real_extract(self, url):
 579         mobj = re.match(self._VALID_URL, url)
 580         name = mobj.group('name')
 581         return self._extract_videos(name, 'https://vimeo.com/groups/%s' % name)
 582
 583
 584 class VimeoReviewIE(InfoExtractor):
 585     IE_NAME = 'vimeo:review'
 586     IE_DESC = 'Review pages on vimeo'
 587     _VALID_URL = r'https://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 588     _TESTS = [{
 589         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 590         'md5': 'c507a72f780cacc12b2248bb4006d253',
 591         'info_dict': {
 592             'id': '75524534',
 593             'ext': 'mp4',
 594             'title': "DICK HARDWICK 'Comedian'",
 595             'uploader': 'Richard Hardwick',
 596         }
 597     }, {
 598         'note': 'video player needs Referer',
 599         'url': 'https://vimeo.com/user22258446/review/91613211/13f927e053',
 600         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 601         'info_dict': {
 602             'id': '91613211',
 603             'ext': 'mp4',
 604             'title': 're:(?i)^Death by dogma versus assembling agile . Sander Hoogendoorn',
 605             'uploader': 'DevWeek Events',
 606             'duration': 2773,
 607             'thumbnail': 're:^https?://.*\.jpg$',
 608         }
 609     }]
 610
 611     def _real_extract(self, url):
 612         mobj = re.match(self._VALID_URL, url)
 613         video_id = mobj.group('id')
 614         player_url = 'https://player.vimeo.com/player/' + video_id
 615         return self.url_result(player_url, 'Vimeo', video_id)
 616
 617
 618 class VimeoWatchLaterIE(VimeoChannelIE):
 619     IE_NAME = 'vimeo:watchlater'
 620     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 621     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
 622     _TITLE = 'Watch Later'
 623     _LOGIN_REQUIRED = True
 624     _TESTS = [{
 625         'url': 'https://vimeo.com/watchlater',
 626         'only_matching': True,
 627     }]
 628
 629     def _real_initialize(self):
 630         self._login()
 631
 632     def _page_url(self, base_url, pagenum):
 633         url = '%s/page:%d/' % (base_url, pagenum)
 634         request = compat_urllib_request.Request(url)
 635         # Set the header to get a partial html page with the ids,
 636         # the normal page doesn't contain them.
 637         request.add_header('X-Requested-With', 'XMLHttpRequest')
 638         return request
 639
 640     def _real_extract(self, url):
 641         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 642
 643
 644 class VimeoLikesIE(InfoExtractor):
 645     _VALID_URL = r'https://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes/?(?:$|[?#]|sort:)'
 646     IE_NAME = 'vimeo:likes'
 647     IE_DESC = 'Vimeo user likes'
 648     _TEST = {
 649         'url': 'https://vimeo.com/user755559/likes/',
 650         'playlist_mincount': 293,
 651         "info_dict": {
 652             'id': 'user755559_likes',
 653             "description": "See all the videos urza likes",
 654             "title": 'Videos urza likes',
 655         },
 656     }
 657
 658     def _real_extract(self, url):
 659         user_id = self._match_id(url)
 660         webpage = self._download_webpage(url, user_id)
 661         page_count = self._int(
 662             self._search_regex(
 663                 r'''(?x)<li><a\s+href="[^"]+"\s+data-page="([0-9]+)">
 664                     .*?</a></li>\s*<li\s+class="pagination_next">
 665                 ''', webpage, 'page count'),
 666             'page count', fatal=True)
 667         PAGE_SIZE = 12
 668         title = self._html_search_regex(
 669             r'(?s)<h1>(.+?)</h1>', webpage, 'title', fatal=False)
 670         description = self._html_search_meta('description', webpage)
 671
 672         def _get_page(idx):
 673             page_url = 'https://vimeo.com/user%s/likes/page:%d/sort:date' % (
 674                 user_id, idx + 1)
 675             webpage = self._download_webpage(
 676                 page_url, user_id,
 677                 note='Downloading page %d/%d' % (idx + 1, page_count))
 678             video_list = self._search_regex(
 679                 r'(?s)<ol class="js-browse_list[^"]+"[^>]*>(.*?)</ol>',
 680                 webpage, 'video content')
 681             paths = re.findall(
 682                 r'<li[^>]*>\s*<a\s+href="([^"]+)"', video_list)
 683             for path in paths:
 684                 yield {
 685                     '_type': 'url',
 686                     'url': compat_urlparse.urljoin(page_url, path),
 687                 }
 688
 689         pl = InAdvancePagedList(_get_page, page_count, PAGE_SIZE)
 690
 691         return {
 692             '_type': 'playlist',
 693             'id': 'user%s_likes' % user_id,
 694             'title': title,
 695             'description': description,
 696             'entries': pl,
 697         }