_ 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 .subtitles import SubtitlesInfoExtractor
  10 from ..utils import (
  11     compat_HTTPError,
  12     compat_urllib_parse,
  13     compat_urllib_request,
  14     clean_html,
  15     get_element_by_attribute,
  16     ExtractorError,
  17     RegexNotFoundError,
  18     smuggle_url,
  19     std_headers,
  20     unsmuggle_url,
  21     urlencode_postdata,
  22     int_or_none,
  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, SubtitlesInfoExtractor):
  54     """Information extractor for vimeo.com."""
  55
  56     # _VALID_URL matches Vimeo URLs
  57     _VALID_URL = r'''(?x)
  58         (?P<proto>(?: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                 'duration': 1595,
  94             },
  95         },
  96         {
  97             'url': 'http://player.vimeo.com/video/54469442',
  98             'md5': '619b811a4417aa4abe78dc653becf511',
  99             'note': 'Videos that embed the url in the player page',
 100             'info_dict': {
 101                 'id': '54469442',
 102                 'ext': 'mp4',
 103                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 104                 'uploader': 'The BLN & Business of Software',
 105                 'uploader_id': 'theblnbusinessofsoftware',
 106                 'duration': 3610,
 107             },
 108         },
 109         {
 110             'url': 'http://vimeo.com/68375962',
 111             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 112             'note': 'Video protected with password',
 113             'info_dict': {
 114                 'id': '68375962',
 115                 'ext': 'mp4',
 116                 'title': 'youtube-dl password protected test video',
 117                 'upload_date': '20130614',
 118                 'uploader_id': 'user18948128',
 119                 'uploader': 'Jaime Marquínez Ferrándiz',
 120                 'duration': 10,
 121             },
 122             'params': {
 123                 'videopassword': 'youtube-dl',
 124             },
 125         },
 126         {
 127             'url': 'http://vimeo.com/channels/keypeele/75629013',
 128             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 129             'note': 'Video is freely available via original URL '
 130                     'and protected with password when accessed via http://vimeo.com/75629013',
 131             'info_dict': {
 132                 'id': '75629013',
 133                 'ext': 'mp4',
 134                 'title': 'Key & Peele: Terrorist Interrogation',
 135                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 136                 'uploader_id': 'atencio',
 137                 'uploader': 'Peter Atencio',
 138                 'duration': 187,
 139             },
 140         },
 141         {
 142             'url': 'http://vimeo.com/76979871',
 143             'md5': '3363dd6ffebe3784d56f4132317fd446',
 144             'note': 'Video with subtitles',
 145             'info_dict': {
 146                 'id': '76979871',
 147                 'ext': 'mp4',
 148                 'title': 'The New Vimeo Player (You Know, For Videos)',
 149                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 150                 'upload_date': '20131015',
 151                 'uploader_id': 'staff',
 152                 'uploader': 'Vimeo Staff',
 153                 'duration': 62,
 154             }
 155         },
 156     ]
 157
 158     def _verify_video_password(self, url, video_id, webpage):
 159         password = self._downloader.params.get('videopassword', None)
 160         if password is None:
 161             raise ExtractorError('This video is protected by a password, use the --video-password option')
 162         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 163         data = compat_urllib_parse.urlencode({
 164             'password': password,
 165             'token': token,
 166         })
 167         # I didn't manage to use the password with https
 168         if url.startswith('https'):
 169             pass_url = url.replace('https', 'http')
 170         else:
 171             pass_url = url
 172         password_request = compat_urllib_request.Request(pass_url + '/password', data)
 173         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 174         password_request.add_header('Cookie', 'xsrft=%s' % token)
 175         self._download_webpage(password_request, video_id,
 176                                'Verifying the password',
 177                                'Wrong password')
 178
 179     def _verify_player_video_password(self, url, video_id):
 180         password = self._downloader.params.get('videopassword', None)
 181         if password is None:
 182             raise ExtractorError('This video is protected by a password, use the --video-password option')
 183         data = compat_urllib_parse.urlencode({'password': password})
 184         pass_url = url + '/check-password'
 185         password_request = compat_urllib_request.Request(pass_url, data)
 186         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 187         return self._download_json(
 188             password_request, video_id,
 189             'Verifying the password',
 190             'Wrong password')
 191
 192     def _real_initialize(self):
 193         self._login()
 194
 195     def _real_extract(self, url):
 196         url, data = unsmuggle_url(url)
 197         headers = std_headers
 198         if data is not None:
 199             headers = headers.copy()
 200             headers.update(data)
 201         if 'Referer' not in headers:
 202             headers['Referer'] = url
 203
 204         # Extract ID from URL
 205         mobj = re.match(self._VALID_URL, url)
 206         video_id = mobj.group('id')
 207         if mobj.group('pro') or mobj.group('player'):
 208             url = 'http://player.vimeo.com/video/' + video_id
 209
 210         # Retrieve video webpage to extract further information
 211         request = compat_urllib_request.Request(url, None, headers)
 212         try:
 213             webpage = self._download_webpage(request, video_id)
 214         except ExtractorError as ee:
 215             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 216                 errmsg = ee.cause.read()
 217                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 218                     raise ExtractorError(
 219                         'Cannot download embed-only video without embedding '
 220                         'URL. Please call youtube-dl with the URL of the page '
 221                         'that embeds this video.',
 222                         expected=True)
 223             raise
 224
 225         # Now we begin extracting as much information as we can from what we
 226         # retrieved. First we extract the information common to all extractors,
 227         # and latter we extract those that are Vimeo specific.
 228         self.report_extraction(video_id)
 229
 230         # Extract the config JSON
 231         try:
 232             try:
 233                 config_url = self._html_search_regex(
 234                     r' data-config-url="(.+?)"', webpage, 'config URL')
 235                 config_json = self._download_webpage(config_url, video_id)
 236                 config = json.loads(config_json)
 237             except RegexNotFoundError:
 238                 # For pro videos or player.vimeo.com urls
 239                 # We try to find out to which variable is assigned the config dic
 240                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 241                 if m_variable_name is not None:
 242                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 243                 else:
 244                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 245                 config = self._search_regex(config_re, webpage, 'info section',
 246                     flags=re.DOTALL)
 247                 config = json.loads(config)
 248         except Exception as e:
 249             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 250                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 251
 252             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 253                 self._verify_video_password(url, video_id, webpage)
 254                 return self._real_extract(url)
 255             else:
 256                 raise ExtractorError('Unable to extract info section',
 257                                      cause=e)
 258         else:
 259             if config.get('view') == 4:
 260                 config = self._verify_player_video_password(url, video_id)
 261
 262         # Extract title
 263         video_title = config["video"]["title"]
 264
 265         # Extract uploader and uploader_id
 266         video_uploader = config["video"]["owner"]["name"]
 267         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 268
 269         # Extract video thumbnail
 270         video_thumbnail = config["video"].get("thumbnail")
 271         if video_thumbnail is None:
 272             video_thumbs = config["video"].get("thumbs")
 273             if video_thumbs and isinstance(video_thumbs, dict):
 274                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 275
 276         # Extract video description
 277         video_description = None
 278         try:
 279             video_description = get_element_by_attribute("class", "description_wrapper", webpage)
 280             if video_description:
 281                 video_description = clean_html(video_description)
 282         except AssertionError as err:
 283             # On some pages like (http://player.vimeo.com/video/54469442) the
 284             # html tags are not closed, python 2.6 cannot handle it
 285             if err.args[0] == 'we should not get here!':
 286                 pass
 287             else:
 288                 raise
 289
 290         # Extract video duration
 291         video_duration = int_or_none(config["video"].get("duration"))
 292
 293         # Extract upload date
 294         video_upload_date = None
 295         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 296         if mobj is not None:
 297             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 298
 299         try:
 300             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 301             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 302             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 303         except RegexNotFoundError:
 304             # This info is only available in vimeo.com/{id} urls
 305             view_count = None
 306             like_count = None
 307             comment_count = None
 308
 309         # Vimeo specific: extract request signature and timestamp
 310         sig = config['request']['signature']
 311         timestamp = config['request']['timestamp']
 312
 313         # Vimeo specific: extract video codec and quality information
 314         # First consider quality, then codecs, then take everything
 315         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 316         files = {'hd': [], 'sd': [], 'other': []}
 317         config_files = config["video"].get("files") or config["request"].get("files")
 318         for codec_name, codec_extension in codecs:
 319             for quality in config_files.get(codec_name, []):
 320                 format_id = '-'.join((codec_name, quality)).lower()
 321                 key = quality if quality in files else 'other'
 322                 video_url = None
 323                 if isinstance(config_files[codec_name], dict):
 324                     file_info = config_files[codec_name][quality]
 325                     video_url = file_info.get('url')
 326                 else:
 327                     file_info = {}
 328                 if video_url is None:
 329                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 330                         % (video_id, sig, timestamp, quality, codec_name.upper())
 331
 332                 files[key].append({
 333                     'ext': codec_extension,
 334                     'url': video_url,
 335                     'format_id': format_id,
 336                     'width': file_info.get('width'),
 337                     'height': file_info.get('height'),
 338                 })
 339         formats = []
 340         for key in ('other', 'sd', 'hd'):
 341             formats += files[key]
 342         if len(formats) == 0:
 343             raise ExtractorError('No known codec found')
 344
 345         subtitles = {}
 346         text_tracks = config['request'].get('text_tracks')
 347         if text_tracks:
 348             for tt in text_tracks:
 349                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
 350
 351         video_subtitles = self.extract_subtitles(video_id, subtitles)
 352         if self._downloader.params.get('listsubtitles', False):
 353             self._list_available_subtitles(video_id, subtitles)
 354             return
 355
 356         return {
 357             'id': video_id,
 358             'uploader': video_uploader,
 359             'uploader_id': video_uploader_id,
 360             'upload_date': video_upload_date,
 361             'title': video_title,
 362             'thumbnail': video_thumbnail,
 363             'description': video_description,
 364             'duration': video_duration,
 365             'formats': formats,
 366             'webpage_url': url,
 367             'view_count': view_count,
 368             'like_count': like_count,
 369             'comment_count': comment_count,
 370             'subtitles': video_subtitles,
 371         }
 372
 373
 374 class VimeoChannelIE(InfoExtractor):
 375     IE_NAME = 'vimeo:channel'
 376     _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 377     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 378     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 379     _TESTS = [{
 380         'url': 'http://vimeo.com/channels/tributes',
 381         'info_dict': {
 382             'title': 'Vimeo Tributes',
 383         },
 384         'playlist_mincount': 25,
 385     }]
 386
 387     def _page_url(self, base_url, pagenum):
 388         return '%s/videos/page:%d/' % (base_url, pagenum)
 389
 390     def _extract_list_title(self, webpage):
 391         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 392
 393     def _extract_videos(self, list_id, base_url):
 394         video_ids = []
 395         for pagenum in itertools.count(1):
 396             webpage = self._download_webpage(
 397                 self._page_url(base_url, pagenum), list_id,
 398                 'Downloading page %s' % pagenum)
 399             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 400             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 401                 break
 402
 403         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 404                    for video_id in video_ids]
 405         return {'_type': 'playlist',
 406                 'id': list_id,
 407                 'title': self._extract_list_title(webpage),
 408                 'entries': entries,
 409                 }
 410
 411     def _real_extract(self, url):
 412         mobj = re.match(self._VALID_URL, url)
 413         channel_id = mobj.group('id')
 414         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 415
 416
 417 class VimeoUserIE(VimeoChannelIE):
 418     IE_NAME = 'vimeo:user'
 419     _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 420     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 421     _TESTS = [{
 422         'url': 'http://vimeo.com/nkistudio/videos',
 423         'info_dict': {
 424             'title': 'Nki',
 425         },
 426         'playlist_mincount': 66,
 427     }]
 428
 429     def _real_extract(self, url):
 430         mobj = re.match(self._VALID_URL, url)
 431         name = mobj.group('name')
 432         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 433
 434
 435 class VimeoAlbumIE(VimeoChannelIE):
 436     IE_NAME = 'vimeo:album'
 437     _VALID_URL = r'https?://vimeo\.com/album/(?P<id>\d+)'
 438     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 439     _TESTS = [{
 440         'url': 'http://vimeo.com/album/2632481',
 441         'info_dict': {
 442             'title': 'Staff Favorites: November 2013',
 443         },
 444         'playlist_mincount': 13,
 445     }]
 446
 447     def _page_url(self, base_url, pagenum):
 448         return '%s/page:%d/' % (base_url, pagenum)
 449
 450     def _real_extract(self, url):
 451         mobj = re.match(self._VALID_URL, url)
 452         album_id = mobj.group('id')
 453         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 454
 455
 456 class VimeoGroupsIE(VimeoAlbumIE):
 457     IE_NAME = 'vimeo:group'
 458     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 459     _TESTS = [{
 460         'url': 'http://vimeo.com/groups/rolexawards',
 461         'info_dict': {
 462             'title': 'Rolex Awards for Enterprise',
 463         },
 464         'playlist_mincount': 73,
 465     }]
 466
 467     def _extract_list_title(self, webpage):
 468         return self._og_search_title(webpage)
 469
 470     def _real_extract(self, url):
 471         mobj = re.match(self._VALID_URL, url)
 472         name = mobj.group('name')
 473         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 474
 475
 476 class VimeoReviewIE(InfoExtractor):
 477     IE_NAME = 'vimeo:review'
 478     IE_DESC = 'Review pages on vimeo'
 479     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 480     _TESTS = [{
 481         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 482         'file': '75524534.mp4',
 483         'md5': 'c507a72f780cacc12b2248bb4006d253',
 484         'info_dict': {
 485             'title': "DICK HARDWICK 'Comedian'",
 486             'uploader': 'Richard Hardwick',
 487         }
 488     }, {
 489         'note': 'video player needs Referer',
 490         'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
 491         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 492         'info_dict': {
 493             'id': '91613211',
 494             'ext': 'mp4',
 495             'title': 'Death by dogma versus assembling agile - Sander Hoogendoorn',
 496             'uploader': 'DevWeek Events',
 497             'duration': 2773,
 498             'thumbnail': 're:^https?://.*\.jpg$',
 499         }
 500     }]
 501
 502     def _real_extract(self, url):
 503         mobj = re.match(self._VALID_URL, url)
 504         video_id = mobj.group('id')
 505         player_url = 'https://player.vimeo.com/player/' + video_id
 506         return self.url_result(player_url, 'Vimeo', video_id)
 507
 508
 509 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 510     IE_NAME = 'vimeo:watchlater'
 511     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 512     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 513     _LOGIN_REQUIRED = True
 514     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 515     _TESTS = [{
 516         'url': 'http://vimeo.com/home/watchlater',
 517         'only_matching': True,
 518     }]
 519
 520     def _real_initialize(self):
 521         self._login()
 522
 523     def _page_url(self, base_url, pagenum):
 524         url = '%s/page:%d/' % (base_url, pagenum)
 525         request = compat_urllib_request.Request(url)
 526         # Set the header to get a partial html page with the ids,
 527         # the normal page doesn't contain them.
 528         request.add_header('X-Requested-With', 'XMLHttpRequest')
 529         return request
 530
 531     def _real_extract(self, url):
 532         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')
 533
 534
 535 class VimeoLikesIE(InfoExtractor):
 536     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/user(?P<id>[0-9]+)/likes(?:$|[?#])'
 537     IE_NAME = 'vimeo:likes'
 538     IE_DESC = 'Vimeo user likes'
 539     _TEST = {
 540         'url': 'https://vimeo.com/user20132939/likes',
 541         'playlist_mincount': 4,
 542         'add_ies': ['Generic'],
 543         "info_dict": {
 544             "description": "Videos Philipp Hagemeister likes on Vimeo.",
 545             "title": "Vimeo / Philipp Hagemeister's likes",
 546         },
 547         'params': {
 548             'extract_flat': False,
 549         },
 550     }
 551
 552     def _real_extract(self, url):
 553         user_id = self._match_id(url)
 554         rss_url = '%s//vimeo.com/user%s/likes/rss' % (
 555             self.http_scheme(), user_id)
 556         surl = smuggle_url(rss_url, {
 557             'force_videoid': '%s_likes' % user_id,
 558             'to_generic': True,
 559         })
 560
 561         return {
 562             '_type': 'url',
 563             'url': surl,
 564         }