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     std_headers,
  19     unsmuggle_url,
  20     urlencode_postdata,
  21     int_or_none,
  22 )
  23
  24
  25 class VimeoBaseInfoExtractor(InfoExtractor):
  26     _NETRC_MACHINE = 'vimeo'
  27     _LOGIN_REQUIRED = False
  28
  29     def _login(self):
  30         (username, password) = self._get_login_info()
  31         if username is None:
  32             if self._LOGIN_REQUIRED:
  33                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  34             return
  35         self.report_login()
  36         login_url = 'https://vimeo.com/log_in'
  37         webpage = self._download_webpage(login_url, None, False)
  38         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
  39         data = urlencode_postdata({
  40             'email': username,
  41             'password': password,
  42             'action': 'login',
  43             'service': 'vimeo',
  44             'token': token,
  45         })
  46         login_request = compat_urllib_request.Request(login_url, data)
  47         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  48         login_request.add_header('Cookie', 'xsrft=%s' % token)
  49         self._download_webpage(login_request, None, False, 'Wrong login info')
  50
  51
  52 class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
  53     """Information extractor for vimeo.com."""
  54
  55     # _VALID_URL matches Vimeo URLs
  56     _VALID_URL = r'''(?x)
  57         (?P<proto>(?:https?:)?//)?
  58         (?:(?:www|(?P<player>player))\.)?
  59         vimeo(?P<pro>pro)?\.com/
  60         (?!channels/[^/?#]+/?(?:$|[?#])|album/)
  61         (?:.*?/)?
  62         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  63         (?:videos?/)?
  64         (?P<id>[0-9]+)
  65         /?(?:[?&].*)?(?:[#].*)?$'''
  66     IE_NAME = 'vimeo'
  67     _TESTS = [
  68         {
  69             'url': 'http://vimeo.com/56015672#at=0',
  70             'md5': '8879b6cc097e987f02484baf890129e5',
  71             'info_dict': {
  72                 'id': '56015672',
  73                 'ext': 'mp4',
  74                 "upload_date": "20121220",
  75                 "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",
  76                 "uploader_id": "user7108434",
  77                 "uploader": "Filippo Valsorda",
  78                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  79                 "duration": 10,
  80             },
  81         },
  82         {
  83             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  84             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  85             'note': 'Vimeo Pro video (#1197)',
  86             'info_dict': {
  87                 'id': '68093876',
  88                 'ext': 'mp4',
  89                 'uploader_id': 'openstreetmapus',
  90                 'uploader': 'OpenStreetMap US',
  91                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
  92                 'duration': 1595,
  93             },
  94         },
  95         {
  96             'url': 'http://player.vimeo.com/video/54469442',
  97             'md5': '619b811a4417aa4abe78dc653becf511',
  98             'note': 'Videos that embed the url in the player page',
  99             'info_dict': {
 100                 'id': '54469442',
 101                 'ext': 'mp4',
 102                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012',
 103                 'uploader': 'The BLN & Business of Software',
 104                 'uploader_id': 'theblnbusinessofsoftware',
 105                 'duration': 3610,
 106             },
 107         },
 108         {
 109             'url': 'http://vimeo.com/68375962',
 110             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 111             'note': 'Video protected with password',
 112             'info_dict': {
 113                 'id': '68375962',
 114                 'ext': 'mp4',
 115                 'title': 'youtube-dl password protected test video',
 116                 'upload_date': '20130614',
 117                 'uploader_id': 'user18948128',
 118                 'uploader': 'Jaime Marquínez Ferrándiz',
 119                 'duration': 10,
 120             },
 121             'params': {
 122                 'videopassword': 'youtube-dl',
 123             },
 124         },
 125         {
 126             'url': 'http://vimeo.com/channels/keypeele/75629013',
 127             'md5': '2f86a05afe9d7abc0b9126d229bbe15d',
 128             'note': 'Video is freely available via original URL '
 129                     'and protected with password when accessed via http://vimeo.com/75629013',
 130             'info_dict': {
 131                 'id': '75629013',
 132                 'ext': 'mp4',
 133                 'title': 'Key & Peele: Terrorist Interrogation',
 134                 'description': 'md5:8678b246399b070816b12313e8b4eb5c',
 135                 'uploader_id': 'atencio',
 136                 'uploader': 'Peter Atencio',
 137                 'duration': 187,
 138             },
 139         },
 140         {
 141             'url': 'http://vimeo.com/76979871',
 142             'md5': '3363dd6ffebe3784d56f4132317fd446',
 143             'note': 'Video with subtitles',
 144             'info_dict': {
 145                 'id': '76979871',
 146                 'ext': 'mp4',
 147                 'title': 'The New Vimeo Player (You Know, For Videos)',
 148                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 149                 'upload_date': '20131015',
 150                 'uploader_id': 'staff',
 151                 'uploader': 'Vimeo Staff',
 152                 'duration': 62,
 153             }
 154         },
 155     ]
 156
 157     def _verify_video_password(self, url, video_id, webpage):
 158         password = self._downloader.params.get('videopassword', None)
 159         if password is None:
 160             raise ExtractorError('This video is protected by a password, use the --video-password option')
 161         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 162         data = compat_urllib_parse.urlencode({
 163             'password': password,
 164             'token': token,
 165         })
 166         # I didn't manage to use the password with https
 167         if url.startswith('https'):
 168             pass_url = url.replace('https', 'http')
 169         else:
 170             pass_url = url
 171         password_request = compat_urllib_request.Request(pass_url + '/password', data)
 172         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 173         password_request.add_header('Cookie', 'xsrft=%s' % token)
 174         self._download_webpage(password_request, video_id,
 175                                'Verifying the password',
 176                                'Wrong password')
 177
 178     def _verify_player_video_password(self, url, video_id):
 179         password = self._downloader.params.get('videopassword', None)
 180         if password is None:
 181             raise ExtractorError('This video is protected by a password, use the --video-password option')
 182         data = compat_urllib_parse.urlencode({'password': password})
 183         pass_url = url + '/check-password'
 184         password_request = compat_urllib_request.Request(pass_url, data)
 185         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 186         return self._download_json(
 187             password_request, video_id,
 188             'Verifying the password',
 189             'Wrong password')
 190
 191     def _real_initialize(self):
 192         self._login()
 193
 194     def _real_extract(self, url):
 195         url, data = unsmuggle_url(url)
 196         headers = std_headers
 197         if data is not None:
 198             headers = headers.copy()
 199             headers.update(data)
 200         if 'Referer' not in headers:
 201             headers['Referer'] = url
 202
 203         # Extract ID from URL
 204         mobj = re.match(self._VALID_URL, url)
 205         video_id = mobj.group('id')
 206         if mobj.group('pro') or mobj.group('player'):
 207             url = 'http://player.vimeo.com/video/' + video_id
 208
 209         # Retrieve video webpage to extract further information
 210         request = compat_urllib_request.Request(url, None, headers)
 211         try:
 212             webpage = self._download_webpage(request, video_id)
 213         except ExtractorError as ee:
 214             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 215                 errmsg = ee.cause.read()
 216                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 217                     raise ExtractorError(
 218                         'Cannot download embed-only video without embedding '
 219                         'URL. Please call youtube-dl with the URL of the page '
 220                         'that embeds this video.',
 221                         expected=True)
 222             raise
 223
 224         # Now we begin extracting as much information as we can from what we
 225         # retrieved. First we extract the information common to all extractors,
 226         # and latter we extract those that are Vimeo specific.
 227         self.report_extraction(video_id)
 228
 229         # Extract the config JSON
 230         try:
 231             try:
 232                 config_url = self._html_search_regex(
 233                     r' data-config-url="(.+?)"', webpage, 'config URL')
 234                 config_json = self._download_webpage(config_url, video_id)
 235                 config = json.loads(config_json)
 236             except RegexNotFoundError:
 237                 # For pro videos or player.vimeo.com urls
 238                 # We try to find out to which variable is assigned the config dic
 239                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 240                 if m_variable_name is not None:
 241                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 242                 else:
 243                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 244                 config = self._search_regex(config_re, webpage, 'info section',
 245                     flags=re.DOTALL)
 246                 config = json.loads(config)
 247         except Exception as e:
 248             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 249                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 250
 251             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 252                 self._verify_video_password(url, video_id, webpage)
 253                 return self._real_extract(url)
 254             else:
 255                 raise ExtractorError('Unable to extract info section',
 256                                      cause=e)
 257         else:
 258             if config.get('view') == 4:
 259                 config = self._verify_player_video_password(url, video_id)
 260
 261         # Extract title
 262         video_title = config["video"]["title"]
 263
 264         # Extract uploader and uploader_id
 265         video_uploader = config["video"]["owner"]["name"]
 266         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 267
 268         # Extract video thumbnail
 269         video_thumbnail = config["video"].get("thumbnail")
 270         if video_thumbnail is None:
 271             video_thumbs = config["video"].get("thumbs")
 272             if video_thumbs and isinstance(video_thumbs, dict):
 273                 _, video_thumbnail = sorted((int(width if width.isdigit() else 0), t_url) for (width, t_url) in video_thumbs.items())[-1]
 274
 275         # Extract video description
 276         video_description = None
 277         try:
 278             video_description = get_element_by_attribute("class", "description_wrapper", webpage)
 279             if video_description:
 280                 video_description = clean_html(video_description)
 281         except AssertionError as err:
 282             # On some pages like (http://player.vimeo.com/video/54469442) the
 283             # html tags are not closed, python 2.6 cannot handle it
 284             if err.args[0] == 'we should not get here!':
 285                 pass
 286             else:
 287                 raise
 288
 289         # Extract video duration
 290         video_duration = int_or_none(config["video"].get("duration"))
 291
 292         # Extract upload date
 293         video_upload_date = None
 294         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 295         if mobj is not None:
 296             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 297
 298         try:
 299             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 300             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 301             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 302         except RegexNotFoundError:
 303             # This info is only available in vimeo.com/{id} urls
 304             view_count = None
 305             like_count = None
 306             comment_count = None
 307
 308         # Vimeo specific: extract request signature and timestamp
 309         sig = config['request']['signature']
 310         timestamp = config['request']['timestamp']
 311
 312         # Vimeo specific: extract video codec and quality information
 313         # First consider quality, then codecs, then take everything
 314         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 315         files = {'hd': [], 'sd': [], 'other': []}
 316         config_files = config["video"].get("files") or config["request"].get("files")
 317         for codec_name, codec_extension in codecs:
 318             for quality in config_files.get(codec_name, []):
 319                 format_id = '-'.join((codec_name, quality)).lower()
 320                 key = quality if quality in files else 'other'
 321                 video_url = None
 322                 if isinstance(config_files[codec_name], dict):
 323                     file_info = config_files[codec_name][quality]
 324                     video_url = file_info.get('url')
 325                 else:
 326                     file_info = {}
 327                 if video_url is None:
 328                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 329                         % (video_id, sig, timestamp, quality, codec_name.upper())
 330
 331                 files[key].append({
 332                     'ext': codec_extension,
 333                     'url': video_url,
 334                     'format_id': format_id,
 335                     'width': file_info.get('width'),
 336                     'height': file_info.get('height'),
 337                 })
 338         formats = []
 339         for key in ('other', 'sd', 'hd'):
 340             formats += files[key]
 341         if len(formats) == 0:
 342             raise ExtractorError('No known codec found')
 343
 344         subtitles = {}
 345         text_tracks = config['request'].get('text_tracks')
 346         if text_tracks:
 347             for tt in text_tracks:
 348                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
 349
 350         video_subtitles = self.extract_subtitles(video_id, subtitles)
 351         if self._downloader.params.get('listsubtitles', False):
 352             self._list_available_subtitles(video_id, subtitles)
 353             return
 354
 355         return {
 356             'id': video_id,
 357             'uploader': video_uploader,
 358             'uploader_id': video_uploader_id,
 359             'upload_date': video_upload_date,
 360             'title': video_title,
 361             'thumbnail': video_thumbnail,
 362             'description': video_description,
 363             'duration': video_duration,
 364             'formats': formats,
 365             'webpage_url': url,
 366             'view_count': view_count,
 367             'like_count': like_count,
 368             'comment_count': comment_count,
 369             'subtitles': video_subtitles,
 370         }
 371
 372
 373 class VimeoChannelIE(InfoExtractor):
 374     IE_NAME = 'vimeo:channel'
 375     _VALID_URL = r'https?://vimeo\.com/channels/(?P<id>[^/?#]+)/?(?:$|[?#])'
 376     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 377     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 378     _TESTS = [{
 379         'url': 'http://vimeo.com/channels/tributes',
 380         'info_dict': {
 381             'title': 'Vimeo Tributes',
 382         },
 383         'playlist_mincount': 25,
 384     }]
 385
 386     def _page_url(self, base_url, pagenum):
 387         return '%s/videos/page:%d/' % (base_url, pagenum)
 388
 389     def _extract_list_title(self, webpage):
 390         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 391
 392     def _extract_videos(self, list_id, base_url):
 393         video_ids = []
 394         for pagenum in itertools.count(1):
 395             webpage = self._download_webpage(
 396                 self._page_url(base_url, pagenum), list_id,
 397                 'Downloading page %s' % pagenum)
 398             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 399             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 400                 break
 401
 402         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 403                    for video_id in video_ids]
 404         return {'_type': 'playlist',
 405                 'id': list_id,
 406                 'title': self._extract_list_title(webpage),
 407                 'entries': entries,
 408                 }
 409
 410     def _real_extract(self, url):
 411         mobj = re.match(self._VALID_URL, url)
 412         channel_id = mobj.group('id')
 413         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 414
 415
 416 class VimeoUserIE(VimeoChannelIE):
 417     IE_NAME = 'vimeo:user'
 418     _VALID_URL = r'https?://vimeo\.com/(?![0-9]+(?:$|[?#/]))(?P<name>[^/]+)(?:/videos|[#?]|$)'
 419     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 420     _TESTS = [{
 421         'url': 'http://vimeo.com/nkistudio/videos',
 422         'info_dict': {
 423             'title': 'Nki',
 424         },
 425         'playlist_mincount': 66,
 426     }]
 427
 428     def _real_extract(self, url):
 429         mobj = re.match(self._VALID_URL, url)
 430         name = mobj.group('name')
 431         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 432
 433
 434 class VimeoAlbumIE(VimeoChannelIE):
 435     IE_NAME = 'vimeo:album'
 436     _VALID_URL = r'https?://vimeo\.com/album/(?P<id>\d+)'
 437     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 438     _TESTS = [{
 439         'url': 'http://vimeo.com/album/2632481',
 440         'info_dict': {
 441             'title': 'Staff Favorites: November 2013',
 442         },
 443         'playlist_mincount': 13,
 444     }]
 445
 446     def _page_url(self, base_url, pagenum):
 447         return '%s/page:%d/' % (base_url, pagenum)
 448
 449     def _real_extract(self, url):
 450         mobj = re.match(self._VALID_URL, url)
 451         album_id = mobj.group('id')
 452         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 453
 454
 455 class VimeoGroupsIE(VimeoAlbumIE):
 456     IE_NAME = 'vimeo:group'
 457     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 458     _TESTS = [{
 459         'url': 'http://vimeo.com/groups/rolexawards',
 460         'info_dict': {
 461             'title': 'Rolex Awards for Enterprise',
 462         },
 463         'playlist_mincount': 73,
 464     }]
 465
 466     def _extract_list_title(self, webpage):
 467         return self._og_search_title(webpage)
 468
 469     def _real_extract(self, url):
 470         mobj = re.match(self._VALID_URL, url)
 471         name = mobj.group('name')
 472         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 473
 474
 475 class VimeoReviewIE(InfoExtractor):
 476     IE_NAME = 'vimeo:review'
 477     IE_DESC = 'Review pages on vimeo'
 478     _VALID_URL = r'https?://vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 479     _TESTS = [{
 480         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 481         'file': '75524534.mp4',
 482         'md5': 'c507a72f780cacc12b2248bb4006d253',
 483         'info_dict': {
 484             'title': "DICK HARDWICK 'Comedian'",
 485             'uploader': 'Richard Hardwick',
 486         }
 487     }, {
 488         'note': 'video player needs Referer',
 489         'url': 'http://vimeo.com/user22258446/review/91613211/13f927e053',
 490         'md5': '6295fdab8f4bf6a002d058b2c6dce276',
 491         'info_dict': {
 492             'id': '91613211',
 493             'ext': 'mp4',
 494             'title': 'Death by dogma versus assembling agile - Sander Hoogendoorn',
 495             'uploader': 'DevWeek Events',
 496             'duration': 2773,
 497             'thumbnail': 're:^https?://.*\.jpg$',
 498         }
 499     }]
 500
 501     def _real_extract(self, url):
 502         mobj = re.match(self._VALID_URL, url)
 503         video_id = mobj.group('id')
 504         player_url = 'https://player.vimeo.com/player/' + video_id
 505         return self.url_result(player_url, 'Vimeo', video_id)
 506
 507
 508 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 509     IE_NAME = 'vimeo:watchlater'
 510     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 511     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 512     _LOGIN_REQUIRED = True
 513     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 514     _TESTS = [{
 515         'url': 'http://vimeo.com/home/watchlater',
 516         'only_matching': True,
 517     }]
 518
 519     def _real_initialize(self):
 520         self._login()
 521
 522     def _page_url(self, base_url, pagenum):
 523         url = '%s/page:%d/' % (base_url, pagenum)
 524         request = compat_urllib_request.Request(url)
 525         # Set the header to get a partial html page with the ids,
 526         # the normal page doesn't contain them.
 527         request.add_header('X-Requested-With', 'XMLHttpRequest')
 528         return request
 529
 530     def _real_extract(self, url):
 531         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')