_ 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     std_headers,
  19     unsmuggle_url,
  20     urlencode_postdata,
  21 )
  22
  23
  24 class VimeoBaseInfoExtractor(InfoExtractor):
  25     _NETRC_MACHINE = 'vimeo'
  26     _LOGIN_REQUIRED = False
  27
  28     def _login(self):
  29         (username, password) = self._get_login_info()
  30         if username is None:
  31             if self._LOGIN_REQUIRED:
  32                 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  33             return
  34         self.report_login()
  35         login_url = 'https://vimeo.com/log_in'
  36         webpage = self._download_webpage(login_url, None, False)
  37         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
  38         data = urlencode_postdata({
  39             'email': username,
  40             'password': password,
  41             'action': 'login',
  42             'service': 'vimeo',
  43             'token': token,
  44         })
  45         login_request = compat_urllib_request.Request(login_url, data)
  46         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  47         login_request.add_header('Cookie', 'xsrft=%s' % token)
  48         self._download_webpage(login_request, None, False, 'Wrong login info')
  49
  50
  51 class VimeoIE(VimeoBaseInfoExtractor, SubtitlesInfoExtractor):
  52     """Information extractor for vimeo.com."""
  53
  54     # _VALID_URL matches Vimeo URLs
  55     _VALID_URL = r'''(?x)
  56         (?P<proto>(?:https?:)?//)?
  57         (?:(?:www|(?P<player>player))\.)?
  58         vimeo(?P<pro>pro)?\.com/
  59         (?:.*?/)?
  60         (?:(?:play_redirect_hls|moogaloop\.swf)\?clip_id=)?
  61         (?:videos?/)?
  62         (?P<id>[0-9]+)
  63         /?(?:[?&].*)?(?:[#].*)?$'''
  64     IE_NAME = 'vimeo'
  65     _TESTS = [
  66         {
  67             'url': 'http://vimeo.com/56015672#at=0',
  68             'md5': '8879b6cc097e987f02484baf890129e5',
  69             'info_dict': {
  70                 'id': '56015672',
  71                 'ext': 'mp4',
  72                 "upload_date": "20121220",
  73                 "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",
  74                 "uploader_id": "user7108434",
  75                 "uploader": "Filippo Valsorda",
  76                 "title": "youtube-dl test video - \u2605 \" ' \u5e78 / \\ \u00e4 \u21ad \U0001d550",
  77             },
  78         },
  79         {
  80             'url': 'http://vimeopro.com/openstreetmapus/state-of-the-map-us-2013/video/68093876',
  81             'file': '68093876.mp4',
  82             'md5': '3b5ca6aa22b60dfeeadf50b72e44ed82',
  83             'note': 'Vimeo Pro video (#1197)',
  84             'info_dict': {
  85                 'uploader_id': 'openstreetmapus',
  86                 'uploader': 'OpenStreetMap US',
  87                 'title': 'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
  88             },
  89         },
  90         {
  91             'url': 'http://player.vimeo.com/video/54469442',
  92             'file': '54469442.mp4',
  93             'md5': '619b811a4417aa4abe78dc653becf511',
  94             'note': 'Videos that embed the url in the player page',
  95             'info_dict': {
  96                 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software',
  97                 'uploader': 'The BLN & Business of Software',
  98                 'uploader_id': 'theblnbusinessofsoftware',
  99             },
 100         },
 101         {
 102             'url': 'http://vimeo.com/68375962',
 103             'file': '68375962.mp4',
 104             'md5': 'aaf896bdb7ddd6476df50007a0ac0ae7',
 105             'note': 'Video protected with password',
 106             'info_dict': {
 107                 'title': 'youtube-dl password protected test video',
 108                 'upload_date': '20130614',
 109                 'uploader_id': 'user18948128',
 110                 'uploader': 'Jaime Marquínez Ferrándiz',
 111             },
 112             'params': {
 113                 'videopassword': 'youtube-dl',
 114             },
 115         },
 116         {
 117             'url': 'http://vimeo.com/76979871',
 118             'md5': '3363dd6ffebe3784d56f4132317fd446',
 119             'note': 'Video with subtitles',
 120             'info_dict': {
 121                 'id': '76979871',
 122                 'ext': 'mp4',
 123                 'title': 'The New Vimeo Player (You Know, For Videos)',
 124                 'description': 'md5:2ec900bf97c3f389378a96aee11260ea',
 125                 'upload_date': '20131015',
 126                 'uploader_id': 'staff',
 127                 'uploader': 'Vimeo Staff',
 128             }
 129         },
 130     ]
 131
 132     @classmethod
 133     def suitable(cls, url):
 134         if VimeoChannelIE.suitable(url):
 135             # Otherwise channel urls like http://vimeo.com/channels/31259 would
 136             # match
 137             return False
 138         else:
 139             return super(VimeoIE, cls).suitable(url)
 140
 141     def _verify_video_password(self, url, video_id, webpage):
 142         password = self._downloader.params.get('videopassword', None)
 143         if password is None:
 144             raise ExtractorError('This video is protected by a password, use the --video-password option')
 145         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 146         data = compat_urllib_parse.urlencode({'password': password,
 147                                               'token': token})
 148         # I didn't manage to use the password with https
 149         if url.startswith('https'):
 150             pass_url = url.replace('https','http')
 151         else:
 152             pass_url = url
 153         password_request = compat_urllib_request.Request(pass_url+'/password', data)
 154         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 155         password_request.add_header('Cookie', 'xsrft=%s' % token)
 156         self._download_webpage(password_request, video_id,
 157                                'Verifying the password',
 158                                'Wrong password')
 159
 160     def _verify_player_video_password(self, url, video_id):
 161         password = self._downloader.params.get('videopassword', None)
 162         if password is None:
 163             raise ExtractorError('This video is protected by a password, use the --video-password option')
 164         data = compat_urllib_parse.urlencode({'password': password})
 165         pass_url = url + '/check-password'
 166         password_request = compat_urllib_request.Request(pass_url, data)
 167         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 168         return self._download_json(
 169             password_request, video_id,
 170             'Verifying the password',
 171             'Wrong password')
 172
 173     def _real_initialize(self):
 174         self._login()
 175
 176     def _real_extract(self, url):
 177         url, data = unsmuggle_url(url)
 178         headers = std_headers
 179         if data is not None:
 180             headers = headers.copy()
 181             headers.update(data)
 182
 183         # Extract ID from URL
 184         mobj = re.match(self._VALID_URL, url)
 185         video_id = mobj.group('id')
 186         if mobj.group('pro') or mobj.group('player'):
 187             url = 'http://player.vimeo.com/video/' + video_id
 188         else:
 189             url = 'https://vimeo.com/' + video_id
 190
 191         # Retrieve video webpage to extract further information
 192         request = compat_urllib_request.Request(url, None, headers)
 193         try:
 194             webpage = self._download_webpage(request, video_id)
 195         except ExtractorError as ee:
 196             if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
 197                 errmsg = ee.cause.read()
 198                 if b'Because of its privacy settings, this video cannot be played here' in errmsg:
 199                     raise ExtractorError(
 200                         'Cannot download embed-only video without embedding '
 201                         'URL. Please call youtube-dl with the URL of the page '
 202                         'that embeds this video.',
 203                         expected=True)
 204             raise
 205
 206         # Now we begin extracting as much information as we can from what we
 207         # retrieved. First we extract the information common to all extractors,
 208         # and latter we extract those that are Vimeo specific.
 209         self.report_extraction(video_id)
 210
 211         # Extract the config JSON
 212         try:
 213             try:
 214                 config_url = self._html_search_regex(
 215                     r' data-config-url="(.+?)"', webpage, 'config URL')
 216                 config_json = self._download_webpage(config_url, video_id)
 217                 config = json.loads(config_json)
 218             except RegexNotFoundError:
 219                 # For pro videos or player.vimeo.com urls
 220                 # We try to find out to which variable is assigned the config dic
 221                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 222                 if m_variable_name is not None:
 223                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 224                 else:
 225                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 226                 config = self._search_regex(config_re, webpage, 'info section',
 227                     flags=re.DOTALL)
 228                 config = json.loads(config)
 229         except Exception as e:
 230             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 231                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 232
 233             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 234                 self._verify_video_password(url, video_id, webpage)
 235                 return self._real_extract(url)
 236             else:
 237                 raise ExtractorError('Unable to extract info section',
 238                                      cause=e)
 239         else:
 240             if config.get('view') == 4:
 241                 config = self._verify_player_video_password(url, video_id)
 242
 243         # Extract title
 244         video_title = config["video"]["title"]
 245
 246         # Extract uploader and uploader_id
 247         video_uploader = config["video"]["owner"]["name"]
 248         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 249
 250         # Extract video thumbnail
 251         video_thumbnail = config["video"].get("thumbnail")
 252         if video_thumbnail is None:
 253             video_thumbs = config["video"].get("thumbs")
 254             if video_thumbs and isinstance(video_thumbs, dict):
 255                 _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in video_thumbs.items())[-1]
 256
 257         # Extract video description
 258         video_description = None
 259         try:
 260             video_description = get_element_by_attribute("itemprop", "description", webpage)
 261             if video_description: video_description = clean_html(video_description)
 262         except AssertionError as err:
 263             # On some pages like (http://player.vimeo.com/video/54469442) the
 264             # html tags are not closed, python 2.6 cannot handle it
 265             if err.args[0] == 'we should not get here!':
 266                 pass
 267             else:
 268                 raise
 269
 270         # Extract upload date
 271         video_upload_date = None
 272         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 273         if mobj is not None:
 274             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 275
 276         try:
 277             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 278             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 279             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 280         except RegexNotFoundError:
 281             # This info is only available in vimeo.com/{id} urls
 282             view_count = None
 283             like_count = None
 284             comment_count = None
 285
 286         # Vimeo specific: extract request signature and timestamp
 287         sig = config['request']['signature']
 288         timestamp = config['request']['timestamp']
 289
 290         # Vimeo specific: extract video codec and quality information
 291         # First consider quality, then codecs, then take everything
 292         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 293         files = {'hd': [], 'sd': [], 'other': []}
 294         config_files = config["video"].get("files") or config["request"].get("files")
 295         for codec_name, codec_extension in codecs:
 296             for quality in config_files.get(codec_name, []):
 297                 format_id = '-'.join((codec_name, quality)).lower()
 298                 key = quality if quality in files else 'other'
 299                 video_url = None
 300                 if isinstance(config_files[codec_name], dict):
 301                     file_info = config_files[codec_name][quality]
 302                     video_url = file_info.get('url')
 303                 else:
 304                     file_info = {}
 305                 if video_url is None:
 306                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 307                         %(video_id, sig, timestamp, quality, codec_name.upper())
 308
 309                 files[key].append({
 310                     'ext': codec_extension,
 311                     'url': video_url,
 312                     'format_id': format_id,
 313                     'width': file_info.get('width'),
 314                     'height': file_info.get('height'),
 315                 })
 316         formats = []
 317         for key in ('other', 'sd', 'hd'):
 318             formats += files[key]
 319         if len(formats) == 0:
 320             raise ExtractorError('No known codec found')
 321
 322         subtitles = {}
 323         text_tracks = config['request'].get('text_tracks')
 324         if text_tracks:
 325             for tt in text_tracks:
 326                 subtitles[tt['lang']] = 'http://vimeo.com' + tt['url']
 327
 328         video_subtitles = self.extract_subtitles(video_id, subtitles)
 329         if self._downloader.params.get('listsubtitles', False):
 330             self._list_available_subtitles(video_id, subtitles)
 331             return
 332
 333         return {
 334             'id': video_id,
 335             'uploader': video_uploader,
 336             'uploader_id': video_uploader_id,
 337             'upload_date': video_upload_date,
 338             'title': video_title,
 339             'thumbnail': video_thumbnail,
 340             'description': video_description,
 341             'formats': formats,
 342             'webpage_url': url,
 343             'view_count': view_count,
 344             'like_count': like_count,
 345             'comment_count': comment_count,
 346             'subtitles': video_subtitles,
 347         }
 348
 349
 350 class VimeoChannelIE(InfoExtractor):
 351     IE_NAME = 'vimeo:channel'
 352     _VALID_URL = r'(?:https?://)?vimeo\.com/channels/(?P<id>[^/]+)/?(\?.*)?$'
 353     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 354     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 355
 356     def _page_url(self, base_url, pagenum):
 357         return '%s/videos/page:%d/' % (base_url, pagenum)
 358
 359     def _extract_list_title(self, webpage):
 360         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 361
 362     def _extract_videos(self, list_id, base_url):
 363         video_ids = []
 364         for pagenum in itertools.count(1):
 365             webpage = self._download_webpage(
 366                 self._page_url(base_url, pagenum) ,list_id,
 367                 'Downloading page %s' % pagenum)
 368             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 369             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 370                 break
 371
 372         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 373                    for video_id in video_ids]
 374         return {'_type': 'playlist',
 375                 'id': list_id,
 376                 'title': self._extract_list_title(webpage),
 377                 'entries': entries,
 378                 }
 379
 380     def _real_extract(self, url):
 381         mobj = re.match(self._VALID_URL, url)
 382         channel_id =  mobj.group('id')
 383         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 384
 385
 386 class VimeoUserIE(VimeoChannelIE):
 387     IE_NAME = 'vimeo:user'
 388     _VALID_URL = r'(?:https?://)?vimeo\.com/(?P<name>[^/]+)(?:/videos|[#?]|$)'
 389     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 390
 391     @classmethod
 392     def suitable(cls, url):
 393         if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
 394             return False
 395         return super(VimeoUserIE, cls).suitable(url)
 396
 397     def _real_extract(self, url):
 398         mobj = re.match(self._VALID_URL, url)
 399         name = mobj.group('name')
 400         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 401
 402
 403 class VimeoAlbumIE(VimeoChannelIE):
 404     IE_NAME = 'vimeo:album'
 405     _VALID_URL = r'(?:https?://)?vimeo\.com/album/(?P<id>\d+)'
 406     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 407
 408     def _page_url(self, base_url, pagenum):
 409         return '%s/page:%d/' % (base_url, pagenum)
 410
 411     def _real_extract(self, url):
 412         mobj = re.match(self._VALID_URL, url)
 413         album_id = mobj.group('id')
 414         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 415
 416
 417 class VimeoGroupsIE(VimeoAlbumIE):
 418     IE_NAME = 'vimeo:group'
 419     _VALID_URL = r'(?:https?://)?vimeo\.com/groups/(?P<name>[^/]+)'
 420
 421     def _extract_list_title(self, webpage):
 422         return self._og_search_title(webpage)
 423
 424     def _real_extract(self, url):
 425         mobj = re.match(self._VALID_URL, url)
 426         name = mobj.group('name')
 427         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 428
 429
 430 class VimeoReviewIE(InfoExtractor):
 431     IE_NAME = 'vimeo:review'
 432     IE_DESC = 'Review pages on vimeo'
 433     _VALID_URL = r'(?:https?://)?vimeo\.com/[^/]+/review/(?P<id>[^/]+)'
 434     _TEST = {
 435         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 436         'file': '75524534.mp4',
 437         'md5': 'c507a72f780cacc12b2248bb4006d253',
 438         'info_dict': {
 439             'title': "DICK HARDWICK 'Comedian'",
 440             'uploader': 'Richard Hardwick',
 441         }
 442     }
 443
 444     def _real_extract(self, url):
 445         mobj = re.match(self._VALID_URL, url)
 446         video_id = mobj.group('id')
 447         player_url = 'https://player.vimeo.com/player/' + video_id
 448         return self.url_result(player_url, 'Vimeo', video_id)
 449
 450
 451 class VimeoWatchLaterIE(VimeoBaseInfoExtractor, VimeoChannelIE):
 452     IE_NAME = 'vimeo:watchlater'
 453     IE_DESC = 'Vimeo watch later list, "vimeowatchlater" keyword (requires authentication)'
 454     _VALID_URL = r'https?://vimeo\.com/home/watchlater|:vimeowatchlater'
 455     _LOGIN_REQUIRED = True
 456     _TITLE_RE = r'href="/home/watchlater".*?>(.*?)<'
 457
 458     def _real_initialize(self):
 459         self._login()
 460
 461     def _page_url(self, base_url, pagenum):
 462         url = '%s/page:%d/' % (base_url, pagenum)
 463         request = compat_urllib_request.Request(url)
 464         # Set the header to get a partial html page with the ids,
 465         # the normal page doesn't contain them.
 466         request.add_header('X-Requested-With', 'XMLHttpRequest')
 467         return request
 468
 469     def _real_extract(self, url):
 470         return self._extract_videos('watchlater', 'https://vimeo.com/home/watchlater')