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