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