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