_ 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 ..utils import (
  10     compat_urllib_parse,
  11     compat_urllib_request,
  12
  13     clean_html,
  14     get_element_by_attribute,
  15     ExtractorError,
  16     RegexNotFoundError,
  17     std_headers,
  18     unsmuggle_url,
  19 )
  20
  21
  22 class VimeoIE(InfoExtractor):
  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
  89     def _login(self):
  90         (username, password) = self._get_login_info()
  91         if username is None:
  92             return
  93         self.report_login()
  94         login_url = 'https://vimeo.com/log_in'
  95         webpage = self._download_webpage(login_url, None, False)
  96         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
  97         data = compat_urllib_parse.urlencode({'email': username,
  98                                               'password': password,
  99                                               'action': 'login',
 100                                               'service': 'vimeo',
 101                                               'token': token,
 102                                               })
 103         login_request = compat_urllib_request.Request(login_url, data)
 104         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 105         login_request.add_header('Cookie', 'xsrft=%s' % token)
 106         self._download_webpage(login_request, None, False, 'Wrong login info')
 107
 108     def _verify_video_password(self, url, video_id, webpage):
 109         password = self._downloader.params.get('videopassword', None)
 110         if password is None:
 111             raise ExtractorError('This video is protected by a password, use the --video-password option')
 112         token = self._search_regex(r'xsrft: \'(.*?)\'', webpage, 'login token')
 113         data = compat_urllib_parse.urlencode({'password': password,
 114                                               'token': token})
 115         # I didn't manage to use the password with https
 116         if url.startswith('https'):
 117             pass_url = url.replace('https','http')
 118         else:
 119             pass_url = url
 120         password_request = compat_urllib_request.Request(pass_url+'/password', data)
 121         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 122         password_request.add_header('Cookie', 'xsrft=%s' % token)
 123         self._download_webpage(password_request, video_id,
 124                                'Verifying the password',
 125                                'Wrong password')
 126
 127     def _verify_player_video_password(self, url, video_id):
 128         password = self._downloader.params.get('videopassword', None)
 129         if password is None:
 130             raise ExtractorError('This video is protected by a password, use the --video-password option')
 131         data = compat_urllib_parse.urlencode({'password': password})
 132         pass_url = url + '/check-password'
 133         password_request = compat_urllib_request.Request(pass_url, data)
 134         password_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 135         return self._download_json(
 136             password_request, video_id,
 137             'Verifying the password',
 138             'Wrong password')
 139
 140     def _real_initialize(self):
 141         self._login()
 142
 143     def _real_extract(self, url):
 144         url, data = unsmuggle_url(url)
 145         headers = std_headers
 146         if data is not None:
 147             headers = headers.copy()
 148             headers.update(data)
 149
 150         # Extract ID from URL
 151         mobj = re.match(self._VALID_URL, url)
 152         if mobj is None:
 153             raise ExtractorError('Invalid URL: %s' % url)
 154
 155         video_id = mobj.group('id')
 156         if mobj.group('pro') or mobj.group('player'):
 157             url = 'http://player.vimeo.com/video/' + video_id
 158         else:
 159             url = 'https://vimeo.com/' + video_id
 160
 161         # Retrieve video webpage to extract further information
 162         request = compat_urllib_request.Request(url, None, headers)
 163         webpage = self._download_webpage(request, video_id)
 164
 165         # Now we begin extracting as much information as we can from what we
 166         # retrieved. First we extract the information common to all extractors,
 167         # and latter we extract those that are Vimeo specific.
 168         self.report_extraction(video_id)
 169
 170         # Extract the config JSON
 171         try:
 172             try:
 173                 config_url = self._html_search_regex(
 174                     r' data-config-url="(.+?)"', webpage, 'config URL')
 175                 config_json = self._download_webpage(config_url, video_id)
 176                 config = json.loads(config_json)
 177             except RegexNotFoundError:
 178                 # For pro videos or player.vimeo.com urls
 179                 # We try to find out to which variable is assigned the config dic
 180                 m_variable_name = re.search('(\w)\.video\.id', webpage)
 181                 if m_variable_name is not None:
 182                     config_re = r'%s=({.+?});' % re.escape(m_variable_name.group(1))
 183                 else:
 184                     config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});']
 185                 config = self._search_regex(config_re, webpage, 'info section',
 186                     flags=re.DOTALL)
 187                 config = json.loads(config)
 188         except Exception as e:
 189             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
 190                 raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option')
 191
 192             if re.search('<form[^>]+?id="pw_form"', webpage) is not None:
 193                 self._verify_video_password(url, video_id, webpage)
 194                 return self._real_extract(url)
 195             else:
 196                 raise ExtractorError('Unable to extract info section',
 197                                      cause=e)
 198         else:
 199             if config.get('view') == 4:
 200                 config = self._verify_player_video_password(url, video_id)
 201
 202         # Extract title
 203         video_title = config["video"]["title"]
 204
 205         # Extract uploader and uploader_id
 206         video_uploader = config["video"]["owner"]["name"]
 207         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 208
 209         # Extract video thumbnail
 210         video_thumbnail = config["video"].get("thumbnail")
 211         if video_thumbnail is None:
 212             _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1]
 213
 214         # Extract video description
 215         video_description = None
 216         try:
 217             video_description = get_element_by_attribute("itemprop", "description", webpage)
 218             if video_description: video_description = clean_html(video_description)
 219         except AssertionError as err:
 220             # On some pages like (http://player.vimeo.com/video/54469442) the
 221             # html tags are not closed, python 2.6 cannot handle it
 222             if err.args[0] == 'we should not get here!':
 223                 pass
 224             else:
 225                 raise
 226
 227         # Extract upload date
 228         video_upload_date = None
 229         mobj = re.search(r'<meta itemprop="dateCreated" content="(\d{4})-(\d{2})-(\d{2})T', webpage)
 230         if mobj is not None:
 231             video_upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3)
 232
 233         try:
 234             view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count'))
 235             like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count'))
 236             comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count'))
 237         except RegexNotFoundError:
 238             # This info is only available in vimeo.com/{id} urls
 239             view_count = None
 240             like_count = None
 241             comment_count = None
 242
 243         # Vimeo specific: extract request signature and timestamp
 244         sig = config['request']['signature']
 245         timestamp = config['request']['timestamp']
 246
 247         # Vimeo specific: extract video codec and quality information
 248         # First consider quality, then codecs, then take everything
 249         codecs = [('vp6', 'flv'), ('vp8', 'flv'), ('h264', 'mp4')]
 250         files = {'hd': [], 'sd': [], 'other': []}
 251         config_files = config["video"].get("files") or config["request"].get("files")
 252         for codec_name, codec_extension in codecs:
 253             for quality in config_files.get(codec_name, []):
 254                 format_id = '-'.join((codec_name, quality)).lower()
 255                 key = quality if quality in files else 'other'
 256                 video_url = None
 257                 if isinstance(config_files[codec_name], dict):
 258                     file_info = config_files[codec_name][quality]
 259                     video_url = file_info.get('url')
 260                 else:
 261                     file_info = {}
 262                 if video_url is None:
 263                     video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
 264                         %(video_id, sig, timestamp, quality, codec_name.upper())
 265
 266                 files[key].append({
 267                     'ext': codec_extension,
 268                     'url': video_url,
 269                     'format_id': format_id,
 270                     'width': file_info.get('width'),
 271                     'height': file_info.get('height'),
 272                 })
 273         formats = []
 274         for key in ('other', 'sd', 'hd'):
 275             formats += files[key]
 276         if len(formats) == 0:
 277             raise ExtractorError('No known codec found')
 278
 279         return {
 280             'id':       video_id,
 281             'uploader': video_uploader,
 282             'uploader_id': video_uploader_id,
 283             'upload_date':  video_upload_date,
 284             'title':    video_title,
 285             'thumbnail':    video_thumbnail,
 286             'description':  video_description,
 287             'formats': formats,
 288             'webpage_url': url,
 289             'view_count': view_count,
 290             'like_count': like_count,
 291             'comment_count': comment_count,
 292         }
 293
 294
 295 class VimeoChannelIE(InfoExtractor):
 296     IE_NAME = 'vimeo:channel'
 297     _VALID_URL = r'(?:https?://)?vimeo.\com/channels/(?P<id>[^/]+)'
 298     _MORE_PAGES_INDICATOR = r'<a.+?rel="next"'
 299     _TITLE_RE = r'<link rel="alternate"[^>]+?title="(.*?)"'
 300
 301     def _page_url(self, base_url, pagenum):
 302         return '%s/videos/page:%d/' % (base_url, pagenum)
 303
 304     def _extract_list_title(self, webpage):
 305         return self._html_search_regex(self._TITLE_RE, webpage, 'list title')
 306
 307     def _extract_videos(self, list_id, base_url):
 308         video_ids = []
 309         for pagenum in itertools.count(1):
 310             webpage = self._download_webpage(
 311                 self._page_url(base_url, pagenum) ,list_id,
 312                 'Downloading page %s' % pagenum)
 313             video_ids.extend(re.findall(r'id="clip_(\d+?)"', webpage))
 314             if re.search(self._MORE_PAGES_INDICATOR, webpage, re.DOTALL) is None:
 315                 break
 316
 317         entries = [self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
 318                    for video_id in video_ids]
 319         return {'_type': 'playlist',
 320                 'id': list_id,
 321                 'title': self._extract_list_title(webpage),
 322                 'entries': entries,
 323                 }
 324
 325     def _real_extract(self, url):
 326         mobj = re.match(self._VALID_URL, url)
 327         channel_id =  mobj.group('id')
 328         return self._extract_videos(channel_id, 'http://vimeo.com/channels/%s' % channel_id)
 329
 330
 331 class VimeoUserIE(VimeoChannelIE):
 332     IE_NAME = 'vimeo:user'
 333     _VALID_URL = r'(?:https?://)?vimeo.\com/(?P<name>[^/]+)(?:/videos|[#?]|$)'
 334     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
 335
 336     @classmethod
 337     def suitable(cls, url):
 338         if VimeoChannelIE.suitable(url) or VimeoIE.suitable(url) or VimeoAlbumIE.suitable(url) or VimeoGroupsIE.suitable(url):
 339             return False
 340         return super(VimeoUserIE, cls).suitable(url)
 341
 342     def _real_extract(self, url):
 343         mobj = re.match(self._VALID_URL, url)
 344         name = mobj.group('name')
 345         return self._extract_videos(name, 'http://vimeo.com/%s' % name)
 346
 347
 348 class VimeoAlbumIE(VimeoChannelIE):
 349     IE_NAME = 'vimeo:album'
 350     _VALID_URL = r'(?:https?://)?vimeo.\com/album/(?P<id>\d+)'
 351     _TITLE_RE = r'<header id="page_header">\n\s*<h1>(.*?)</h1>'
 352
 353     def _page_url(self, base_url, pagenum):
 354         return '%s/page:%d/' % (base_url, pagenum)
 355
 356     def _real_extract(self, url):
 357         mobj = re.match(self._VALID_URL, url)
 358         album_id = mobj.group('id')
 359         return self._extract_videos(album_id, 'http://vimeo.com/album/%s' % album_id)
 360
 361
 362 class VimeoGroupsIE(VimeoAlbumIE):
 363     IE_NAME = 'vimeo:group'
 364     _VALID_URL = r'(?:https?://)?vimeo.\com/groups/(?P<name>[^/]+)'
 365
 366     def _extract_list_title(self, webpage):
 367         return self._og_search_title(webpage)
 368
 369     def _real_extract(self, url):
 370         mobj = re.match(self._VALID_URL, url)
 371         name = mobj.group('name')
 372         return self._extract_videos(name, 'http://vimeo.com/groups/%s' % name)
 373
 374
 375 class VimeoReviewIE(InfoExtractor):
 376     IE_NAME = 'vimeo:review'
 377     IE_DESC = 'Review pages on vimeo'
 378     _VALID_URL = r'(?:https?://)?vimeo.\com/[^/]+/review/(?P<id>[^/]+)'
 379     _TEST = {
 380         'url': 'https://vimeo.com/user21297594/review/75524534/3c257a1b5d',
 381         'file': '75524534.mp4',
 382         'md5': 'c507a72f780cacc12b2248bb4006d253',
 383         'info_dict': {
 384             'title': "DICK HARDWICK 'Comedian'",
 385             'uploader': 'Richard Hardwick',
 386         }
 387     }
 388
 389     def _real_extract(self, url):
 390         mobj = re.match(self._VALID_URL, url)
 391         video_id = mobj.group('id')
 392         player_url = 'https://player.vimeo.com/player/' + video_id
 393         return self.url_result(player_url, 'Vimeo', video_id)