Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/metacafe.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_parse_qs,
   8     compat_urllib_parse,
   9     compat_urllib_request,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14     int_or_none,
  15 )
  16
  17
  18 class MetacafeIE(InfoExtractor):
  19     _VALID_URL = r'http://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
  20     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
  21     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
  22     IE_NAME = 'metacafe'
  23     _TESTS = [
  24         # Youtube video
  25         {
  26             'add_ie': ['Youtube'],
  27             'url': 'http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/',
  28             'info_dict': {
  29                 'id': '_aUehQsCQtM',
  30                 'ext': 'mp4',
  31                 'upload_date': '20090102',
  32                 'title': 'The Electric Company | "Short I" | PBS KIDS GO!',
  33                 'description': 'md5:2439a8ef6d5a70e380c22f5ad323e5a8',
  34                 'uploader': 'PBS',
  35                 'uploader_id': 'PBS'
  36             }
  37         },
  38         # Normal metacafe video
  39         {
  40             'url': 'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
  41             'md5': '6e0bca200eaad2552e6915ed6fd4d9ad',
  42             'info_dict': {
  43                 'id': '11121940',
  44                 'ext': 'mp4',
  45                 'title': 'News: Stuff You Won\'t Do with Your PlayStation 4',
  46                 'uploader': 'ign',
  47                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
  48             },
  49         },
  50         # AnyClip video
  51         {
  52             'url': 'http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/',
  53             'info_dict': {
  54                 'id': 'an-dVVXnuY7Jh77J',
  55                 'ext': 'mp4',
  56                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
  57                 'uploader': 'anyclip',
  58                 'description': 'md5:38c711dd98f5bb87acf973d573442e67',
  59             },
  60         },
  61         # age-restricted video
  62         {
  63             'url': 'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
  64             'md5': '98dde7c1a35d02178e8ab7560fe8bd09',
  65             'info_dict': {
  66                 'id': '5186653',
  67                 'ext': 'mp4',
  68                 'title': 'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
  69                 'uploader': 'Dwayne Pipe',
  70                 'description': 'md5:950bf4c581e2c059911fa3ffbe377e4b',
  71                 'age_limit': 18,
  72             },
  73         },
  74         # cbs video
  75         {
  76             'url': 'http://www.metacafe.com/watch/cb-8VD4r_Zws8VP/open_this_is_face_the_nation_february_9/',
  77             'info_dict': {
  78                 'id': '8VD4r_Zws8VP',
  79                 'ext': 'flv',
  80                 'title': 'Open: This is Face the Nation, February 9',
  81                 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
  82                 'duration': 96,
  83             },
  84             'params': {
  85                 # rtmp download
  86                 'skip_download': True,
  87             },
  88         },
  89         # Movieclips.com video
  90         {
  91             'url': 'http://www.metacafe.com/watch/mv-Wy7ZU/my_week_with_marilyn_do_you_love_me/',
  92             'info_dict': {
  93                 'id': 'mv-Wy7ZU',
  94                 'ext': 'mp4',
  95                 'title': 'My Week with Marilyn - Do You Love Me?',
  96                 'description': 'From the movie My Week with Marilyn - Colin (Eddie Redmayne) professes his love to Marilyn (Michelle Williams) and gets her to promise to return to set and finish the movie.',
  97                 'uploader': 'movie_trailers',
  98                 'duration': 176,
  99             },
 100             'params': {
 101                 'skip_download': 'requires rtmpdump',
 102             }
 103         }
 104     ]
 105
 106     def report_disclaimer(self):
 107         self.to_screen('Retrieving disclaimer')
 108
 109     def _real_initialize(self):
 110         # Retrieve disclaimer
 111         self.report_disclaimer()
 112         self._download_webpage(self._DISCLAIMER, None, False, 'Unable to retrieve disclaimer')
 113
 114         # Confirm age
 115         disclaimer_form = {
 116             'filters': '0',
 117             'submit': "Continue - I'm over 18",
 118         }
 119         request = compat_urllib_request.Request(self._FILTER_POST, compat_urllib_parse.urlencode(disclaimer_form))
 120         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 121         self.report_age_confirmation()
 122         self._download_webpage(request, None, False, 'Unable to confirm age')
 123
 124     def _real_extract(self, url):
 125         # Extract id and simplified title from URL
 126         mobj = re.match(self._VALID_URL, url)
 127         if mobj is None:
 128             raise ExtractorError('Invalid URL: %s' % url)
 129
 130         video_id = mobj.group(1)
 131
 132         # the video may come from an external site
 133         m_external = re.match('^(\w{2})-(.*)$', video_id)
 134         if m_external is not None:
 135             prefix, ext_id = m_external.groups()
 136             # Check if video comes from YouTube
 137             if prefix == 'yt':
 138                 return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
 139             # CBS videos use theplatform.com
 140             if prefix == 'cb':
 141                 return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 142
 143         # Retrieve video webpage to extract further information
 144         req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)
 145
 146         # AnyClip videos require the flashversion cookie so that we get the link
 147         # to the mp4 file
 148         mobj_an = re.match(r'^an-(.*?)$', video_id)
 149         if mobj_an:
 150             req.headers['Cookie'] = 'flashVersion=0;'
 151         webpage = self._download_webpage(req, video_id)
 152
 153         # Extract URL, uploader and title from webpage
 154         self.report_extraction(video_id)
 155         video_url = None
 156         mobj = re.search(r'(?m)&mediaURL=([^&]+)', webpage)
 157         if mobj is not None:
 158             mediaURL = compat_urllib_parse.unquote(mobj.group(1))
 159             video_ext = mediaURL[-3:]
 160
 161             # Extract gdaKey if available
 162             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
 163             if mobj is None:
 164                 video_url = mediaURL
 165             else:
 166                 gdaKey = mobj.group(1)
 167                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
 168         if video_url is None:
 169             mobj = re.search(r'<video src="([^"]+)"', webpage)
 170             if mobj:
 171                 video_url = mobj.group(1)
 172                 video_ext = 'mp4'
 173         if video_url is None:
 174             flashvars = self._search_regex(
 175                 r' name="flashvars" value="(.*?)"', webpage, 'flashvars',
 176                 default=None)
 177             if flashvars:
 178                 vardict = compat_parse_qs(flashvars)
 179                 if 'mediaData' not in vardict:
 180                     raise ExtractorError('Unable to extract media URL')
 181                 mobj = re.search(
 182                     r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
 183                 if mobj is None:
 184                     raise ExtractorError('Unable to extract media URL')
 185                 mediaURL = mobj.group('mediaURL').replace('\\/', '/')
 186                 video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
 187                 video_ext = determine_ext(video_url)
 188         if video_url is None:
 189             player_url = self._search_regex(
 190                 r"swfobject\.embedSWF\('([^']+)'",
 191                 webpage, 'config URL', default=None)
 192             if player_url:
 193                 config_url = self._search_regex(
 194                     r'config=(.+)$', player_url, 'config URL')
 195                 config_doc = self._download_xml(
 196                     config_url, video_id,
 197                     note='Downloading video config')
 198                 smil_url = config_doc.find('.//properties').attrib['smil_file']
 199                 smil_doc = self._download_xml(
 200                     smil_url, video_id,
 201                     note='Downloading SMIL document')
 202                 base_url = smil_doc.find('./head/meta').attrib['base']
 203                 video_url = []
 204                 for vn in smil_doc.findall('.//video'):
 205                     br = int(vn.attrib['system-bitrate'])
 206                     play_path = vn.attrib['src']
 207                     video_url.append({
 208                         'format_id': 'smil-%d' % br,
 209                         'url': base_url,
 210                         'play_path': play_path,
 211                         'page_url': url,
 212                         'player_url': player_url,
 213                         'ext': play_path.partition(':')[0],
 214                     })
 215
 216         if video_url is None:
 217             raise ExtractorError('Unsupported video type')
 218
 219         video_title = self._html_search_regex(
 220             r'(?im)<title>(.*) - Video</title>', webpage, 'title')
 221         description = self._og_search_description(webpage)
 222         thumbnail = self._og_search_thumbnail(webpage)
 223         video_uploader = self._html_search_regex(
 224             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
 225             webpage, 'uploader nickname', fatal=False)
 226         duration = int_or_none(
 227             self._html_search_meta('video:duration', webpage))
 228
 229         age_limit = (
 230             18
 231             if re.search(r'"contentRating":"restricted"', webpage)
 232             else 0)
 233
 234         if isinstance(video_url, list):
 235             formats = video_url
 236         else:
 237             formats = [{
 238                 'url': video_url,
 239                 'ext': video_ext,
 240             }]
 241
 242         self._sort_formats(formats)
 243         return {
 244             'id': video_id,
 245             'description': description,
 246             'uploader': video_uploader,
 247             'title': video_title,
 248             'thumbnail': thumbnail,
 249             'age_limit': age_limit,
 250             'formats': formats,
 251             'duration': duration,
 252         }