_ Git - youtube-dl/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_unquote,
   9 )
  10 from ..utils import (
  11     determine_ext,
  12     ExtractorError,
  13     int_or_none,
  14     sanitized_Request,
  15     urlencode_postdata,
  16 )
  17
  18
  19 class MetacafeIE(InfoExtractor):
  20     _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
  21     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
  22     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
  23     IE_NAME = 'metacafe'
  24     _TESTS = [
  25         # Youtube video
  26         {
  27             'add_ie': ['Youtube'],
  28             'url': 'http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/',
  29             'info_dict': {
  30                 'id': '_aUehQsCQtM',
  31                 'ext': 'mp4',
  32                 'upload_date': '20090102',
  33                 'title': 'The Electric Company | "Short I" | PBS KIDS GO!',
  34                 'description': 'md5:2439a8ef6d5a70e380c22f5ad323e5a8',
  35                 'uploader': 'PBS',
  36                 'uploader_id': 'PBS'
  37             }
  38         },
  39         # Normal metacafe video
  40         {
  41             'url': 'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
  42             'md5': '6e0bca200eaad2552e6915ed6fd4d9ad',
  43             'info_dict': {
  44                 'id': '11121940',
  45                 'ext': 'mp4',
  46                 'title': 'News: Stuff You Won\'t Do with Your PlayStation 4',
  47                 'uploader': 'ign',
  48                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
  49             },
  50         },
  51         # AnyClip video
  52         {
  53             'url': 'http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/',
  54             'info_dict': {
  55                 'id': 'an-dVVXnuY7Jh77J',
  56                 'ext': 'mp4',
  57                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
  58                 'uploader': 'anyclip',
  59                 'description': 'md5:38c711dd98f5bb87acf973d573442e67',
  60             },
  61         },
  62         # age-restricted video
  63         {
  64             'url': 'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
  65             'md5': '98dde7c1a35d02178e8ab7560fe8bd09',
  66             'info_dict': {
  67                 'id': '5186653',
  68                 'ext': 'mp4',
  69                 'title': 'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
  70                 'uploader': 'Dwayne Pipe',
  71                 'description': 'md5:950bf4c581e2c059911fa3ffbe377e4b',
  72                 'age_limit': 18,
  73             },
  74         },
  75         # cbs video
  76         {
  77             'url': 'http://www.metacafe.com/watch/cb-8VD4r_Zws8VP/open_this_is_face_the_nation_february_9/',
  78             'info_dict': {
  79                 'id': '8VD4r_Zws8VP',
  80                 'ext': 'flv',
  81                 'title': 'Open: This is Face the Nation, February 9',
  82                 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
  83                 'duration': 96,
  84                 'uploader': 'CBSI-NEW',
  85                 'upload_date': '20140209',
  86                 'timestamp': 1391959800,
  87             },
  88             'params': {
  89                 # rtmp download
  90                 'skip_download': True,
  91             },
  92         },
  93         # Movieclips.com video
  94         {
  95             'url': 'http://www.metacafe.com/watch/mv-Wy7ZU/my_week_with_marilyn_do_you_love_me/',
  96             'info_dict': {
  97                 'id': 'mv-Wy7ZU',
  98                 'ext': 'mp4',
  99                 'title': 'My Week with Marilyn - Do You Love Me?',
 100                 '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.',
 101                 'uploader': 'movie_trailers',
 102                 'duration': 176,
 103             },
 104             'params': {
 105                 'skip_download': 'requires rtmpdump',
 106             }
 107         }
 108     ]
 109
 110     def report_disclaimer(self):
 111         self.to_screen('Retrieving disclaimer')
 112
 113     def _real_initialize(self):
 114         # Retrieve disclaimer
 115         self.report_disclaimer()
 116         self._download_webpage(self._DISCLAIMER, None, False, 'Unable to retrieve disclaimer')
 117
 118         # Confirm age
 119         disclaimer_form = {
 120             'filters': '0',
 121             'submit': "Continue - I'm over 18",
 122         }
 123         request = sanitized_Request(self._FILTER_POST, urlencode_postdata(disclaimer_form))
 124         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 125         self.report_age_confirmation()
 126         self._download_webpage(request, None, False, 'Unable to confirm age')
 127
 128     def _real_extract(self, url):
 129         # Extract id and simplified title from URL
 130         mobj = re.match(self._VALID_URL, url)
 131         if mobj is None:
 132             raise ExtractorError('Invalid URL: %s' % url)
 133
 134         video_id = mobj.group(1)
 135
 136         # the video may come from an external site
 137         m_external = re.match('^(\w{2})-(.*)$', video_id)
 138         if m_external is not None:
 139             prefix, ext_id = m_external.groups()
 140             # Check if video comes from YouTube
 141             if prefix == 'yt':
 142                 return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
 143             # CBS videos use theplatform.com
 144             if prefix == 'cb':
 145                 return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 146
 147         # Retrieve video webpage to extract further information
 148         req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id)
 149
 150         # AnyClip videos require the flashversion cookie so that we get the link
 151         # to the mp4 file
 152         mobj_an = re.match(r'^an-(.*?)$', video_id)
 153         if mobj_an:
 154             req.headers['Cookie'] = 'flashVersion=0;'
 155         webpage = self._download_webpage(req, video_id)
 156
 157         # Extract URL, uploader and title from webpage
 158         self.report_extraction(video_id)
 159         video_url = None
 160         mobj = re.search(r'(?m)&(?:media|video)URL=([^&]+)', webpage)
 161         if mobj is not None:
 162             mediaURL = compat_urllib_parse_unquote(mobj.group(1))
 163             video_ext = determine_ext(mediaURL)
 164
 165             # Extract gdaKey if available
 166             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
 167             if mobj is None:
 168                 video_url = mediaURL
 169             else:
 170                 gdaKey = mobj.group(1)
 171                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
 172         if video_url is None:
 173             mobj = re.search(r'<video src="([^"]+)"', webpage)
 174             if mobj:
 175                 video_url = mobj.group(1)
 176                 video_ext = 'mp4'
 177         if video_url is None:
 178             flashvars = self._search_regex(
 179                 r' name="flashvars" value="(.*?)"', webpage, 'flashvars',
 180                 default=None)
 181             if flashvars:
 182                 vardict = compat_parse_qs(flashvars)
 183                 if 'mediaData' not in vardict:
 184                     raise ExtractorError('Unable to extract media URL')
 185                 mobj = re.search(
 186                     r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
 187                 if mobj is None:
 188                     raise ExtractorError('Unable to extract media URL')
 189                 mediaURL = mobj.group('mediaURL').replace('\\/', '/')
 190                 video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
 191                 video_ext = determine_ext(video_url)
 192         if video_url is None:
 193             player_url = self._search_regex(
 194                 r"swfobject\.embedSWF\('([^']+)'",
 195                 webpage, 'config URL', default=None)
 196             if player_url:
 197                 config_url = self._search_regex(
 198                     r'config=(.+)$', player_url, 'config URL')
 199                 config_doc = self._download_xml(
 200                     config_url, video_id,
 201                     note='Downloading video config')
 202                 smil_url = config_doc.find('.//properties').attrib['smil_file']
 203                 smil_doc = self._download_xml(
 204                     smil_url, video_id,
 205                     note='Downloading SMIL document')
 206                 base_url = smil_doc.find('./head/meta').attrib['base']
 207                 video_url = []
 208                 for vn in smil_doc.findall('.//video'):
 209                     br = int(vn.attrib['system-bitrate'])
 210                     play_path = vn.attrib['src']
 211                     video_url.append({
 212                         'format_id': 'smil-%d' % br,
 213                         'url': base_url,
 214                         'play_path': play_path,
 215                         'page_url': url,
 216                         'player_url': player_url,
 217                         'ext': play_path.partition(':')[0],
 218                     })
 219
 220         if video_url is None:
 221             raise ExtractorError('Unsupported video type')
 222
 223         video_title = self._html_search_regex(
 224             r'(?im)<title>(.*) - Video</title>', webpage, 'title')
 225         description = self._og_search_description(webpage)
 226         thumbnail = self._og_search_thumbnail(webpage)
 227         video_uploader = self._html_search_regex(
 228             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
 229             webpage, 'uploader nickname', fatal=False)
 230         duration = int_or_none(
 231             self._html_search_meta('video:duration', webpage))
 232
 233         age_limit = (
 234             18
 235             if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage)
 236             else 0)
 237
 238         if isinstance(video_url, list):
 239             formats = video_url
 240         else:
 241             formats = [{
 242                 'url': video_url,
 243                 'ext': video_ext,
 244             }]
 245
 246         self._sort_formats(formats)
 247         return {
 248             'id': video_id,
 249             'description': description,
 250             'uploader': video_uploader,
 251             'title': video_title,
 252             'thumbnail': thumbnail,
 253             'age_limit': age_limit,
 254             'formats': formats,
 255             'duration': duration,
 256         }