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