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_unquote,
   9     compat_urllib_parse_urlencode,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14     int_or_none,
  15     get_element_by_attribute,
  16     mimetype2ext,
  17 )
  18
  19
  20 class MetacafeIE(InfoExtractor):
  21     _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/(?P<video_id>[^/]+)/(?P<display_id>[^/?#]+)'
  22     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
  23     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
  24     IE_NAME = 'metacafe'
  25     _TESTS = [
  26         # Youtube video
  27         {
  28             'add_ie': ['Youtube'],
  29             'url': 'http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/',
  30             'info_dict': {
  31                 'id': '_aUehQsCQtM',
  32                 'ext': 'mp4',
  33                 'upload_date': '20090102',
  34                 'title': 'The Electric Company | "Short I" | PBS KIDS GO!',
  35                 'description': 'md5:2439a8ef6d5a70e380c22f5ad323e5a8',
  36                 'uploader': 'PBS',
  37                 'uploader_id': 'PBS'
  38             }
  39         },
  40         # Normal metacafe video
  41         {
  42             'url': 'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
  43             'md5': '6e0bca200eaad2552e6915ed6fd4d9ad',
  44             'info_dict': {
  45                 'id': '11121940',
  46                 'ext': 'mp4',
  47                 'title': 'News: Stuff You Won\'t Do with Your PlayStation 4',
  48                 'uploader': 'ign',
  49                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
  50             },
  51             'skip': 'Page is temporarily unavailable.',
  52         },
  53         # metacafe video with family filter
  54         {
  55             'url': 'http://www.metacafe.com/watch/2155630/adult_art_by_david_hart_156/',
  56             'md5': 'b06082c5079bbdcde677a6291fbdf376',
  57             'info_dict': {
  58                 'id': '2155630',
  59                 'ext': 'mp4',
  60                 'title': 'Adult Art By David Hart 156',
  61                 'uploader': '63346',
  62                 'description': 'md5:9afac8fc885252201ad14563694040fc',
  63             },
  64             'params': {
  65                 'skip_download': True,
  66             },
  67         },
  68         # AnyClip video
  69         {
  70             'url': 'http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/',
  71             'info_dict': {
  72                 'id': 'an-dVVXnuY7Jh77J',
  73                 'ext': 'mp4',
  74                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
  75                 'uploader': 'AnyClip',
  76                 'description': 'md5:cbef0460d31e3807f6feb4e7a5952e5b',
  77             },
  78         },
  79         # age-restricted video
  80         {
  81             'url': 'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
  82             'md5': '98dde7c1a35d02178e8ab7560fe8bd09',
  83             'info_dict': {
  84                 'id': '5186653',
  85                 'ext': 'mp4',
  86                 'title': 'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
  87                 'uploader': 'Dwayne Pipe',
  88                 'description': 'md5:950bf4c581e2c059911fa3ffbe377e4b',
  89                 'age_limit': 18,
  90             },
  91         },
  92         # cbs video
  93         {
  94             'url': 'http://www.metacafe.com/watch/cb-8VD4r_Zws8VP/open_this_is_face_the_nation_february_9/',
  95             'info_dict': {
  96                 'id': '8VD4r_Zws8VP',
  97                 'ext': 'flv',
  98                 'title': 'Open: This is Face the Nation, February 9',
  99                 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
 100                 'duration': 96,
 101                 'uploader': 'CBSI-NEW',
 102                 'upload_date': '20140209',
 103                 'timestamp': 1391959800,
 104             },
 105             'params': {
 106                 # rtmp download
 107                 'skip_download': True,
 108             },
 109         },
 110         # Movieclips.com video
 111         {
 112             'url': 'http://www.metacafe.com/watch/mv-Wy7ZU/my_week_with_marilyn_do_you_love_me/',
 113             'info_dict': {
 114                 'id': 'mv-Wy7ZU',
 115                 'ext': 'mp4',
 116                 'title': 'My Week with Marilyn - Do You Love Me?',
 117                 '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.',
 118                 'uploader': 'movie_trailers',
 119                 'duration': 176,
 120             },
 121             'params': {
 122                 'skip_download': 'requires rtmpdump',
 123             }
 124         }
 125     ]
 126
 127     def report_disclaimer(self):
 128         self.to_screen('Retrieving disclaimer')
 129
 130     def _real_extract(self, url):
 131         # Extract id and simplified title from URL
 132         video_id, display_id = re.match(self._VALID_URL, url).groups()
 133
 134         # the video may come from an external site
 135         m_external = re.match(r'^(\w{2})-(.*)$', video_id)
 136         if m_external is not None:
 137             prefix, ext_id = m_external.groups()
 138             # Check if video comes from YouTube
 139             if prefix == 'yt':
 140                 return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
 141             # CBS videos use theplatform.com
 142             if prefix == 'cb':
 143                 return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 144
 145         headers = {
 146             # Disable family filter
 147             'Cookie': 'user=%s; ' % compat_urllib_parse_urlencode({'ffilter': False})
 148         }
 149
 150         # AnyClip videos require the flashversion cookie so that we get the link
 151         # to the mp4 file
 152         if video_id.startswith('an-'):
 153             headers['Cookie'] += 'flashVersion=0; '
 154
 155         # Retrieve video webpage to extract further information
 156         webpage = self._download_webpage(url, video_id, headers=headers)
 157
 158         error = get_element_by_attribute(
 159             'class', 'notfound-page-title', webpage)
 160         if error:
 161             raise ExtractorError(error, expected=True)
 162
 163         video_title = self._html_search_meta(
 164             ['og:title', 'twitter:title'], webpage, 'title', default=None) or self._search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
 165
 166         # Extract URL, uploader and title from webpage
 167         self.report_extraction(video_id)
 168         video_url = None
 169         mobj = re.search(r'(?m)&(?:media|video)URL=([^&]+)', webpage)
 170         if mobj is not None:
 171             mediaURL = compat_urllib_parse_unquote(mobj.group(1))
 172             video_ext = determine_ext(mediaURL)
 173
 174             # Extract gdaKey if available
 175             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
 176             if mobj is None:
 177                 video_url = mediaURL
 178             else:
 179                 gdaKey = mobj.group(1)
 180                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
 181         if video_url is None:
 182             mobj = re.search(r'<video src="([^"]+)"', webpage)
 183             if mobj:
 184                 video_url = mobj.group(1)
 185                 video_ext = 'mp4'
 186         if video_url is None:
 187             flashvars = self._search_regex(
 188                 r' name="flashvars" value="(.*?)"', webpage, 'flashvars',
 189                 default=None)
 190             if flashvars:
 191                 vardict = compat_parse_qs(flashvars)
 192                 if 'mediaData' not in vardict:
 193                     raise ExtractorError('Unable to extract media URL')
 194                 mobj = re.search(
 195                     r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
 196                 if mobj is None:
 197                     raise ExtractorError('Unable to extract media URL')
 198                 mediaURL = mobj.group('mediaURL').replace('\\/', '/')
 199                 video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
 200                 video_ext = determine_ext(video_url)
 201         if video_url is None:
 202             player_url = self._search_regex(
 203                 r"swfobject\.embedSWF\('([^']+)'",
 204                 webpage, 'config URL', default=None)
 205             if player_url:
 206                 config_url = self._search_regex(
 207                     r'config=(.+)$', player_url, 'config URL')
 208                 config_doc = self._download_xml(
 209                     config_url, video_id,
 210                     note='Downloading video config')
 211                 smil_url = config_doc.find('.//properties').attrib['smil_file']
 212                 smil_doc = self._download_xml(
 213                     smil_url, video_id,
 214                     note='Downloading SMIL document')
 215                 base_url = smil_doc.find('./head/meta').attrib['base']
 216                 video_url = []
 217                 for vn in smil_doc.findall('.//video'):
 218                     br = int(vn.attrib['system-bitrate'])
 219                     play_path = vn.attrib['src']
 220                     video_url.append({
 221                         'format_id': 'smil-%d' % br,
 222                         'url': base_url,
 223                         'play_path': play_path,
 224                         'page_url': url,
 225                         'player_url': player_url,
 226                         'ext': play_path.partition(':')[0],
 227                     })
 228         if video_url is None:
 229             flashvars = self._parse_json(self._search_regex(
 230                 r'flashvars\s*=\s*({.*});', webpage, 'flashvars',
 231                 default=None), video_id, fatal=False)
 232             if flashvars:
 233                 video_url = []
 234                 for source in flashvars.get('sources'):
 235                     source_url = source.get('src')
 236                     if not source_url:
 237                         continue
 238                     ext = mimetype2ext(source.get('type')) or determine_ext(source_url)
 239                     if ext == 'm3u8':
 240                         video_url.extend(self._extract_m3u8_formats(
 241                             source_url, video_id, 'mp4',
 242                             'm3u8_native', m3u8_id='hls', fatal=False))
 243                     else:
 244                         video_url.append({
 245                             'url': source_url,
 246                             'ext': ext,
 247                         })
 248
 249         if video_url is None:
 250             raise ExtractorError('Unsupported video type')
 251
 252         description = self._html_search_meta(
 253             ['og:description', 'twitter:description', 'description'],
 254             webpage, 'title', fatal=False)
 255         thumbnail = self._html_search_meta(
 256             ['og:image', 'twitter:image'], webpage, 'title', fatal=False)
 257         video_uploader = self._html_search_regex(
 258             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
 259             webpage, 'uploader nickname', fatal=False)
 260         duration = int_or_none(
 261             self._html_search_meta('video:duration', webpage, default=None))
 262         age_limit = (
 263             18
 264             if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage)
 265             else 0)
 266
 267         if isinstance(video_url, list):
 268             formats = video_url
 269         else:
 270             formats = [{
 271                 'url': video_url,
 272                 'ext': video_ext,
 273             }]
 274         self._sort_formats(formats)
 275
 276         return {
 277             'id': video_id,
 278             'display_id': display_id,
 279             'description': description,
 280             'uploader': video_uploader,
 281             'title': video_title,
 282             'thumbnail': thumbnail,
 283             'age_limit': age_limit,
 284             'formats': formats,
 285             'duration': duration,
 286         }