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