yt_dlp/extractor/metacafe.py

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