Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/metacafe.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_parse_qs,
   9     compat_urllib_parse,
  10     compat_urllib_parse_unquote,
  11 )
  12 from ..utils import (
  13     determine_ext,
  14     ExtractorError,
  15     int_or_none,
  16     get_element_by_attribute,
  17     mimetype2ext,
  18 )
  19
  20
  21 class MetacafeIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/(?P<video_id>[^/]+)/(?P<display_id>[^/?#]+)'
  23     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
  24     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
  25     IE_NAME = 'metacafe'
  26     _TESTS = [
  27         # Youtube video
  28         {
  29             'add_ie': ['Youtube'],
  30             'url': 'http://metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/',
  31             'info_dict': {
  32                 'id': '_aUehQsCQtM',
  33                 'ext': 'mp4',
  34                 'upload_date': '20090102',
  35                 'title': 'The Electric Company | "Short I" | PBS KIDS GO!',
  36                 'description': 'md5:2439a8ef6d5a70e380c22f5ad323e5a8',
  37                 'uploader': 'PBS',
  38                 'uploader_id': 'PBS'
  39             }
  40         },
  41         # Normal metacafe video
  42         {
  43             'url': 'http://www.metacafe.com/watch/11121940/news_stuff_you_wont_do_with_your_playstation_4/',
  44             'md5': '6e0bca200eaad2552e6915ed6fd4d9ad',
  45             'info_dict': {
  46                 'id': '11121940',
  47                 'ext': 'mp4',
  48                 'title': 'News: Stuff You Won\'t Do with Your PlayStation 4',
  49                 'uploader': 'ign',
  50                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
  51             },
  52             'skip': 'Page is temporarily unavailable.',
  53         },
  54         # metacafe video with family filter
  55         {
  56             'url': 'http://www.metacafe.com/watch/2155630/adult_art_by_david_hart_156/',
  57             'md5': 'b06082c5079bbdcde677a6291fbdf376',
  58             'info_dict': {
  59                 'id': '2155630',
  60                 'ext': 'mp4',
  61                 'title': 'Adult Art By David Hart 156',
  62                 'uploader': '63346',
  63                 'description': 'md5:9afac8fc885252201ad14563694040fc',
  64             },
  65             'params': {
  66                 'skip_download': True,
  67             },
  68         },
  69         # AnyClip video
  70         {
  71             'url': 'http://www.metacafe.com/watch/an-dVVXnuY7Jh77J/the_andromeda_strain_1971_stop_the_bomb_part_3/',
  72             'info_dict': {
  73                 'id': 'an-dVVXnuY7Jh77J',
  74                 'ext': 'mp4',
  75                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
  76                 'uploader': 'AnyClip',
  77                 'description': 'md5:cbef0460d31e3807f6feb4e7a5952e5b',
  78             },
  79         },
  80         # age-restricted video
  81         {
  82             'url': 'http://www.metacafe.com/watch/5186653/bbc_internal_christmas_tape_79_uncensored_outtakes_etc/',
  83             'md5': '98dde7c1a35d02178e8ab7560fe8bd09',
  84             'info_dict': {
  85                 'id': '5186653',
  86                 'ext': 'mp4',
  87                 'title': 'BBC INTERNAL Christmas Tape \'79 - UNCENSORED Outtakes, Etc.',
  88                 'uploader': 'Dwayne Pipe',
  89                 'description': 'md5:950bf4c581e2c059911fa3ffbe377e4b',
  90                 'age_limit': 18,
  91             },
  92         },
  93         # cbs video
  94         {
  95             'url': 'http://www.metacafe.com/watch/cb-8VD4r_Zws8VP/open_this_is_face_the_nation_february_9/',
  96             'info_dict': {
  97                 'id': '8VD4r_Zws8VP',
  98                 'ext': 'flv',
  99                 'title': 'Open: This is Face the Nation, February 9',
 100                 'description': 'md5:8a9ceec26d1f7ed6eab610834cc1a476',
 101                 'duration': 96,
 102                 'uploader': 'CBSI-NEW',
 103                 'upload_date': '20140209',
 104                 'timestamp': 1391959800,
 105             },
 106             'params': {
 107                 # rtmp download
 108                 'skip_download': True,
 109             },
 110         },
 111         # Movieclips.com video
 112         {
 113             'url': 'http://www.metacafe.com/watch/mv-Wy7ZU/my_week_with_marilyn_do_you_love_me/',
 114             'info_dict': {
 115                 'id': 'mv-Wy7ZU',
 116                 'ext': 'mp4',
 117                 'title': 'My Week with Marilyn - Do You Love Me?',
 118                 '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.',
 119                 'uploader': 'movie_trailers',
 120                 'duration': 176,
 121             },
 122             'params': {
 123                 'skip_download': 'requires rtmpdump',
 124             }
 125         }
 126     ]
 127
 128     def report_disclaimer(self):
 129         self.to_screen('Retrieving disclaimer')
 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(r'^(\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         headers = {
 147             # Disable family filter
 148             'Cookie': 'user=%s; ' % compat_urllib_parse.quote(json.dumps({'ffilter': False}))
 149         }
 150
 151         # AnyClip videos require the flashversion cookie so that we get the link
 152         # to the mp4 file
 153         if video_id.startswith('an-'):
 154             headers['Cookie'] += 'flashVersion=0; '
 155
 156         # Retrieve video webpage to extract further information
 157         webpage = self._download_webpage(url, video_id, headers=headers)
 158
 159         error = get_element_by_attribute(
 160             'class', 'notfound-page-title', webpage)
 161         if error:
 162             raise ExtractorError(error, expected=True)
 163
 164         video_title = self._html_search_meta(
 165             ['og:title', 'twitter:title'], webpage, 'title', default=None) or self._search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
 166
 167         # Extract URL, uploader and title from webpage
 168         self.report_extraction(video_id)
 169         video_url = None
 170         mobj = re.search(r'(?m)&(?:media|video)URL=([^&]+)', webpage)
 171         if mobj is not None:
 172             mediaURL = compat_urllib_parse_unquote(mobj.group(1))
 173             video_ext = determine_ext(mediaURL)
 174
 175             # Extract gdaKey if available
 176             mobj = re.search(r'(?m)&gdaKey=(.*?)&', webpage)
 177             if mobj is None:
 178                 video_url = mediaURL
 179             else:
 180                 gdaKey = mobj.group(1)
 181                 video_url = '%s?__gda__=%s' % (mediaURL, gdaKey)
 182         if video_url is None:
 183             mobj = re.search(r'<video src="([^"]+)"', webpage)
 184             if mobj:
 185                 video_url = mobj.group(1)
 186                 video_ext = 'mp4'
 187         if video_url is None:
 188             flashvars = self._search_regex(
 189                 r' name="flashvars" value="(.*?)"', webpage, 'flashvars',
 190                 default=None)
 191             if flashvars:
 192                 vardict = compat_parse_qs(flashvars)
 193                 if 'mediaData' not in vardict:
 194                     raise ExtractorError('Unable to extract media URL')
 195                 mobj = re.search(
 196                     r'"mediaURL":"(?P<mediaURL>http.*?)",(.*?)"key":"(?P<key>.*?)"', vardict['mediaData'][0])
 197                 if mobj is None:
 198                     raise ExtractorError('Unable to extract media URL')
 199                 mediaURL = mobj.group('mediaURL').replace('\\/', '/')
 200                 video_url = '%s?__gda__=%s' % (mediaURL, mobj.group('key'))
 201                 video_ext = determine_ext(video_url)
 202         if video_url is None:
 203             player_url = self._search_regex(
 204                 r"swfobject\.embedSWF\('([^']+)'",
 205                 webpage, 'config URL', default=None)
 206             if player_url:
 207                 config_url = self._search_regex(
 208                     r'config=(.+)$', player_url, 'config URL')
 209                 config_doc = self._download_xml(
 210                     config_url, video_id,
 211                     note='Downloading video config')
 212                 smil_url = config_doc.find('.//properties').attrib['smil_file']
 213                 smil_doc = self._download_xml(
 214                     smil_url, video_id,
 215                     note='Downloading SMIL document')
 216                 base_url = smil_doc.find('./head/meta').attrib['base']
 217                 video_url = []
 218                 for vn in smil_doc.findall('.//video'):
 219                     br = int(vn.attrib['system-bitrate'])
 220                     play_path = vn.attrib['src']
 221                     video_url.append({
 222                         'format_id': 'smil-%d' % br,
 223                         'url': base_url,
 224                         'play_path': play_path,
 225                         'page_url': url,
 226                         'player_url': player_url,
 227                         'ext': play_path.partition(':')[0],
 228                     })
 229         if video_url is None:
 230             flashvars = self._parse_json(self._search_regex(
 231                 r'flashvars\s*=\s*({.*});', webpage, 'flashvars',
 232                 default=None), video_id, fatal=False)
 233             if flashvars:
 234                 video_url = []
 235                 for source in flashvars.get('sources'):
 236                     source_url = source.get('src')
 237                     if not source_url:
 238                         continue
 239                     ext = mimetype2ext(source.get('type')) or determine_ext(source_url)
 240                     if ext == 'm3u8':
 241                         video_url.extend(self._extract_m3u8_formats(
 242                             source_url, video_id, 'mp4',
 243                             'm3u8_native', m3u8_id='hls', fatal=False))
 244                     else:
 245                         video_url.append({
 246                             'url': source_url,
 247                             'ext': ext,
 248                         })
 249
 250         if video_url is None:
 251             raise ExtractorError('Unsupported video type')
 252
 253         description = self._html_search_meta(
 254             ['og:description', 'twitter:description', 'description'],
 255             webpage, 'title', fatal=False)
 256         thumbnail = self._html_search_meta(
 257             ['og:image', 'twitter:image'], webpage, 'title', fatal=False)
 258         video_uploader = self._html_search_regex(
 259             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
 260             webpage, 'uploader nickname', fatal=False)
 261         duration = int_or_none(
 262             self._html_search_meta('video:duration', webpage, default=None))
 263         age_limit = (
 264             18
 265             if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage)
 266             else 0)
 267
 268         if isinstance(video_url, list):
 269             formats = video_url
 270         else:
 271             formats = [{
 272                 'url': video_url,
 273                 'ext': video_ext,
 274             }]
 275         self._sort_formats(formats)
 276
 277         return {
 278             'id': video_id,
 279             'display_id': display_id,
 280             'description': description,
 281             'uploader': video_uploader,
 282             'title': video_title,
 283             'thumbnail': thumbnail,
 284             'age_limit': age_limit,
 285             'formats': formats,
 286             'duration': duration,
 287         }