_ Git - youtube-dl/blob - youtube_dl/extractor/comedycentral.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from .mtv import MTVIE, _media_xml_tag
   5 from ..utils import (
   6     compat_str,
   7     compat_urllib_parse,
   8
   9     ExtractorError,
  10     unified_strdate,
  11 )
  12
  13
  14 class ComedyCentralIE(MTVIE):
  15     _VALID_URL = r'https?://(?:www\.)?comedycentral\.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
  16     _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
  17
  18     _TEST = {
  19         u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
  20         u'md5': u'4167875aae411f903b751a21f357f1ee',
  21         u'info_dict': {
  22             u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
  23             u'ext': u'mp4',
  24             u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
  25             u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
  26         },
  27     }
  28     # Overwrite MTVIE properties we don't want
  29     _TESTS = []
  30
  31     def _get_thumbnail_url(self, uri, itemdoc):
  32         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
  33         return itemdoc.find(search_path).attrib['url']
  34
  35     def _real_extract(self, url):
  36         mobj = re.match(self._VALID_URL, url)
  37         title = mobj.group('title')
  38         webpage = self._download_webpage(url, title)
  39         mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
  40                                   webpage, u'mgid')
  41         return self._get_videos_info(mgid)
  42
  43
  44 class ComedyCentralShowsIE(InfoExtractor):
  45     IE_DESC = u'The Daily Show / Colbert Report'
  46     # urls can be abbreviations like :thedailyshow or :colbert
  47     # urls for episodes like:
  48     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
  49     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
  50     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
  51     _VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
  52                       |(https?://)?(www\.)?
  53                           (?P<showname>thedailyshow|colbertnation)\.com/
  54                          (full-episodes/(?P<episode>.*)|
  55                           (?P<clip>
  56                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  57                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
  58                           (?P<interview>
  59                               extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
  60                      $"""
  61     _TEST = {
  62         u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
  63         u'file': u'422212.mp4',
  64         u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
  65         u'info_dict': {
  66             u"upload_date": u"20121214",
  67             u"description": u"Kristen Stewart",
  68             u"uploader": u"thedailyshow",
  69             u"title": u"thedailyshow-kristen-stewart part 1"
  70         }
  71     }
  72
  73     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  74
  75     _video_extensions = {
  76         '3500': 'mp4',
  77         '2200': 'mp4',
  78         '1700': 'mp4',
  79         '1200': 'mp4',
  80         '750': 'mp4',
  81         '400': 'mp4',
  82     }
  83     _video_dimensions = {
  84         '3500': (1280, 720),
  85         '2200': (960, 540),
  86         '1700': (768, 432),
  87         '1200': (640, 360),
  88         '750': (512, 288),
  89         '400': (384, 216),
  90     }
  91
  92     @classmethod
  93     def suitable(cls, url):
  94         """Receives a URL and returns True if suitable for this IE."""
  95         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  96
  97     @staticmethod
  98     def _transform_rtmp_url(rtmp_video_url):
  99         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
 100         if not m:
 101             raise ExtractorError(u'Cannot transform RTMP url')
 102         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 103         return base + m.group('finalid')
 104
 105     def _real_extract(self, url):
 106         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 107         if mobj is None:
 108             raise ExtractorError(u'Invalid URL: %s' % url)
 109
 110         if mobj.group('shortname'):
 111             if mobj.group('shortname') in ('tds', 'thedailyshow'):
 112                 url = u'http://www.thedailyshow.com/full-episodes/'
 113             else:
 114                 url = u'http://www.colbertnation.com/full-episodes/'
 115             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 116             assert mobj is not None
 117
 118         if mobj.group('clip'):
 119             if mobj.group('showname') == 'thedailyshow':
 120                 epTitle = mobj.group('tdstitle')
 121             else:
 122                 epTitle = mobj.group('cntitle')
 123             dlNewest = False
 124         elif mobj.group('interview'):
 125             epTitle = mobj.group('interview_title')
 126             dlNewest = False
 127         else:
 128             dlNewest = not mobj.group('episode')
 129             if dlNewest:
 130                 epTitle = mobj.group('showname')
 131             else:
 132                 epTitle = mobj.group('episode')
 133
 134         self.report_extraction(epTitle)
 135         webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
 136         if dlNewest:
 137             url = htmlHandle.geturl()
 138             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 139             if mobj is None:
 140                 raise ExtractorError(u'Invalid redirected URL: ' + url)
 141             if mobj.group('episode') == '':
 142                 raise ExtractorError(u'Redirected URL is still not specific: ' + url)
 143             epTitle = mobj.group('episode')
 144
 145         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 146
 147         if len(mMovieParams) == 0:
 148             # The Colbert Report embeds the information in a without
 149             # a URL prefix; so extract the alternate reference
 150             # and then add the URL prefix manually.
 151
 152             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
 153             if len(altMovieParams) == 0:
 154                 raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
 155             else:
 156                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 157
 158         uri = mMovieParams[0][1]
 159         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
 160         idoc = self._download_xml(indexUrl, epTitle,
 161                                           u'Downloading show index',
 162                                           u'unable to download episode index')
 163
 164         results = []
 165
 166         itemEls = idoc.findall('.//item')
 167         for partNum,itemEl in enumerate(itemEls):
 168             mediaId = itemEl.findall('./guid')[0].text
 169             shortMediaId = mediaId.split(':')[-1]
 170             showId = mediaId.split(':')[-2].replace('.com', '')
 171             officialTitle = itemEl.findall('./title')[0].text
 172             officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
 173
 174             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
 175                         compat_urllib_parse.urlencode({'uri': mediaId}))
 176             cdoc = self._download_xml(configUrl, epTitle,
 177                                                u'Downloading configuration for %s' % shortMediaId)
 178
 179             turls = []
 180             for rendition in cdoc.findall('.//rendition'):
 181                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 182                 turls.append(finfo)
 183
 184             if len(turls) == 0:
 185                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
 186                 continue
 187
 188             formats = []
 189             for format, rtmp_video_url in turls:
 190                 w, h = self._video_dimensions.get(format, (None, None))
 191                 formats.append({
 192                     'url': self._transform_rtmp_url(rtmp_video_url),
 193                     'ext': self._video_extensions.get(format, 'mp4'),
 194                     'format_id': format,
 195                     'height': h,
 196                     'width': w,
 197                 })
 198
 199             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
 200             info = {
 201                 'id': shortMediaId,
 202                 'formats': formats,
 203                 'uploader': showId,
 204                 'upload_date': officialDate,
 205                 'title': effTitle,
 206                 'thumbnail': None,
 207                 'description': compat_str(officialTitle),
 208             }
 209
 210             # TODO: Remove when #980 has been merged
 211             info.update(info['formats'][-1])
 212
 213             results.append(info)
 214
 215         return results