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

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from .mtv import MTVIE, _media_xml_tag
   6 from ..utils import (
   7     compat_str,
   8     compat_urllib_parse,
   9
  10     ExtractorError,
  11     unified_strdate,
  12 )
  13
  14
  15 class ComedyCentralIE(MTVIE):
  16     _VALID_URL = r'http://www.comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
  17     _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
  18
  19     _TEST = {
  20         u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
  21         u'md5': u'4167875aae411f903b751a21f357f1ee',
  22         u'info_dict': {
  23             u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
  24             u'ext': u'mp4',
  25             u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
  26             u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
  27         },
  28     }
  29     # Overwrite MTVIE properties we don't want
  30     _TESTS = []
  31
  32     def _get_thumbnail_url(self, uri, itemdoc):
  33         search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail'))
  34         return itemdoc.find(search_path).attrib['url']
  35
  36     def _real_extract(self, url):
  37         mobj = re.match(self._VALID_URL, url)
  38         title = mobj.group('title')
  39         webpage = self._download_webpage(url, title)
  40         mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
  41                                   webpage, u'mgid')
  42         return self._get_videos_info(mgid)
  43
  44
  45 class ComedyCentralShowsIE(InfoExtractor):
  46     IE_DESC = u'The Daily Show / Colbert Report'
  47     # urls can be abbreviations like :thedailyshow or :colbert
  48     # urls for episodes like:
  49     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
  50     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
  51     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
  52     _VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
  53                       |(https?://)?(www\.)?
  54                           (?P<showname>thedailyshow|colbertnation)\.com/
  55                          (full-episodes/(?P<episode>.*)|
  56                           (?P<clip>
  57                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  58                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
  59                           (?P<interview>
  60                               extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
  61                      $"""
  62     _TEST = {
  63         u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
  64         u'file': u'422212.mp4',
  65         u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
  66         u'info_dict': {
  67             u"upload_date": u"20121214",
  68             u"description": u"Kristen Stewart",
  69             u"uploader": u"thedailyshow",
  70             u"title": u"thedailyshow-kristen-stewart part 1"
  71         }
  72     }
  73
  74     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  75
  76     _video_extensions = {
  77         '3500': 'mp4',
  78         '2200': 'mp4',
  79         '1700': 'mp4',
  80         '1200': 'mp4',
  81         '750': 'mp4',
  82         '400': 'mp4',
  83     }
  84     _video_dimensions = {
  85         '3500': (1280, 720),
  86         '2200': (960, 540),
  87         '1700': (768, 432),
  88         '1200': (640, 360),
  89         '750': (512, 288),
  90         '400': (384, 216),
  91     }
  92
  93     @classmethod
  94     def suitable(cls, url):
  95         """Receives a URL and returns True if suitable for this IE."""
  96         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  97
  98     @staticmethod
  99     def _transform_rtmp_url(rtmp_video_url):
 100         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
 101         if not m:
 102             raise ExtractorError(u'Cannot transform RTMP url')
 103         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 104         return base + m.group('finalid')
 105
 106     def _real_extract(self, url):
 107         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 108         if mobj is None:
 109             raise ExtractorError(u'Invalid URL: %s' % url)
 110
 111         if mobj.group('shortname'):
 112             if mobj.group('shortname') in ('tds', 'thedailyshow'):
 113                 url = u'http://www.thedailyshow.com/full-episodes/'
 114             else:
 115                 url = u'http://www.colbertnation.com/full-episodes/'
 116             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 117             assert mobj is not None
 118
 119         if mobj.group('clip'):
 120             if mobj.group('showname') == 'thedailyshow':
 121                 epTitle = mobj.group('tdstitle')
 122             else:
 123                 epTitle = mobj.group('cntitle')
 124             dlNewest = False
 125         elif mobj.group('interview'):
 126             epTitle = mobj.group('interview_title')
 127             dlNewest = False
 128         else:
 129             dlNewest = not mobj.group('episode')
 130             if dlNewest:
 131                 epTitle = mobj.group('showname')
 132             else:
 133                 epTitle = mobj.group('episode')
 134
 135         self.report_extraction(epTitle)
 136         webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
 137         if dlNewest:
 138             url = htmlHandle.geturl()
 139             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 140             if mobj is None:
 141                 raise ExtractorError(u'Invalid redirected URL: ' + url)
 142             if mobj.group('episode') == '':
 143                 raise ExtractorError(u'Redirected URL is still not specific: ' + url)
 144             epTitle = mobj.group('episode')
 145
 146         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 147
 148         if len(mMovieParams) == 0:
 149             # The Colbert Report embeds the information in a without
 150             # a URL prefix; so extract the alternate reference
 151             # and then add the URL prefix manually.
 152
 153             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
 154             if len(altMovieParams) == 0:
 155                 raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
 156             else:
 157                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 158
 159         uri = mMovieParams[0][1]
 160         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
 161         indexXml = self._download_webpage(indexUrl, epTitle,
 162                                           u'Downloading show index',
 163                                           u'unable to download episode index')
 164
 165         results = []
 166
 167         idoc = xml.etree.ElementTree.fromstring(indexXml)
 168         itemEls = idoc.findall('.//item')
 169         for partNum,itemEl in enumerate(itemEls):
 170             mediaId = itemEl.findall('./guid')[0].text
 171             shortMediaId = mediaId.split(':')[-1]
 172             showId = mediaId.split(':')[-2].replace('.com', '')
 173             officialTitle = itemEl.findall('./title')[0].text
 174             officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
 175
 176             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
 177                         compat_urllib_parse.urlencode({'uri': mediaId}))
 178             configXml = self._download_webpage(configUrl, epTitle,
 179                                                u'Downloading configuration for %s' % shortMediaId)
 180
 181             cdoc = xml.etree.ElementTree.fromstring(configXml)
 182             turls = []
 183             for rendition in cdoc.findall('.//rendition'):
 184                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 185                 turls.append(finfo)
 186
 187             if len(turls) == 0:
 188                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
 189                 continue
 190
 191             formats = []
 192             for format, rtmp_video_url in turls:
 193                 w, h = self._video_dimensions.get(format, (None, None))
 194                 formats.append({
 195                     'url': self._transform_rtmp_url(rtmp_video_url),
 196                     'ext': self._video_extensions.get(format, 'mp4'),
 197                     'format_id': format,
 198                     'height': h,
 199                     'width': w,
 200                 })
 201
 202             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
 203             info = {
 204                 'id': shortMediaId,
 205                 'formats': formats,
 206                 'uploader': showId,
 207                 'upload_date': officialDate,
 208                 'title': effTitle,
 209                 'thumbnail': None,
 210                 'description': compat_str(officialTitle),
 211             }
 212
 213             # TODO: Remove when #980 has been merged
 214             info.update(info['formats'][-1])
 215
 216             results.append(info)
 217
 218         return results