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

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