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

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