youtube_dl/extractor/comedycentral.py

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