_ 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 ..utils import (
   6     compat_str,
   7     compat_urllib_parse,
   8
   9     ExtractorError,
  10     unified_strdate,
  11 )
  12
  13
  14 class ComedyCentralIE(InfoExtractor):
  15     """Information extractor for The Daily Show and Colbert Report """
  16
  17     # urls can be abbreviations like :thedailyshow or :colbert
  18     # urls for episodes like:
  19     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
  20     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
  21     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
  22     _VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
  23                       |(https?://)?(www\.)?
  24                           (?P<showname>thedailyshow|colbertnation)\.com/
  25                          (full-episodes/(?P<episode>.*)|
  26                           (?P<clip>
  27                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  28                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))))
  29                      $"""
  30
  31     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  32
  33     _video_extensions = {
  34         '3500': 'mp4',
  35         '2200': 'mp4',
  36         '1700': 'mp4',
  37         '1200': 'mp4',
  38         '750': 'mp4',
  39         '400': 'mp4',
  40     }
  41     _video_dimensions = {
  42         '3500': '1280x720',
  43         '2200': '960x540',
  44         '1700': '768x432',
  45         '1200': '640x360',
  46         '750': '512x288',
  47         '400': '384x216',
  48     }
  49
  50     @classmethod
  51     def suitable(cls, url):
  52         """Receives a URL and returns True if suitable for this IE."""
  53         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  54
  55     def _print_formats(self, formats):
  56         print('Available formats:')
  57         for x in formats:
  58             print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
  59
  60
  61     def _real_extract(self, url):
  62         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  63         if mobj is None:
  64             raise ExtractorError(u'Invalid URL: %s' % url)
  65
  66         if mobj.group('shortname'):
  67             if mobj.group('shortname') in ('tds', 'thedailyshow'):
  68                 url = u'http://www.thedailyshow.com/full-episodes/'
  69             else:
  70                 url = u'http://www.colbertnation.com/full-episodes/'
  71             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  72             assert mobj is not None
  73
  74         if mobj.group('clip'):
  75             if mobj.group('showname') == 'thedailyshow':
  76                 epTitle = mobj.group('tdstitle')
  77             else:
  78                 epTitle = mobj.group('cntitle')
  79             dlNewest = False
  80         else:
  81             dlNewest = not mobj.group('episode')
  82             if dlNewest:
  83                 epTitle = mobj.group('showname')
  84             else:
  85                 epTitle = mobj.group('episode')
  86
  87         self.report_extraction(epTitle)
  88         webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
  89         if dlNewest:
  90             url = htmlHandle.geturl()
  91             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  92             if mobj is None:
  93                 raise ExtractorError(u'Invalid redirected URL: ' + url)
  94             if mobj.group('episode') == '':
  95                 raise ExtractorError(u'Redirected URL is still not specific: ' + url)
  96             epTitle = mobj.group('episode')
  97
  98         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
  99
 100         if len(mMovieParams) == 0:
 101             # The Colbert Report embeds the information in a without
 102             # a URL prefix; so extract the alternate reference
 103             # and then add the URL prefix manually.
 104
 105             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
 106             if len(altMovieParams) == 0:
 107                 raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
 108             else:
 109                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 110
 111         uri = mMovieParams[0][1]
 112         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
 113         indexXml = self._download_webpage(indexUrl, epTitle,
 114                                           u'Downloading show index',
 115                                           u'unable to download episode index')
 116
 117         results = []
 118
 119         idoc = xml.etree.ElementTree.fromstring(indexXml)
 120         itemEls = idoc.findall('.//item')
 121         for partNum,itemEl in enumerate(itemEls):
 122             mediaId = itemEl.findall('./guid')[0].text
 123             shortMediaId = mediaId.split(':')[-1]
 124             showId = mediaId.split(':')[-2].replace('.com', '')
 125             officialTitle = itemEl.findall('./title')[0].text
 126             officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
 127
 128             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
 129                         compat_urllib_parse.urlencode({'uri': mediaId}))
 130             configXml = self._download_webpage(configUrl, epTitle,
 131                                                u'Downloading configuration for %s' % shortMediaId)
 132
 133             cdoc = xml.etree.ElementTree.fromstring(configXml)
 134             turls = []
 135             for rendition in cdoc.findall('.//rendition'):
 136                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 137                 turls.append(finfo)
 138
 139             if len(turls) == 0:
 140                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
 141                 continue
 142
 143             if self._downloader.params.get('listformats', None):
 144                 self._print_formats([i[0] for i in turls])
 145                 return
 146
 147             # For now, just pick the highest bitrate
 148             format,rtmp_video_url = turls[-1]
 149
 150             # Get the format arg from the arg stream
 151             req_format = self._downloader.params.get('format', None)
 152
 153             # Select format if we can find one
 154             for f,v in turls:
 155                 if f == req_format:
 156                     format, rtmp_video_url = f, v
 157                     break
 158
 159             m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
 160             if not m:
 161                 raise ExtractorError(u'Cannot transform RTMP url')
 162             base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 163             video_url = base + m.group('finalid')
 164
 165             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
 166             info = {
 167                 'id': shortMediaId,
 168                 'url': video_url,
 169                 'uploader': showId,
 170                 'upload_date': officialDate,
 171                 'title': effTitle,
 172                 'ext': 'mp4',
 173                 'format': format,
 174                 'thumbnail': None,
 175                 'description': compat_str(officialTitle),
 176             }
 177             results.append(info)
 178
 179         return results