_ 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     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': '1280x720',
  55         '2200': '960x540',
  56         '1700': '768x432',
  57         '1200': '640x360',
  58         '750': '512x288',
  59         '400': '384x216',
  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     def _print_formats(self, formats):
  68         print('Available formats:')
  69         for x in formats:
  70             print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
  71
  72
  73     def _real_extract(self, url):
  74         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  75         if mobj is None:
  76             raise ExtractorError(u'Invalid URL: %s' % url)
  77
  78         if mobj.group('shortname'):
  79             if mobj.group('shortname') in ('tds', 'thedailyshow'):
  80                 url = u'http://www.thedailyshow.com/full-episodes/'
  81             else:
  82                 url = u'http://www.colbertnation.com/full-episodes/'
  83             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  84             assert mobj is not None
  85
  86         if mobj.group('clip'):
  87             if mobj.group('showname') == 'thedailyshow':
  88                 epTitle = mobj.group('tdstitle')
  89             else:
  90                 epTitle = mobj.group('cntitle')
  91             dlNewest = False
  92         elif mobj.group('interview'):
  93             epTitle = mobj.group('interview_title')
  94             dlNewest = False
  95         else:
  96             dlNewest = not mobj.group('episode')
  97             if dlNewest:
  98                 epTitle = mobj.group('showname')
  99             else:
 100                 epTitle = mobj.group('episode')
 101
 102         self.report_extraction(epTitle)
 103         webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
 104         if dlNewest:
 105             url = htmlHandle.geturl()
 106             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 107             if mobj is None:
 108                 raise ExtractorError(u'Invalid redirected URL: ' + url)
 109             if mobj.group('episode') == '':
 110                 raise ExtractorError(u'Redirected URL is still not specific: ' + url)
 111             epTitle = mobj.group('episode')
 112
 113         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 114
 115         if len(mMovieParams) == 0:
 116             # The Colbert Report embeds the information in a without
 117             # a URL prefix; so extract the alternate reference
 118             # and then add the URL prefix manually.
 119
 120             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
 121             if len(altMovieParams) == 0:
 122                 raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
 123             else:
 124                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 125
 126         uri = mMovieParams[0][1]
 127         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
 128         indexXml = self._download_webpage(indexUrl, epTitle,
 129                                           u'Downloading show index',
 130                                           u'unable to download episode index')
 131
 132         results = []
 133
 134         idoc = xml.etree.ElementTree.fromstring(indexXml)
 135         itemEls = idoc.findall('.//item')
 136         for partNum,itemEl in enumerate(itemEls):
 137             mediaId = itemEl.findall('./guid')[0].text
 138             shortMediaId = mediaId.split(':')[-1]
 139             showId = mediaId.split(':')[-2].replace('.com', '')
 140             officialTitle = itemEl.findall('./title')[0].text
 141             officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
 142
 143             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
 144                         compat_urllib_parse.urlencode({'uri': mediaId}))
 145             configXml = self._download_webpage(configUrl, epTitle,
 146                                                u'Downloading configuration for %s' % shortMediaId)
 147
 148             cdoc = xml.etree.ElementTree.fromstring(configXml)
 149             turls = []
 150             for rendition in cdoc.findall('.//rendition'):
 151                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 152                 turls.append(finfo)
 153
 154             if len(turls) == 0:
 155                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
 156                 continue
 157
 158             if self._downloader.params.get('listformats', None):
 159                 self._print_formats([i[0] for i in turls])
 160                 return
 161
 162             # For now, just pick the highest bitrate
 163             format,rtmp_video_url = turls[-1]
 164
 165             # Get the format arg from the arg stream
 166             req_format = self._downloader.params.get('format', None)
 167
 168             # Select format if we can find one
 169             for f,v in turls:
 170                 if f == req_format:
 171                     format, rtmp_video_url = f, v
 172                     break
 173
 174             m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
 175             if not m:
 176                 raise ExtractorError(u'Cannot transform RTMP url')
 177             base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 178             video_url = base + m.group('finalid')
 179
 180             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
 181             info = {
 182                 'id': shortMediaId,
 183                 'url': video_url,
 184                 'uploader': showId,
 185                 'upload_date': officialDate,
 186                 'title': effTitle,
 187                 'ext': 'mp4',
 188                 'format': format,
 189                 'thumbnail': None,
 190                 'description': compat_str(officialTitle),
 191             }
 192             results.append(info)
 193
 194         return results