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     _TEST = {
  31         u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
  32         u'file': u'422212.mp4',
  33         u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
  34         u'info_dict': {
  35             u"upload_date": u"20121214",
  36             u"description": u"Kristen Stewart",
  37             u"uploader": u"thedailyshow",
  38             u"title": u"thedailyshow-kristen-stewart part 1"
  39         }
  40     }
  41
  42     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  43
  44     _video_extensions = {
  45         '3500': 'mp4',
  46         '2200': 'mp4',
  47         '1700': 'mp4',
  48         '1200': 'mp4',
  49         '750': 'mp4',
  50         '400': 'mp4',
  51     }
  52     _video_dimensions = {
  53         '3500': '1280x720',
  54         '2200': '960x540',
  55         '1700': '768x432',
  56         '1200': '640x360',
  57         '750': '512x288',
  58         '400': '384x216',
  59     }
  60
  61     @classmethod
  62     def suitable(cls, url):
  63         """Receives a URL and returns True if suitable for this IE."""
  64         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
  65
  66     def _print_formats(self, formats):
  67         print('Available formats:')
  68         for x in formats:
  69             print('%s\t:\t%s\t[%s]' %(x, self._video_extensions.get(x, 'mp4'), self._video_dimensions.get(x, '???')))
  70
  71
  72     def _real_extract(self, url):
  73         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  74         if mobj is None:
  75             raise ExtractorError(u'Invalid URL: %s' % url)
  76
  77         if mobj.group('shortname'):
  78             if mobj.group('shortname') in ('tds', 'thedailyshow'):
  79                 url = u'http://www.thedailyshow.com/full-episodes/'
  80             else:
  81                 url = u'http://www.colbertnation.com/full-episodes/'
  82             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  83             assert mobj is not None
  84
  85         if mobj.group('clip'):
  86             if mobj.group('showname') == 'thedailyshow':
  87                 epTitle = mobj.group('tdstitle')
  88             else:
  89                 epTitle = mobj.group('cntitle')
  90             dlNewest = False
  91         else:
  92             dlNewest = not mobj.group('episode')
  93             if dlNewest:
  94                 epTitle = mobj.group('showname')
  95             else:
  96                 epTitle = mobj.group('episode')
  97
  98         self.report_extraction(epTitle)
  99         webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
 100         if dlNewest:
 101             url = htmlHandle.geturl()
 102             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 103             if mobj is None:
 104                 raise ExtractorError(u'Invalid redirected URL: ' + url)
 105             if mobj.group('episode') == '':
 106                 raise ExtractorError(u'Redirected URL is still not specific: ' + url)
 107             epTitle = mobj.group('episode')
 108
 109         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 110
 111         if len(mMovieParams) == 0:
 112             # The Colbert Report embeds the information in a without
 113             # a URL prefix; so extract the alternate reference
 114             # and then add the URL prefix manually.
 115
 116             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
 117             if len(altMovieParams) == 0:
 118                 raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
 119             else:
 120                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 121
 122         uri = mMovieParams[0][1]
 123         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
 124         indexXml = self._download_webpage(indexUrl, epTitle,
 125                                           u'Downloading show index',
 126                                           u'unable to download episode index')
 127
 128         results = []
 129
 130         idoc = xml.etree.ElementTree.fromstring(indexXml)
 131         itemEls = idoc.findall('.//item')
 132         for partNum,itemEl in enumerate(itemEls):
 133             mediaId = itemEl.findall('./guid')[0].text
 134             shortMediaId = mediaId.split(':')[-1]
 135             showId = mediaId.split(':')[-2].replace('.com', '')
 136             officialTitle = itemEl.findall('./title')[0].text
 137             officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
 138
 139             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
 140                         compat_urllib_parse.urlencode({'uri': mediaId}))
 141             configXml = self._download_webpage(configUrl, epTitle,
 142                                                u'Downloading configuration for %s' % shortMediaId)
 143
 144             cdoc = xml.etree.ElementTree.fromstring(configXml)
 145             turls = []
 146             for rendition in cdoc.findall('.//rendition'):
 147                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 148                 turls.append(finfo)
 149
 150             if len(turls) == 0:
 151                 self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
 152                 continue
 153
 154             if self._downloader.params.get('listformats', None):
 155                 self._print_formats([i[0] for i in turls])
 156                 return
 157
 158             # For now, just pick the highest bitrate
 159             format,rtmp_video_url = turls[-1]
 160
 161             # Get the format arg from the arg stream
 162             req_format = self._downloader.params.get('format', None)
 163
 164             # Select format if we can find one
 165             for f,v in turls:
 166                 if f == req_format:
 167                     format, rtmp_video_url = f, v
 168                     break
 169
 170             m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
 171             if not m:
 172                 raise ExtractorError(u'Cannot transform RTMP url')
 173             base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 174             video_url = base + m.group('finalid')
 175
 176             effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
 177             info = {
 178                 'id': shortMediaId,
 179                 'url': video_url,
 180                 'uploader': showId,
 181                 'upload_date': officialDate,
 182                 'title': effTitle,
 183                 'ext': 'mp4',
 184                 'format': format,
 185                 'thumbnail': None,
 186                 'description': compat_str(officialTitle),
 187             }
 188             results.append(info)
 189
 190         return results