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