_ Git - youtube-dl/blob - youtube_dl/extractor/comedycentral.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from .mtv import MTVServicesInfoExtractor
   7 from ..utils import (
   8     compat_str,
   9     compat_urllib_parse,
  10     ExtractorError,
  11     float_or_none,
  12     unified_strdate,
  13 )
  14
  15
  16 class ComedyCentralIE(MTVServicesInfoExtractor):
  17     _VALID_URL = r'''(?x)https?://(?:www\.)?(comedycentral|cc)\.com/
  18         (video-clips|episodes|cc-studios|video-collections)
  19         /(?P<title>.*)'''
  20     _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
  21
  22     _TEST = {
  23         'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
  24         'md5': '4167875aae411f903b751a21f357f1ee',
  25         'info_dict': {
  26             'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
  27             'ext': 'mp4',
  28             'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
  29             'description': 'After a certain point, breastfeeding becomes c**kblocking.',
  30         },
  31     }
  32
  33
  34 class ComedyCentralShowsIE(InfoExtractor):
  35     IE_DESC = 'The Daily Show / The Colbert Report'
  36     # urls can be abbreviations like :thedailyshow or :colbert
  37     # urls for episodes like:
  38     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
  39     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
  40     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
  41     _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
  42                       |https?://(:www\.)?
  43                           (?P<showname>thedailyshow|thecolbertreport)\.(?:cc\.)?com/
  44                          (full-episodes/(?P<episode>.*)|
  45                           (?P<clip>
  46                               (?:videos/[^/]+/(?P<videotitle>[^/?#]+))
  47                               |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  48                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
  49                           (?P<interview>
  50                               extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
  51                      $'''
  52     _TEST = {
  53         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
  54         'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
  55         'info_dict': {
  56             'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
  57             'ext': 'mp4',
  58             'upload_date': '20121213',
  59             'description': 'Kristen Stewart learns to let loose in "On the Road."',
  60             'uploader': 'thedailyshow',
  61             'title': 'thedailyshow-kristen-stewart part 1',
  62         }
  63     }
  64
  65     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  66
  67     _video_extensions = {
  68         '3500': 'mp4',
  69         '2200': 'mp4',
  70         '1700': 'mp4',
  71         '1200': 'mp4',
  72         '750': 'mp4',
  73         '400': 'mp4',
  74     }
  75     _video_dimensions = {
  76         '3500': (1280, 720),
  77         '2200': (960, 540),
  78         '1700': (768, 432),
  79         '1200': (640, 360),
  80         '750': (512, 288),
  81         '400': (384, 216),
  82     }
  83
  84     @staticmethod
  85     def _transform_rtmp_url(rtmp_video_url):
  86         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
  87         if not m:
  88             raise ExtractorError('Cannot transform RTMP url')
  89         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
  90         return base + m.group('finalid')
  91
  92     def _real_extract(self, url):
  93         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  94         if mobj is None:
  95             raise ExtractorError('Invalid URL: %s' % url)
  96
  97         if mobj.group('shortname'):
  98             if mobj.group('shortname') in ('tds', 'thedailyshow'):
  99                 url = 'http://thedailyshow.cc.com/full-episodes/'
 100             else:
 101                 url = 'http://thecolbertreport.cc.com/full-episodes/'
 102             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 103             assert mobj is not None
 104
 105         if mobj.group('clip'):
 106             if mobj.group('videotitle'):
 107                 epTitle = mobj.group('videotitle')
 108             elif mobj.group('showname') == 'thedailyshow':
 109                 epTitle = mobj.group('tdstitle')
 110             else:
 111                 epTitle = mobj.group('cntitle')
 112             dlNewest = False
 113         elif mobj.group('interview'):
 114             epTitle = mobj.group('interview_title')
 115             dlNewest = False
 116         else:
 117             dlNewest = not mobj.group('episode')
 118             if dlNewest:
 119                 epTitle = mobj.group('showname')
 120             else:
 121                 epTitle = mobj.group('episode')
 122         show_name = mobj.group('showname')
 123
 124         webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
 125         if dlNewest:
 126             url = htmlHandle.geturl()
 127             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 128             if mobj is None:
 129                 raise ExtractorError('Invalid redirected URL: ' + url)
 130             if mobj.group('episode') == '':
 131                 raise ExtractorError('Redirected URL is still not specific: ' + url)
 132             epTitle = mobj.group('episode').rpartition('/')[-1]
 133
 134         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 135         if len(mMovieParams) == 0:
 136             # The Colbert Report embeds the information in a without
 137             # a URL prefix; so extract the alternate reference
 138             # and then add the URL prefix manually.
 139
 140             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
 141             if len(altMovieParams) == 0:
 142                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
 143             else:
 144                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 145
 146         uri = mMovieParams[0][1]
 147         # Correct cc.com in uri
 148         uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.cc.com', uri)
 149
 150         index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse.urlencode({'uri': uri}))
 151         idoc = self._download_xml(
 152             index_url, epTitle,
 153             'Downloading show index', 'Unable to download episode index')
 154
 155         title = idoc.find('./channel/title').text
 156         description = idoc.find('./channel/description').text
 157
 158         entries = []
 159         item_els = idoc.findall('.//item')
 160         for part_num, itemEl in enumerate(item_els):
 161             upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
 162             thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
 163
 164             content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
 165             duration = float_or_none(content.attrib.get('duration'))
 166             mediagen_url = content.attrib['url']
 167             guid = itemEl.find('.//guid').text.rpartition(':')[-1]
 168
 169             cdoc = self._download_xml(
 170                 mediagen_url, epTitle,
 171                 'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 172
 173             turls = []
 174             for rendition in cdoc.findall('.//rendition'):
 175                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 176                 turls.append(finfo)
 177
 178             formats = []
 179             for format, rtmp_video_url in turls:
 180                 w, h = self._video_dimensions.get(format, (None, None))
 181                 formats.append({
 182                     'format_id': 'vhttp-%s' % format,
 183                     'url': self._transform_rtmp_url(rtmp_video_url),
 184                     'ext': self._video_extensions.get(format, 'mp4'),
 185                     'height': h,
 186                     'width': w,
 187                 })
 188                 formats.append({
 189                     'format_id': 'rtmp-%s' % format,
 190                     'url': rtmp_video_url,
 191                     'ext': self._video_extensions.get(format, 'mp4'),
 192                     'height': h,
 193                     'width': w,
 194                 })
 195                 self._sort_formats(formats)
 196
 197             virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
 198             entries.append({
 199                 'id': guid,
 200                 'title': virtual_id,
 201                 'formats': formats,
 202                 'uploader': show_name,
 203                 'upload_date': upload_date,
 204                 'duration': duration,
 205                 'thumbnail': thumbnail,
 206                 'description': description,
 207             })
 208
 209         return {
 210             '_type': 'playlist',
 211             'entries': entries,
 212             'title': show_name + ' ' + title,
 213             'description': description,
 214         }