_ 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                               (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  47                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
  48                           (?P<interview>
  49                               extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
  50                      $'''
  51     _TEST = {
  52         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
  53         'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
  54         'info_dict': {
  55             'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
  56             'ext': 'mp4',
  57             'upload_date': '20121213',
  58             'description': 'Kristen Stewart learns to let loose in "On the Road."',
  59             'uploader': 'thedailyshow',
  60             'title': 'thedailyshow-kristen-stewart part 1',
  61         }
  62     }
  63
  64     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  65
  66     _video_extensions = {
  67         '3500': 'mp4',
  68         '2200': 'mp4',
  69         '1700': 'mp4',
  70         '1200': 'mp4',
  71         '750': 'mp4',
  72         '400': 'mp4',
  73     }
  74     _video_dimensions = {
  75         '3500': (1280, 720),
  76         '2200': (960, 540),
  77         '1700': (768, 432),
  78         '1200': (640, 360),
  79         '750': (512, 288),
  80         '400': (384, 216),
  81     }
  82
  83     @staticmethod
  84     def _transform_rtmp_url(rtmp_video_url):
  85         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
  86         if not m:
  87             raise ExtractorError('Cannot transform RTMP url')
  88         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
  89         return base + m.group('finalid')
  90
  91     def _real_extract(self, url):
  92         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  93         if mobj is None:
  94             raise ExtractorError('Invalid URL: %s' % url)
  95
  96         if mobj.group('shortname'):
  97             if mobj.group('shortname') in ('tds', 'thedailyshow'):
  98                 url = 'http://thedailyshow.cc.com/full-episodes/'
  99             else:
 100                 url = 'http://thecolbertreport.cc.com/full-episodes/'
 101             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 102             assert mobj is not None
 103
 104         if mobj.group('clip'):
 105             if mobj.group('showname') == 'thedailyshow':
 106                 epTitle = mobj.group('tdstitle')
 107             else:
 108                 epTitle = mobj.group('cntitle')
 109             dlNewest = False
 110         elif mobj.group('interview'):
 111             epTitle = mobj.group('interview_title')
 112             dlNewest = False
 113         else:
 114             dlNewest = not mobj.group('episode')
 115             if dlNewest:
 116                 epTitle = mobj.group('showname')
 117             else:
 118                 epTitle = mobj.group('episode')
 119         show_name = mobj.group('showname')
 120
 121         webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
 122         if dlNewest:
 123             url = htmlHandle.geturl()
 124             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 125             if mobj is None:
 126                 raise ExtractorError('Invalid redirected URL: ' + url)
 127             if mobj.group('episode') == '':
 128                 raise ExtractorError('Redirected URL is still not specific: ' + url)
 129             epTitle = mobj.group('episode').rpartition('/')[-1]
 130
 131         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 132         if len(mMovieParams) == 0:
 133             # The Colbert Report embeds the information in a without
 134             # a URL prefix; so extract the alternate reference
 135             # and then add the URL prefix manually.
 136
 137             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
 138             if len(altMovieParams) == 0:
 139                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
 140             else:
 141                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 142
 143         uri = mMovieParams[0][1]
 144         # Correct cc.com in uri
 145         uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.cc.com', uri)
 146
 147         index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse.urlencode({'uri': uri}))
 148         idoc = self._download_xml(
 149             index_url, epTitle,
 150             'Downloading show index', 'Unable to download episode index')
 151
 152         title = idoc.find('./channel/title').text
 153         description = idoc.find('./channel/description').text
 154
 155         entries = []
 156         item_els = idoc.findall('.//item')
 157         for part_num, itemEl in enumerate(item_els):
 158             upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
 159             thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
 160
 161             content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
 162             duration = float_or_none(content.attrib.get('duration'))
 163             mediagen_url = content.attrib['url']
 164             guid = itemEl.find('.//guid').text.rpartition(':')[-1]
 165
 166             cdoc = self._download_xml(
 167                 mediagen_url, epTitle,
 168                 'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 169
 170             turls = []
 171             for rendition in cdoc.findall('.//rendition'):
 172                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 173                 turls.append(finfo)
 174
 175             formats = []
 176             for format, rtmp_video_url in turls:
 177                 w, h = self._video_dimensions.get(format, (None, None))
 178                 formats.append({
 179                     'format_id': 'vhttp-%s' % format,
 180                     'url': self._transform_rtmp_url(rtmp_video_url),
 181                     'ext': self._video_extensions.get(format, 'mp4'),
 182                     'height': h,
 183                     'width': w,
 184                 })
 185                 formats.append({
 186                     'format_id': 'rtmp-%s' % format,
 187                     'url': rtmp_video_url,
 188                     'ext': self._video_extensions.get(format, 'mp4'),
 189                     'height': h,
 190                     'width': w,
 191                 })
 192                 self._sort_formats(formats)
 193
 194             virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
 195             entries.append({
 196                 'id': guid,
 197                 'title': virtual_id,
 198                 'formats': formats,
 199                 'uploader': show_name,
 200                 'upload_date': upload_date,
 201                 'duration': duration,
 202                 'thumbnail': thumbnail,
 203                 'description': description,
 204             })
 205
 206         return {
 207             '_type': 'playlist',
 208             'entries': entries,
 209             'title': show_name + ' ' + title,
 210             'description': description,
 211         }