Raphaël G. Git Repositories - youtubedl/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\.)?cc\.com/
  18         (video-clips|episodes|cc-studios|video-collections|full-episodes)
  19         /(?P<title>.*)'''
  20     _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
  21
  22     _TEST = {
  23         'url': 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
  24         'md5': 'c4f48e9eda1b16dd10add0744344b6d8',
  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/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
  45                           (?P<clip>
  46                               (?:(?:guests/[^/]+|videos|video-playlists|special-editions)/[^/]+/(?P<videotitle>[^/?#]+))
  47                               |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
  48                               |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
  49                           )|
  50                           (?P<interview>
  51                               extended-interviews/(?P<interID>[0-9a-z]+)/(?:playlist_tds_extended_)?(?P<interview_title>.*?)(/.*?)?)))
  52                      (?:[?#].*|$)'''
  53     _TEST = {
  54         'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
  55         'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
  56         'info_dict': {
  57             'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
  58             'ext': 'mp4',
  59             'upload_date': '20121213',
  60             'description': 'Kristen Stewart learns to let loose in "On the Road."',
  61             'uploader': 'thedailyshow',
  62             'title': 'thedailyshow kristen-stewart part 1',
  63         }
  64     }
  65
  66     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  67
  68     _video_extensions = {
  69         '3500': 'mp4',
  70         '2200': 'mp4',
  71         '1700': 'mp4',
  72         '1200': 'mp4',
  73         '750': 'mp4',
  74         '400': 'mp4',
  75     }
  76     _video_dimensions = {
  77         '3500': (1280, 720),
  78         '2200': (960, 540),
  79         '1700': (768, 432),
  80         '1200': (640, 360),
  81         '750': (512, 288),
  82         '400': (384, 216),
  83     }
  84
  85     @staticmethod
  86     def _transform_rtmp_url(rtmp_video_url):
  87         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
  88         if not m:
  89             raise ExtractorError('Cannot transform RTMP url')
  90         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
  91         return base + m.group('finalid')
  92
  93     def _real_extract(self, url):
  94         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
  95         if mobj is None:
  96             raise ExtractorError('Invalid URL: %s' % url)
  97
  98         if mobj.group('shortname'):
  99             if mobj.group('shortname') in ('tds', 'thedailyshow'):
 100                 url = 'http://thedailyshow.cc.com/full-episodes/'
 101             else:
 102                 url = 'http://thecolbertreport.cc.com/full-episodes/'
 103             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 104             assert mobj is not None
 105
 106         if mobj.group('clip'):
 107             if mobj.group('videotitle'):
 108                 epTitle = mobj.group('videotitle')
 109             elif mobj.group('showname') == 'thedailyshow':
 110                 epTitle = mobj.group('tdstitle')
 111             else:
 112                 epTitle = mobj.group('cntitle')
 113             dlNewest = False
 114         elif mobj.group('interview'):
 115             epTitle = mobj.group('interview_title')
 116             dlNewest = False
 117         else:
 118             dlNewest = not mobj.group('episode')
 119             if dlNewest:
 120                 epTitle = mobj.group('showname')
 121             else:
 122                 epTitle = mobj.group('episode')
 123         show_name = mobj.group('showname')
 124
 125         webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
 126         if dlNewest:
 127             url = htmlHandle.geturl()
 128             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 129             if mobj is None:
 130                 raise ExtractorError('Invalid redirected URL: ' + url)
 131             if mobj.group('episode') == '':
 132                 raise ExtractorError('Redirected URL is still not specific: ' + url)
 133             epTitle = (mobj.group('episode') or mobj.group('videotitle')).rpartition('/')[-1]
 134
 135         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 136         if len(mMovieParams) == 0:
 137             # The Colbert Report embeds the information in a without
 138             # a URL prefix; so extract the alternate reference
 139             # and then add the URL prefix manually.
 140
 141             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
 142             if len(altMovieParams) == 0:
 143                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
 144             else:
 145                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 146
 147         uri = mMovieParams[0][1]
 148         # Correct cc.com in uri
 149         uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.cc.com', uri)
 150
 151         index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse.urlencode({'uri': uri}))
 152         idoc = self._download_xml(
 153             index_url, epTitle,
 154             'Downloading show index', 'Unable to download episode index')
 155
 156         title = idoc.find('./channel/title').text
 157         description = idoc.find('./channel/description').text
 158
 159         entries = []
 160         item_els = idoc.findall('.//item')
 161         for part_num, itemEl in enumerate(item_els):
 162             upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
 163             thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
 164
 165             content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
 166             duration = float_or_none(content.attrib.get('duration'))
 167             mediagen_url = content.attrib['url']
 168             guid = itemEl.find('./guid').text.rpartition(':')[-1]
 169
 170             cdoc = self._download_xml(
 171                 mediagen_url, epTitle,
 172                 'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 173
 174             turls = []
 175             for rendition in cdoc.findall('.//rendition'):
 176                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 177                 turls.append(finfo)
 178
 179             formats = []
 180             for format, rtmp_video_url in turls:
 181                 w, h = self._video_dimensions.get(format, (None, None))
 182                 formats.append({
 183                     'format_id': 'vhttp-%s' % format,
 184                     'url': self._transform_rtmp_url(rtmp_video_url),
 185                     'ext': self._video_extensions.get(format, 'mp4'),
 186                     'height': h,
 187                     'width': w,
 188                 })
 189                 formats.append({
 190                     'format_id': 'rtmp-%s' % format,
 191                     'url': rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm'),
 192                     'ext': self._video_extensions.get(format, 'mp4'),
 193                     'height': h,
 194                     'width': w,
 195                 })
 196                 self._sort_formats(formats)
 197
 198             virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
 199             entries.append({
 200                 'id': guid,
 201                 'title': virtual_id,
 202                 'formats': formats,
 203                 'uploader': show_name,
 204                 'upload_date': upload_date,
 205                 'duration': duration,
 206                 'thumbnail': thumbnail,
 207                 'description': description,
 208             })
 209
 210         return {
 211             '_type': 'playlist',
 212             'entries': entries,
 213             'title': show_name + ' ' + title,
 214             'description': description,
 215         }