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|news-team/[^/]+)/[^/]+/(?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     _TESTS = [{
  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         'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
  66         'only_matching': True,
  67     }, {
  68         'url': 'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
  69         'only_matching': True,
  70     }, {
  71         'url': 'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
  72         'only_matching': True,
  73     }, {
  74         'url': 'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
  75         'only_matching': True,
  76     }, {
  77         'url': 'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
  78         'only_matching': True,
  79     }, {
  80         'url': 'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
  81         'only_matching': True,
  82     }, {
  83         'url': 'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
  84         'only_matching': True,
  85     }, {
  86         'url': 'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
  87         'only_matching': True,
  88     }, {
  89         'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
  90         'only_matching': True,
  91     }]
  92
  93     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
  94
  95     _video_extensions = {
  96         '3500': 'mp4',
  97         '2200': 'mp4',
  98         '1700': 'mp4',
  99         '1200': 'mp4',
 100         '750': 'mp4',
 101         '400': 'mp4',
 102     }
 103     _video_dimensions = {
 104         '3500': (1280, 720),
 105         '2200': (960, 540),
 106         '1700': (768, 432),
 107         '1200': (640, 360),
 108         '750': (512, 288),
 109         '400': (384, 216),
 110     }
 111
 112     @staticmethod
 113     def _transform_rtmp_url(rtmp_video_url):
 114         m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
 115         if not m:
 116             raise ExtractorError('Cannot transform RTMP url')
 117         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
 118         return base + m.group('finalid')
 119
 120     def _real_extract(self, url):
 121         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 122         if mobj is None:
 123             raise ExtractorError('Invalid URL: %s' % url)
 124
 125         if mobj.group('shortname'):
 126             if mobj.group('shortname') in ('tds', 'thedailyshow'):
 127                 url = 'http://thedailyshow.cc.com/full-episodes/'
 128             else:
 129                 url = 'http://thecolbertreport.cc.com/full-episodes/'
 130             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 131             assert mobj is not None
 132
 133         if mobj.group('clip'):
 134             if mobj.group('videotitle'):
 135                 epTitle = mobj.group('videotitle')
 136             elif mobj.group('showname') == 'thedailyshow':
 137                 epTitle = mobj.group('tdstitle')
 138             else:
 139                 epTitle = mobj.group('cntitle')
 140             dlNewest = False
 141         elif mobj.group('interview'):
 142             epTitle = mobj.group('interview_title')
 143             dlNewest = False
 144         else:
 145             dlNewest = not mobj.group('episode')
 146             if dlNewest:
 147                 epTitle = mobj.group('showname')
 148             else:
 149                 epTitle = mobj.group('episode')
 150         show_name = mobj.group('showname')
 151
 152         webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
 153         if dlNewest:
 154             url = htmlHandle.geturl()
 155             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
 156             if mobj is None:
 157                 raise ExtractorError('Invalid redirected URL: ' + url)
 158             if mobj.group('episode') == '':
 159                 raise ExtractorError('Redirected URL is still not specific: ' + url)
 160             epTitle = (mobj.group('episode') or mobj.group('videotitle')).rpartition('/')[-1]
 161
 162         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
 163         if len(mMovieParams) == 0:
 164             # The Colbert Report embeds the information in a without
 165             # a URL prefix; so extract the alternate reference
 166             # and then add the URL prefix manually.
 167
 168             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
 169             if len(altMovieParams) == 0:
 170                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
 171             else:
 172                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 173
 174         uri = mMovieParams[0][1]
 175         # Correct cc.com in uri
 176         uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.cc.com', uri)
 177
 178         index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse.urlencode({'uri': uri}))
 179         idoc = self._download_xml(
 180             index_url, epTitle,
 181             'Downloading show index', 'Unable to download episode index')
 182
 183         title = idoc.find('./channel/title').text
 184         description = idoc.find('./channel/description').text
 185
 186         entries = []
 187         item_els = idoc.findall('.//item')
 188         for part_num, itemEl in enumerate(item_els):
 189             upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
 190             thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
 191
 192             content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
 193             duration = float_or_none(content.attrib.get('duration'))
 194             mediagen_url = content.attrib['url']
 195             guid = itemEl.find('./guid').text.rpartition(':')[-1]
 196
 197             cdoc = self._download_xml(
 198                 mediagen_url, epTitle,
 199                 'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 200
 201             turls = []
 202             for rendition in cdoc.findall('.//rendition'):
 203                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
 204                 turls.append(finfo)
 205
 206             formats = []
 207             for format, rtmp_video_url in turls:
 208                 w, h = self._video_dimensions.get(format, (None, None))
 209                 formats.append({
 210                     'format_id': 'vhttp-%s' % format,
 211                     'url': self._transform_rtmp_url(rtmp_video_url),
 212                     'ext': self._video_extensions.get(format, 'mp4'),
 213                     'height': h,
 214                     'width': w,
 215
 216                     'format_note': 'HTTP 400 at the moment (patches welcome!)',
 217                     'preference': -100,
 218                 })
 219                 formats.append({
 220                     'format_id': 'rtmp-%s' % format,
 221                     'url': rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm'),
 222                     'ext': self._video_extensions.get(format, 'mp4'),
 223                     'height': h,
 224                     'width': w,
 225                 })
 226                 self._sort_formats(formats)
 227
 228             virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
 229             entries.append({
 230                 'id': guid,
 231                 'title': virtual_id,
 232                 'formats': formats,
 233                 'uploader': show_name,
 234                 'upload_date': upload_date,
 235                 'duration': duration,
 236                 'thumbnail': thumbnail,
 237                 'description': description,
 238             })
 239
 240         return {
 241             '_type': 'playlist',
 242             'entries': entries,
 243             'title': show_name + ' ' + title,
 244             'description': description,
 245         }