]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/comedycentral.py
Merge tag 'upstream/2016.06.25'
[youtubedl] / youtube_dl / extractor / comedycentral.py
index ed3986f313a149f0db4a69dc92762730297ced1a..2b6aaa3aa47541669a59b7936477023d54ae7437 100644 (file)
@@ -2,63 +2,137 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
 from .mtv import MTVServicesInfoExtractor
-from ..utils import (
+from ..compat import (
     compat_str,
-    compat_urllib_parse,
-
+    compat_urllib_parse_urlencode,
+)
+from ..utils import (
     ExtractorError,
+    float_or_none,
     unified_strdate,
 )
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?comedycentral\.com/
-        (video-clips|episodes|cc-studios|video-collections)
+    _VALID_URL = r'''(?x)https?://(?:www\.)?cc\.com/
+        (video-clips|episodes|cc-studios|video-collections|full-episodes|shows)
         /(?P<title>.*)'''
     _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
 
-    _TEST = {
-        'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
-        'md5': '4167875aae411f903b751a21f357f1ee',
+    _TESTS = [{
+        'url': 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
+        'md5': 'c4f48e9eda1b16dd10add0744344b6d8',
         'info_dict': {
             'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
             'ext': 'mp4',
             'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
             'description': 'After a certain point, breastfeeding becomes c**kblocking.',
         },
-    }
+    }, {
+        'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/interviews/6yx39d/exclusive-rand-paul-extended-interview',
+        'only_matching': True,
+    }]
 
 
-class ComedyCentralShowsIE(InfoExtractor):
-    IE_DESC = 'The Daily Show / Colbert Report'
-    # urls can be abbreviations like :thedailyshow or :colbert
+class ComedyCentralShowsIE(MTVServicesInfoExtractor):
+    IE_DESC = 'The Daily Show / The Colbert Report'
+    # urls can be abbreviations like :thedailyshow
     # urls for episodes like:
     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
     #                     or: http://www.colbertnation.com/the-colbert-report-videos/421667/november-29-2012/moon-shattering-news
     #                     or: http://www.colbertnation.com/the-colbert-report-collections/422008/festival-of-lights/79524
-    _VALID_URL = r"""^(:(?P<shortname>tds|thedailyshow|cr|colbert|colbertnation|colbertreport)
-                      |(https?://)?(www\.)?
-                          (?P<showname>thedailyshow|colbertnation)\.com/
-                         (full-episodes/(?P<episode>.*)|
+    _VALID_URL = r'''(?x)^(:(?P<shortname>tds|thedailyshow)
+                      |https?://(:www\.)?
+                          (?P<showname>thedailyshow|thecolbertreport|tosh)\.(?:cc\.)?com/
+                         ((?:full-)?episodes/(?:[0-9a-z]{6}/)?(?P<episode>.*)|
                           (?P<clip>
-                              (the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
-                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*)))|
+                              (?:(?:guests/[^/]+|videos|video-(?:clips|playlists)|special-editions|news-team/[^/]+)/[^/]+/(?P<videotitle>[^/?#]+))
+                              |(the-colbert-report-(videos|collections)/(?P<clipID>[0-9]+)/[^/]*/(?P<cntitle>.*?))
+                              |(watch/(?P<date>[^/]*)/(?P<tdstitle>.*))
+                          )|
                           (?P<interview>
-                              extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
-                     $"""
-    _TEST = {
-        'url': 'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
-        'file': '422212.mp4',
+                              extended-interviews/(?P<interID>[0-9a-z]+)/
+                              (?:playlist_tds_extended_)?(?P<interview_title>[^/?#]*?)
+                              (?:/[^/?#]?|[?#]|$))))
+                     '''
+    _TESTS = [{
+        'url': 'http://thedailyshow.cc.com/watch/thu-december-13-2012/kristen-stewart',
         'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
         'info_dict': {
-            "upload_date": "20121214",
-            "description": "Kristen Stewart",
-            "uploader": "thedailyshow",
-            "title": "thedailyshow-kristen-stewart part 1"
+            'id': 'ab9ab3e7-5a98-4dbe-8b21-551dc0523d55',
+            'ext': 'mp4',
+            'upload_date': '20121213',
+            'description': 'Kristen Stewart learns to let loose in "On the Road."',
+            'uploader': 'thedailyshow',
+            'title': 'thedailyshow kristen-stewart part 1',
         }
-    }
+    }, {
+        'url': 'http://thedailyshow.cc.com/extended-interviews/b6364d/sarah-chayes-extended-interview',
+        'info_dict': {
+            'id': 'sarah-chayes-extended-interview',
+            'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+            'title': 'thedailyshow Sarah Chayes Extended Interview',
+        },
+        'playlist': [
+            {
+                'info_dict': {
+                    'id': '0baad492-cbec-4ec1-9e50-ad91c291127f',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 1',
+                },
+            },
+            {
+                'info_dict': {
+                    'id': '1e4fb91b-8ce7-4277-bd7c-98c9f1bbd283',
+                    'ext': 'mp4',
+                    'upload_date': '20150129',
+                    'description': 'Carnegie Endowment Senior Associate Sarah Chayes discusses how corrupt institutions function throughout the world in her book "Thieves of State: Why Corruption Threatens Global Security."',
+                    'uploader': 'thedailyshow',
+                    'title': 'thedailyshow sarah-chayes-extended-interview part 2',
+                },
+            },
+        ],
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://thedailyshow.cc.com/extended-interviews/xm3fnq/andrew-napolitano-extended-interview',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/videos/29w6fx/-realhumanpraise-for-fox-news',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/videos/gh6urb/neil-degrasse-tyson-pt--1?xrs=eml_col_031114',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/guests/michael-lewis/3efna8/exclusive---michael-lewis-extended-interview-pt--3',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/episodes/sy7yv0/april-8--2014---denis-leary',
+        'only_matching': True,
+    }, {
+        'url': 'http://thecolbertreport.cc.com/episodes/8ase07/april-8--2014---jane-goodall',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/video-playlists/npde3s/the-daily-show-19088-highlights',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/video-playlists/t6d9sg/the-daily-show-20038-highlights/be3cwo',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/special-editions/2l8fdb/special-edition---a-look-back-at-food',
+        'only_matching': True,
+    }, {
+        'url': 'http://thedailyshow.cc.com/news-team/michael-che/7wnfel/we-need-to-talk-about-israel',
+        'only_matching': True,
+    }, {
+        'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans',
+        'only_matching': True,
+    }]
 
     _available_formats = ['3500', '2200', '1700', '1200', '750', '400']
 
@@ -79,34 +153,16 @@ class ComedyCentralShowsIE(InfoExtractor):
         '400': (384, 216),
     }
 
-    @classmethod
-    def suitable(cls, url):
-        """Receives a URL and returns True if suitable for this IE."""
-        return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
-
-    @staticmethod
-    def _transform_rtmp_url(rtmp_video_url):
-        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
-        if not m:
-            raise ExtractorError('Cannot transform RTMP url')
-        base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
-        return base + m.group('finalid')
-
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, re.VERBOSE)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
+        mobj = re.match(self._VALID_URL, url)
 
         if mobj.group('shortname'):
-            if mobj.group('shortname') in ('tds', 'thedailyshow'):
-                url = 'http://www.thedailyshow.com/full-episodes/'
-            else:
-                url = 'http://www.colbertnation.com/full-episodes/'
-            mobj = re.match(self._VALID_URL, url, re.VERBOSE)
-            assert mobj is not None
+            return self.url_result('http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes')
 
         if mobj.group('clip'):
-            if mobj.group('showname') == 'thedailyshow':
+            if mobj.group('videotitle'):
+                epTitle = mobj.group('videotitle')
+            elif mobj.group('showname') == 'thedailyshow':
                 epTitle = mobj.group('tdstitle')
             else:
                 epTitle = mobj.group('cntitle')
@@ -120,9 +176,9 @@ class ComedyCentralShowsIE(InfoExtractor):
                 epTitle = mobj.group('showname')
             else:
                 epTitle = mobj.group('episode')
+        show_name = mobj.group('showname')
 
-        self.report_extraction(epTitle)
-        webpage,htmlHandle = self._download_webpage_handle(url, epTitle)
+        webpage, htmlHandle = self._download_webpage_handle(url, epTitle)
         if dlNewest:
             url = htmlHandle.geturl()
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
@@ -130,71 +186,90 @@ class ComedyCentralShowsIE(InfoExtractor):
                 raise ExtractorError('Invalid redirected URL: ' + url)
             if mobj.group('episode') == '':
                 raise ExtractorError('Redirected URL is still not specific: ' + url)
-            epTitle = mobj.group('episode')
+            epTitle = (mobj.group('episode') or mobj.group('videotitle')).rpartition('/')[-1]
 
         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
-
         if len(mMovieParams) == 0:
             # The Colbert Report embeds the information in a without
             # a URL prefix; so extract the alternate reference
             # and then add the URL prefix manually.
 
-            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
+            altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video|playlist).*?:.*?)"', webpage)
             if len(altMovieParams) == 0:
                 raise ExtractorError('unable to find Flash URL in webpage ' + url)
             else:
-                mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
+                mMovieParams = [('http://media.mtvnservices.com/' + altMovieParams[0], altMovieParams[0])]
 
         uri = mMovieParams[0][1]
-        indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
-        idoc = self._download_xml(indexUrl, epTitle,
-                                          'Downloading show index',
-                                          'unable to download episode index')
-
-        results = []
-
-        itemEls = idoc.findall('.//item')
-        for partNum,itemEl in enumerate(itemEls):
-            mediaId = itemEl.findall('./guid')[0].text
-            shortMediaId = mediaId.split(':')[-1]
-            showId = mediaId.split(':')[-2].replace('.com', '')
-            officialTitle = itemEl.findall('./title')[0].text
-            officialDate = unified_strdate(itemEl.findall('./pubDate')[0].text)
-
-            configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
-                        compat_urllib_parse.urlencode({'uri': mediaId}))
-            cdoc = self._download_xml(configUrl, epTitle,
-                                               'Downloading configuration for %s' % shortMediaId)
+        # Correct cc.com in uri
+        uri = re.sub(r'(episode:[^.]+)(\.cc)?\.com', r'\1.com', uri)
+
+        index_url = 'http://%s.cc.com/feeds/mrss?%s' % (show_name, compat_urllib_parse_urlencode({'uri': uri}))
+        idoc = self._download_xml(
+            index_url, epTitle,
+            'Downloading show index', 'Unable to download episode index')
+
+        title = idoc.find('./channel/title').text
+        description = idoc.find('./channel/description').text
+
+        entries = []
+        item_els = idoc.findall('.//item')
+        for part_num, itemEl in enumerate(item_els):
+            upload_date = unified_strdate(itemEl.findall('./pubDate')[0].text)
+            thumbnail = itemEl.find('.//{http://search.yahoo.com/mrss/}thumbnail').attrib.get('url')
+
+            content = itemEl.find('.//{http://search.yahoo.com/mrss/}content')
+            duration = float_or_none(content.attrib.get('duration'))
+            mediagen_url = content.attrib['url']
+            guid = itemEl.find('./guid').text.rpartition(':')[-1]
+
+            cdoc = self._download_xml(
+                mediagen_url, epTitle,
+                'Downloading configuration for segment %d / %d' % (part_num + 1, len(item_els)))
 
             turls = []
             for rendition in cdoc.findall('.//rendition'):
                 finfo = (rendition.attrib['bitrate'], rendition.findall('./src')[0].text)
                 turls.append(finfo)
 
-            if len(turls) == 0:
-                self._downloader.report_error('unable to download ' + mediaId + ': No videos found')
-                continue
-
             formats = []
             for format, rtmp_video_url in turls:
                 w, h = self._video_dimensions.get(format, (None, None))
                 formats.append({
+                    'format_id': 'vhttp-%s' % format,
                     'url': self._transform_rtmp_url(rtmp_video_url),
                     'ext': self._video_extensions.get(format, 'mp4'),
-                    'format_id': format,
                     'height': h,
                     'width': w,
                 })
+                formats.append({
+                    'format_id': 'rtmp-%s' % format,
+                    'url': rtmp_video_url.replace('viacomccstrm', 'viacommtvstrm'),
+                    'ext': self._video_extensions.get(format, 'mp4'),
+                    'height': h,
+                    'width': w,
+                })
+                self._sort_formats(formats)
+
+            subtitles = self._extract_subtitles(cdoc, guid)
 
-            effTitle = showId + '-' + epTitle + ' part ' + compat_str(partNum+1)
-            results.append({
-                'id': shortMediaId,
+            virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
+            entries.append({
+                'id': guid,
+                'title': virtual_id,
                 'formats': formats,
-                'uploader': showId,
-                'upload_date': officialDate,
-                'title': effTitle,
-                'thumbnail': None,
-                'description': compat_str(officialTitle),
+                'uploader': show_name,
+                'upload_date': upload_date,
+                'duration': duration,
+                'thumbnail': thumbnail,
+                'description': description,
+                'subtitles': subtitles,
             })
 
-        return results
+        return {
+            '_type': 'playlist',
+            'id': epTitle,
+            'entries': entries,
+            'title': show_name + ' ' + title,
+            'description': description,
+        }