[cbs, cbsnews] fix extraction(fixes #10393)

youtube_dl/extractor/cbs.py

@@ -4,6 +4,7 @@
 from ..utils import (
     int_or_none,
     find_xpath_attr,
+    ExtractorError,
 )
 
 
@@ -17,19 +18,6 @@ def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
             }]
         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
-    def _extract_video_info(self, filter_query, video_id):
-        return self._extract_feed_info(
-            'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: {
-                'series': entry.get('cbs$SeriesTitle'),
-                'season_number': int_or_none(entry.get('cbs$SeasonNumber')),
-                'episode': entry.get('cbs$EpisodeTitle'),
-                'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')),
-            }, {
-                'StreamPack': {
-                    'manifest': 'm3u',
-                }
-            })
-
 
 class CBSIE(CBSBaseIE):
     _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
@@ -38,7 +26,6 @@ class CBSIE(CBSBaseIE):
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
         'info_dict': {
             'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
-            'display_id': 'connect-chat-feat-garth-brooks',
             'ext': 'mp4',
             'title': 'Connect Chat feat. Garth Brooks',
             'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
@@ -47,7 +34,10 @@ class CBSIE(CBSBaseIE):
             'upload_date': '20131127',
             'uploader': 'CBSI-NEW',
         },
-        'expected_warnings': ['Failed to download m3u8 information'],
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
         '_skip': 'Blocked outside the US',
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
@@ -56,8 +46,31 @@ class CBSIE(CBSBaseIE):
         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
         'only_matching': True,
     }]
-    TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
+
+    def _extract_video_info(self, guid):
+        path = 'dJ5BDC/media/guid/2198311517/' + guid
+        smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path
+        formats, subtitles = self._extract_theplatform_smil(smil_url + '&manifest=m3u', guid)
+        for r in ('HLS&formats=M3U', 'RTMP', 'WIFI', '3G'):
+            try:
+                tp_formats, _ = self._extract_theplatform_smil(smil_url + '&assetTypes=' + r, guid, 'Downloading %s SMIL data' % r.split('&')[0])
+                formats.extend(tp_formats)
+            except ExtractorError:
+                continue
+        self._sort_formats(formats)
+        metadata = self._download_theplatform_metadata(path, guid)
+        info = self._parse_theplatform_metadata(metadata)
+        info.update({
+            'id': guid,
+            'formats': formats,
+            'subtitles': subtitles,
+            'series': metadata.get('cbs$SeriesTitle'),
+            'season_number': int_or_none(metadata.get('cbs$SeasonNumber')),
+            'episode': metadata.get('cbs$EpisodeTitle'),
+            'episode_number': int_or_none(metadata.get('cbs$EpisodeNumber')),
+        })
+        return info
 
     def _real_extract(self, url):
         content_id = self._match_id(url)
-        return self._extract_video_info('byGuid=%s' % content_id, content_id)
+        return self._extract_video_info(content_id)

youtube_dl/extractor/cbsnews.py

@@ -2,13 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from .cbs import CBSBaseIE
+from .cbs import CBSIE
 from ..utils import (
     parse_duration,
 )
 
 
-class CBSNewsIE(CBSBaseIE):
+class CBSNewsIE(CBSIE):
     IE_DESC = 'CBS News'
     _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
 
@@ -35,7 +35,8 @@ class CBSNewsIE(CBSBaseIE):
                 'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
                 'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
-                'upload_date': '19700101',
+                'upload_date': '20140404',
+                'timestamp': 1396650660,
                 'uploader': 'CBSI-NEW',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
@@ -63,7 +64,7 @@ def _real_extract(self, url):
 
         item = video_info['item'] if 'item' in video_info else video_info
         guid = item['mpxRefId']
-        return self._extract_video_info('byGuid=%s' % guid, guid)
+        return self._extract_video_info(guid)
 
 
 class CBSNewsLiveVideoIE(InfoExtractor):

youtube_dl/extractor/cbssports.py

@@ -23,6 +23,9 @@ class CBSSportsIE(CBSBaseIE):
         }
     }]
 
+    def _extract_video_info(self, filter_query, video_id):
+        return self._extract_feed_info('dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         return self._extract_video_info('byId=%s' % video_id, video_id)