[cnbc] Add new extractor(closes #8012)

[youtube-dl] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 2230dfe021527e3e1fc64beb57b758795d34a82c..236c999724ee90f86ecad4bc2b771aefca0bf6f5 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -31,10 +31,9 @@ _x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 class ThePlatformBaseIE(OnceIE):
     def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
         meta = self._download_xml(smil_url, video_id, note=note, query={'format': 'SMIL'})
-        error_element = find_xpath_attr(
-            meta, _x('.//smil:ref'), 'src',
-            'http://link.theplatform.com/s/errorFiles/Unavailable.mp4')
-        if error_element is not None:
+        error_element = find_xpath_attr(meta, _x('.//smil:ref'), 'src')
+        if error_element is not None and error_element.attrib['src'].startswith(
+                'http://link.theplatform.com/s/errorFiles/Unavailable.'):
             raise ExtractorError(error_element.attrib['abstract'], expected=True)
 
         smil_formats = self._parse_smil_formats(
@@ -83,7 +82,7 @@ class ThePlatformBaseIE(OnceIE):
 class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
         (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
-           (?:(?P<media>(?:(?:[^/]+/)+select/)?media/)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
+           (?:(?:(?:[^/]+/)+select/)?(?P<media>media/(?:guid/\d+/)?)|(?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/))?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TESTS = [{
@@ -171,10 +170,10 @@ class ThePlatformIE(ThePlatformBaseIE):
         if not provider_id:
             provider_id = 'dJ5BDC'
 
-        path = provider_id
+        path = provider_id + '/'
         if mobj.group('media'):
-            path += '/media'
-        path += '/' + video_id
+            path += mobj.group('media')
+        path += video_id
 
         qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
         if 'guid' in qs_dict: