[srgssrplay] Improve _VALID_URL (closes #21155)

meunierd · Feb 13, 2020 · 4797754 · 4797754
1 parent 26ed248
commit 4797754
Showing 1 changed file with 15 additions and 11 deletions.
diff --git a/youtube_dl/extractor/srgssr.py b/youtube_dl/extractor/srgssr.py
@@ -106,7 +106,16 @@ def _real_extract(self, url):
 
 class SRGSSRPlayIE(InfoExtractor):
     IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
-    _VALID_URL = r'https?://(?:(?:www|play)\.)?(?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/(?:[^/]+/|popup)(?P<type>video|audio)(?:/[^?]+|player)\?id=(?P<id>[0-9a-f\-]{36}|\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:(?:www|play)\.)?
+                        (?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/
+                        (?:
+                            [^/]+/(?P<type>video|audio)/[^?]+|
+                            popup(?P<type_2>video|audio)player
+                        )
+                        \?id=(?P<id>[0-9a-f\-]{36}|\d+)
+                    '''
 
     _TESTS = [{
         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
@@ -165,18 +174,13 @@ class SRGSSRPlayIE(InfoExtractor):
         }
     }, {
         'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01',
-        'md5': 'f6247aa7c905b81c9ba7f50fb22e2fbd',
-        'info_dict': {
-            'id': 'c4dba0ca-e75b-43b2-a34f-f708a4932e01',
-            'ext': 'mp4',
-            'upload_date': '20190122',
-            'title': 'Erster Selfie-Stick (1983)',
-            'description': 'md5:23a6b40024e583137e4137f5946543c1',
-            'timestamp': 1548155133,
-        }
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+        mobj = re.match(self._VALID_URL, url)
+        bu = mobj.group('bu')
+        media_type = mobj.group('type') or mobj.group('type_2')
+        media_id = mobj.group('id')
         # other info can be extracted from url + '&layout=json'
         return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR')