[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/restudy.py b/youtube_dl/extractor/restudy.py
index 56a6c0f93b4c317a2460c08cdb5df96791623695..d47fb45ca5f7ce73e2905136f579cdaa6937fb32 100644 (file)
--- a/youtube_dl/extractor/restudy.py
+++ b/youtube_dl/extractor/restudy.py
@@ -5,37 +5,40 @@ from .common import InfoExtractor
 
 
 class RestudyIE(InfoExtractor):
-    _VALID_URL = r'https://www.restudy.dk/video/play/id/(?P<id>[0-9]+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:(?:www|portal)\.)?restudy\.dk/video/[^/]+/id/(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'https://www.restudy.dk/video/play/id/1637',
-        # MD5 sum of first 10241 bytes of the video file, as reported by
-        # head -c 10241 Leiden-frosteffekt-1637.mp4 | md5sum
-        'md5': '4e755c4287f292a1fe5363834a683818',
         'info_dict': {
             'id': '1637',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'Leiden-frosteffekt',
+            'description': 'Denne video er et eksperiment med flydende kvælstof.',
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
         }
-    }
+    }, {
+        'url': 'https://portal.restudy.dk/video/leiden-frosteffekt/id/1637',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
-        xml_url = (
-            'https://www.restudy.dk/awsmedia/SmilDirectory/video_%s.xml'
-            % video_id)
-        xml = self._download_webpage(xml_url, video_id)
-
-        base = self._search_regex(
-            r'<meta base="([^"]+)', xml, 'meta base')
-        # TODO: Provide multiple video qualities instead of forcing highest
-        filename = self._search_regex(
-            r'<video src="mp4:([^"]+_high\.mp4)', xml, 'filename')
-        url = '%s%s' % (base, filename)
-        title = self._og_search_title(webpage)
+
+        title = self._og_search_title(webpage).strip()
+        description = self._og_search_description(webpage).strip()
+
+        formats = self._extract_smil_formats(
+            'https://cdn.portal.restudy.dk/dynamic/themes/front/awsmedia/SmilDirectory/video_%s.xml' % video_id,
+            video_id)
+        self._sort_formats(formats)
+
         return {
             'id': video_id,
             'title': title,
-            'url': url,
-            'protocol': 'rtmp',
+            'description': description,
+            'formats': formats,
         }