Update to reflect website changes.

youtube_dlc/extractor/urplay.py

@@ -3,6 +3,7 @@ from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import unified_timestamp
+import re
 
 
 class URPlayIE(InfoExtractor):
@@ -13,10 +14,10 @@ class URPlayIE(InfoExtractor):
         'info_dict': {
             'id': '203704',
             'ext': 'mp4',
-            'title': 'UR Samtiden - Livet, universum och rymdens märkliga musik : Om vetenskap, kritiskt tänkande och motstånd',
+            'title': 'Om vetenskap, kritiskt tänkande och motstånd',
             'description': 'md5:5344508a52aa78c1ced6c1b8b9e44e9a',
-            'timestamp': 1513512768,
-            'upload_date': '20171217',
+            'timestamp': 1513292400,
+            'upload_date': '20171214',
         },
     }, {
         'url': 'https://urskola.se/Produkter/190031-Tripp-Trapp-Trad-Sovkudde',
@@ -37,35 +38,41 @@ class URPlayIE(InfoExtractor):
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
-        urplayer_data = self._parse_json(self._search_regex(
-            r'urPlayer\.init\(({.+?})\);', webpage, 'urplayer data'), video_id)
-        host = self._download_json('http://streaming-loadbalancer.ur.se/loadbalancer.json', video_id)['redirect']
+        urplayer_data = re.sub(""", "\"", self._search_regex(
+            r'components\/Player\/Player\" data-react-props=\"({.+?})\"',
+            webpage, 'urplayer data'))
+        urplayer_data = self._parse_json(urplayer_data, video_id)
+        for i in range(len(urplayer_data['accessibleEpisodes'])):
+            if urplayer_data.get('accessibleEpisodes', {})[i].get('id') == int(video_id):
+                urplayer_data = urplayer_data['accessibleEpisodes'][i]
+                break
 
+        host = self._download_json('http://streaming-loadbalancer.ur.se/loadbalancer.json', video_id)['redirect']
         formats = []
-        for quality_attr, quality, preference in (('', 'sd', 0), ('_hd', 'hd', 1)):
-            file_http = urplayer_data.get('file_http' + quality_attr) or urplayer_data.get('file_http_sub' + quality_attr)
-            if file_http:
+        urplayer_streams = urplayer_data.get("streamingInfo")
+        for quality in ('sd'), ('hd'):
+            location = (urplayer_streams.get("raw", {}).get(quality, {}).get("location")
+                        or urplayer_streams.get("sweComplete", {}).get(quality, {}).get("location"))
+            if location:
                 formats.extend(self._extract_wowza_formats(
-                    'http://%s/%splaylist.m3u8' % (host, file_http), video_id, skip_protocols=['rtmp', 'rtsp']))
+                               'http://%s/%s/playlist.m3u8' % (host, location), video_id,
+                               skip_protocols=['f4m', 'rtmp', 'rtsp']))
         self._sort_formats(formats)
-
         subtitles = {}
-        for subtitle in urplayer_data.get('subtitles', []):
-            subtitle_url = subtitle.get('file')
-            kind = subtitle.get('kind')
-            if not subtitle_url or (kind and kind != 'captions'):
-                continue
-            subtitles.setdefault(subtitle.get('label', 'Svenska'), []).append({
-                'url': subtitle_url,
+        subs = urplayer_streams.get("sweComplete", {}).get("tt", {}).get("location")
+        if subs:
+            subtitles.setdefault('Svenska', []).append({
+                'url': subs,
             })
 
         return {
             'id': video_id,
             'title': urplayer_data['title'],
             'description': self._og_search_description(webpage),
-            'thumbnail': urplayer_data.get('image'),
-            'timestamp': unified_timestamp(self._html_search_meta(('uploadDate', 'schema:uploadDate'), webpage, 'timestamp')),
-            'series': urplayer_data.get('series_title'),
+            'thumbnail': urplayer_data.get('image', {}).get('1280x720'),
+            'timestamp': unified_timestamp(self._html_search_meta(('uploadDate', 'schema:uploadDate'),
+                                           webpage, 'timestamp')),
+            'series': urplayer_data.get('seriesTitle'),
             'subtitles': subtitles,
             'formats': formats,
         }