Fix WDRMaus; extend URL matching for other Maus pages; improve ID extraction
authordirkf <fieldhouse@gmx.net>
Tue, 25 Jan 2022 12:59:31 +0000 (12:59 +0000)
committerdirkf <fieldhouse@gmx.net>
Sat, 5 Feb 2022 02:22:45 +0000 (02:22 +0000)
youtube_dl/extractor/wdr.py

index 2903d189ea387673f872c629409cad8d34caefc6..a5488f3fd4e8a05115aa2697e33a75d7f9ab605b 100644 (file)
@@ -10,6 +10,7 @@ from ..compat import (
 )
 from ..utils import (
     determine_ext,
+    dict_get,
     ExtractorError,
     js_to_json,
     strip_jsonp,
@@ -22,9 +23,10 @@ from ..utils import (
 
 
 class WDRIE(InfoExtractor):
-    _VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
+    __API_URL_TPL = '//deviceids-medp.wdr.de/ondemand/%s/%s'
+    _VALID_URL = (r'(?:https?:' + __API_URL_TPL) % (r'\d+', r'(?=\d+\.js)|wdr:)(?P<id>\d{6,})')
     _GEO_COUNTRIES = ['DE']
-    _TEST = {
+    _TESTS = [{
         'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
         'info_dict': {
             'id': 'mdb-1557833',
@@ -32,11 +34,20 @@ class WDRIE(InfoExtractor):
             'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
             'upload_date': '20180112',
         },
-    }
+    },
+    ]
+
+    def _asset_url(self, wdr_id):
+        id_len = max(len(wdr_id), 5)
+        return ''.join(('https:', self.__API_URL_TPL % (wdr_id[:id_len - 4], wdr_id, ), '.js'))
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        if url.startswith('wdr:'):
+            video_id = url[4:]
+            url = self._asset_url(video_id)
+
         metadata = self._download_json(
             url, video_id, transform_source=strip_jsonp)
 
@@ -115,10 +126,10 @@ class WDRIE(InfoExtractor):
         }
 
 
-class WDRPageIE(InfoExtractor):
-    _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
+class WDRPageIE(WDRIE):
+    _MAUS_REGEX = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/)*?(?P<maus_id>[^/?#.]+)(?:/?|/index\.php5|\.php5)$'
     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
-    _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
+    _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _MAUS_REGEX
 
     _TESTS = [
         {
@@ -180,12 +191,12 @@ class WDRPageIE(InfoExtractor):
         {
             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
             'info_dict': {
-                'id': 'mdb-1552552',
+                'id': 'mdb-2627637',
                 'ext': 'mp4',
                 'upload_date': 're:^[0-9]{8}$',
-                'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
+                'title': 're:^Die Sendung (?:mit der Maus )?vom [0-9.]{10}$',
             },
-            'skip': 'The id changes from week to week because of the new episode'
+            'skip': 'The id changes from week to week because of the new episode'
         },
         {
             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
@@ -234,7 +245,7 @@ class WDRPageIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('display_id')
+        display_id = dict_get(mobj.groupdict(), ('display_id', 'maus_id'), 'wdrmaus')
         webpage = self._download_webpage(url, display_id)
 
         entries = []
@@ -260,6 +271,14 @@ class WDRPageIE(InfoExtractor):
             jsonp_url = try_get(
                 media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
             if jsonp_url:
+                # metadata, or player JS with ['ref'] giving WDR id, or just media, perhaps
+                clip_id = media_link_obj['mediaObj'].get('ref')
+                if jsonp_url.endswith('.assetjsonp'):
+                    asset = self._download_json(
+                        jsonp_url, display_id, fatal=False, transform_source=strip_jsonp)
+                    clip_id = try_get(asset, lambda x: x['trackerData']['trackerClipId'], compat_str)
+                if clip_id:
+                    jsonp_url = self._asset_url(clip_id[4:])
                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 
         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)