[discoverynetworks] Add support for disco-api videos(closes #16724)

youtube_dl/extractor/discoverynetworks.py

@@ -3,8 +3,8 @@
 
 import re
 
-from .common import InfoExtractor
 from .brightcove import BrightcoveLegacyIE
+from .dplay import DPlayIE
 from ..compat import (
     compat_parse_qs,
     compat_urlparse,
@@ -12,8 +12,13 @@
 from ..utils import smuggle_url
 
 
-class DiscoveryNetworksDeIE(InfoExtractor):
+class DiscoveryNetworksDeIE(DPlayIE):
-    _VALID_URL = r'https?://(?:www\.)?(?:discovery|tlc|animalplanet|dmax)\.de/(?:.*#(?P<id>\d+)|(?:[^/]+/)*videos/(?P<title>[^/?#]+))'
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?P<site>discovery|tlc|animalplanet|dmax)\.de/
+                        (?:
+                           .*\#(?P<id>\d+)|
+                           (?:[^/]+/)*videos/(?P<display_id>[^/?#]+)|
+                           programme/(?P<programme>[^/]+)/video/(?P<alternate_id>[^/]+)
+                        )'''
 
     _TESTS = [{
         'url': 'http://www.tlc.de/sendungen/breaking-amish/videos/#3235167922001',
@@ -40,6 +45,14 @@ class DiscoveryNetworksDeIE(InfoExtractor):
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        alternate_id = mobj.group('alternate_id')
+        if alternate_id:
+            self._initialize_geo_bypass({
+                'countries': ['DE'],
+            })
+            return self._get_disco_api_info(
+                url, '%s/%s' % (mobj.group('programme'), alternate_id),
+                'sonic-eu1-prod.disco-api.com', mobj.group('site') + 'de')
         brightcove_id = mobj.group('id')
         if not brightcove_id:
             title = mobj.group('title')

youtube_dl/extractor/dplay.py

@@ -97,34 +97,20 @@ class DPlayIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    def _real_extract(self, url):
+    def _get_disco_api_info(self, url, display_id, disco_host, realm):
-        mobj = re.match(self._VALID_URL, url)
+        disco_base = 'https://' + disco_host
-        display_id = mobj.group('id')
+        token = self._download_json(
-        domain = mobj.group('domain')
-
-        self._initialize_geo_bypass({
-            'countries': [mobj.group('country').upper()],
-        })
-
-        webpage = self._download_webpage(url, display_id)
-
-        video_id = self._search_regex(
-            r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
-
-        if not video_id:
-            host = mobj.group('host')
-            disco_base = 'https://disco-api.%s' % host
-            self._download_json(
             '%s/token' % disco_base, display_id, 'Downloading token',
             query={
-                    'realm': host.replace('.', ''),
+                'realm': realm,
-                })
+            })['data']['attributes']['token']
+        headers = {
+            'Referer': url,
+            'Authorization': 'Bearer ' + token,
+        }
         video = self._download_json(
             '%s/content/videos/%s' % (disco_base, display_id), display_id,
-                headers={
+            headers=headers, query={
-                    'Referer': url,
-                    'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1',
-                }, query={
                 'include': 'show'
             })
         video_id = video['data']['id']
@@ -133,7 +119,7 @@ def _real_extract(self, url):
         formats = []
         for format_id, format_dict in self._download_json(
                 '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
-                    display_id)['data']['attributes']['streaming'].items():
+                display_id, headers=headers)['data']['attributes']['streaming'].items():
             if not isinstance(format_dict, dict):
                 continue
             format_url = format_dict.get('url')
@@ -180,6 +166,25 @@ def _real_extract(self, url):
             'formats': formats,
         }
 
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+        domain = mobj.group('domain')
+
+        self._initialize_geo_bypass({
+            'countries': [mobj.group('country').upper()],
+        })
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_id = self._search_regex(
+            r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
+
+        if not video_id:
+            host = mobj.group('host')
+            return self._get_disco_api_info(
+                url, display_id, 'disco-api.' + host, host.replace('.', ''))
+
         info = self._download_json(
             'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
             video_id)['data'][0]