youtube-dl

commit: 26499ba823ecba99e18cf5cc76e001f4bbbe4293
parent 58f6c2112d55cdd77ad76b323760bb934d7e7576
Author: Remita Amine <remitamine@gmail.com>
Date:   Sun, 17 Jan 2021 17:35:10 +0100

[aljazeera] fix extraction(closes #20911)(closes #27779)

Diffstat:
M
youtube_dl/extractor/aljazeera.py
41
++++++++++++++++++++++++++++++++---------

1 file changed, 32 insertions(+), 9 deletions(-)
diff --git a/youtube_dl/extractor/aljazeera.py b/youtube_dl/extractor/aljazeera.py
@@ -1,13 +1,16 @@
 from __future__ import unicode_literals
 
+import json
+import re
+
 from .common import InfoExtractor
 
 
 class AlJazeeraIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?:programmes|video)/.*?/(?P<id>[^/]+)\.html'
+    _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?P<type>program/[^/]+|(?:feature|video)s)/\d{4}/\d{1,2}/\d{1,2}/(?P<id>[^/?&#]+)'
 
     _TESTS = [{
-        'url': 'http://www.aljazeera.com/programmes/the-slum/2014/08/deliverance-201482883754237240.html',
+        'url': 'https://www.aljazeera.com/program/episode/2014/9/19/deliverance',
         'info_dict': {
             'id': '3792260579001',
             'ext': 'mp4',
@@ -20,14 +23,34 @@ class AlJazeeraIE(InfoExtractor):
         'add_ie': ['BrightcoveNew'],
         'skip': 'Not accessible from Travis CI server',
     }, {
-        'url': 'http://www.aljazeera.com/video/news/2017/05/sierra-leone-709-carat-diamond-auctioned-170511100111930.html',
+        'url': 'https://www.aljazeera.com/videos/2017/5/11/sierra-leone-709-carat-diamond-to-be-auctioned-off',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.aljazeera.com/features/2017/8/21/transforming-pakistans-buses-into-art',
         'only_matching': True,
     }]
-    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/665003303001/default_default/index.html?videoId=%s'
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
 
     def _real_extract(self, url):
-        program_name = self._match_id(url)
-        webpage = self._download_webpage(url, program_name)
-        brightcove_id = self._search_regex(
-            r'RenderPagesVideo\(\'(.+?)\'', webpage, 'brightcove id')
-        return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id)
+        post_type, name = re.match(self._VALID_URL, url).groups()
+        post_type = {
+            'features': 'post',
+            'program': 'episode',
+            'videos': 'video',
+        }[post_type.split('/')[0]]
+        video = self._download_json(
+            'https://www.aljazeera.com/graphql', name, query={
+                'operationName': 'SingleArticleQuery',
+                'variables': json.dumps({
+                    'name': name,
+                    'postType': post_type,
+                }),
+            }, headers={
+                'wp-site': 'aje',
+            })['data']['article']['video']
+        video_id = video['id']
+        account_id = video.get('accountId') or '665003303001'
+        player_id = video.get('playerId') or 'BkeSH5BDb'
+        return self.url_result(
+            self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id),
+            'BrightcoveNew', video_id)