From db6364df21e02ac7f400a25a7aef8e4dd0824c5d Mon Sep 17 00:00:00 2001
From: Jakub Wieczorek <jakub.adam.wieczorek@gmail.com>
Date: Mon, 20 Jul 2020 11:07:22 +0200
Subject: [PATCH] [polskieradio] Fix extraction for the updated article page
 theme

---
 youtube_dl/extractor/polskieradio.py | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/youtube_dl/extractor/polskieradio.py b/youtube_dl/extractor/polskieradio.py
index 978d6f813..6728e2f05 100644
--- a/youtube_dl/extractor/polskieradio.py
+++ b/youtube_dl/extractor/polskieradio.py
@@ -15,6 +15,7 @@ from ..utils import (
     int_or_none,
     strip_or_none,
     unified_timestamp,
+    unescapeHTML,
 )
 
 
@@ -39,6 +40,25 @@ class PolskieRadioIE(InfoExtractor):
                 'thumbnail': r're:^https?://static\.prsa\.pl/images/.*\.jpg$'
             },
         }],
+    }, {
+        'url': 'https://www.polskieradio.pl/8/2382/Artykul/2534482,Zagarysci-Poezja-jak-spoiwo',
+        'info_dict': {
+            'id': '2534482',
+            'title': 'Żagaryści. Poezja jak spoiwo',
+            'description': 'md5:f18d95d5dcba747a09b635e21a4c0695',
+        },
+        'playlist': [{
+            'md5': 'd07559829f61d5a93a75755987ded760',
+            'info_dict': {
+                'id': '2516679',
+                'ext': 'mp3',
+                'title': 'md5:c6e1234e0b747ad883cb91b7ad06b98c',
+                'timestamp': 1592654400,
+                'upload_date': '20200620',
+                'duration': 1430,
+                'thumbnail': r're:^https?://static\.prsa\.pl/images/.*\.jpg$'
+            },
+        }],
     }, {
         'url': 'http://www.polskieradio.pl/265/5217/Artykul/1635803,Euro-2016-nie-ma-miejsca-na-blad-Polacy-graja-ze-Szwajcaria-o-cwiercfinal',
         'info_dict': {
@@ -78,8 +98,8 @@ class PolskieRadioIE(InfoExtractor):
 
         media_urls = set()
 
-        for data_media in re.findall(r'<[^>]+data-media=({[^>]+})', content):
-            media = self._parse_json(data_media, playlist_id, fatal=False)
+        for data_media in re.findall(r'<[^>]+data-media="?({[^>]+})"?', content):
+            media = self._parse_json(data_media, playlist_id, transform_source=unescapeHTML, fatal=False)
             if not media.get('file') or not media.get('desc'):
                 continue
             media_url = self._proto_relative_url(media['file'], 'http:')
@@ -98,6 +118,7 @@ class PolskieRadioIE(InfoExtractor):
 
         title = self._og_search_title(webpage).strip()
         description = strip_or_none(self._og_search_description(webpage))
+        description = description.replace('\xa0', ' ') if description is not None else None
 
         return self.playlist_result(entries, playlist_id, title, description)