From aca2fd222fc951dd8e177d4c5584f5276e719825 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Sergey=20M=E2=80=A4?= <dstftw@gmail.com>
Date: Sat, 11 Jan 2020 02:18:36 +0700
Subject: [PATCH] [orf:radio] Clean description and improve extraction

---
 youtube_dl/extractor/orf.py | 46 ++++++++++++++++++++++---------------
 1 file changed, 28 insertions(+), 18 deletions(-)

diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py
index 816bd212e..45fc745a3 100644
--- a/youtube_dl/extractor/orf.py
+++ b/youtube_dl/extractor/orf.py
@@ -6,12 +6,14 @@
 from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
+    clean_html,
     determine_ext,
     float_or_none,
     HEADRequest,
     int_or_none,
     orderedSet,
     remove_end,
+    str_or_none,
     strip_jsonp,
     unescapeHTML,
     unified_strdate,
@@ -162,30 +164,37 @@ def _real_extract(self, url):
         show_id = mobj.group('show')
 
         data = self._download_json(
-            'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s' % (station, show_id, show_date),
-            show_id
-        )
+            'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s'
+            % (station, show_id, show_date), show_id)
 
-        def extract_entry_dict(info, title, subtitle):
-            return {
-                'id': info['loopStreamId'].replace('.mp3', ''),
-                'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (station, info['loopStreamId']),
+        entries = []
+        for info in data['streams']:
+            loop_stream_id = str_or_none(info.get('loopStreamId'))
+            if not loop_stream_id:
+                continue
+            title = str_or_none(data.get('title'))
+            if not title:
+                continue
+            start = int_or_none(info.get('start'), scale=1000)
+            end = int_or_none(info.get('end'), scale=1000)
+            duration = end - start if end and start else None
+            entries.append({
+                'id': loop_stream_id.replace('.mp3', ''),
+                'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (station, loop_stream_id),
                 'title': title,
-                'description': subtitle,
-                'duration': (info['end'] - info['start']) / 1000,
-                'timestamp': info['start'] / 1000,
+                'description': clean_html(data.get('subtitle')),
+                'duration': duration,
+                'timestamp': start,
                 'ext': 'mp3',
-                'series': data.get('programTitle')
-            }
-
-        entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
+                'series': data.get('programTitle'),
+            })
 
         return {
             '_type': 'playlist',
             'id': show_id,
-            'title': data['title'],
-            'description': data['subtitle'],
-            'entries': entries
+            'title': data.get('title'),
+            'description': clean_html(data.get('subtitle')),
+            'entries': entries,
         }
 
 
@@ -206,7 +215,8 @@ class ORFFM4IE(ORFRadioIE):
             'timestamp': 1483819257,
             'upload_date': '20170107',
         },
-        'skip': 'Shows from ORF radios are only available for 7 days.'
+        'skip': 'Shows from ORF radios are only available for 7 days.',
+        'only_matching': True,
     }