[ard:mediathek] Fix title and description extraction (closes #18349)

[youtube-dl] / youtube_dl / extractor / ard.py

diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py
index cff8ca4a54a80915ea6a29cd76ca82762e81dd37..84e96f76984224945d6ff8af93302e69b78dca02 100644 (file)
--- a/youtube_dl/extractor/ard.py
+++ b/youtube_dl/extractor/ard.py
@@ -21,7 +21,7 @@ from ..compat import compat_etree_fromstring
 
 class ARDMediathekIE(InfoExtractor):
     IE_NAME = 'ARD:mediathek'
-    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
 
     _TESTS = [{
         # available till 26.07.2022
@@ -37,6 +37,9 @@ class ARDMediathekIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
+        'only_matching': True,
     }, {
         # audio
         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
@@ -170,13 +173,18 @@ class ARDMediathekIE(InfoExtractor):
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
              r'<meta name="dcterms\.title" content="(.*?)"/>',
-             r'<h4 class="headline">(.*?)</h4>'],
+             r'<h4 class="headline">(.*?)</h4>',
+             r'<title[^>]*>(.*?)</title>'],
             webpage, 'title')
         description = self._html_search_meta(
             'dcterms.abstract', webpage, 'description', default=None)
         if description is None:
             description = self._html_search_meta(
-                'description', webpage, 'meta description')
+                'description', webpage, 'meta description', default=None)
+        if description is None:
+            description = self._html_search_regex(
+                r'<p\s+class="teasertext">(.+?)</p>',
+                webpage, 'teaser text', default=None)
 
         # Thumbnail is sometimes not present.
         # It is in the mobile version, but that seems to use a different URL
@@ -307,7 +315,7 @@ class ARDBetaMediathekIE(InfoExtractor):
         display_id = mobj.group('display_id')
 
         webpage = self._download_webpage(url, display_id)
-        data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);', webpage, 'json')
+        data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
         data = self._parse_json(data_json, display_id)
 
         res = {
@@ -338,12 +346,9 @@ class ARDBetaMediathekIE(InfoExtractor):
                 format_url = widget['_stream']['json'][0]
 
                 if format_url.endswith('.f4m'):
-                    # Skip f4m - these URLs just return a 403
-                    formats.append({
-                        'format_id': 'f4m-' + widget['_quality'],
-                        'url': format_url,
-                        'preference': -1001,
-                    })
+                    formats.extend(self._extract_f4m_formats(
+                        format_url + '?hdcore=3.11.0',
+                        video_id, f4m_id='hds', fatal=False))
                 elif format_url.endswith('m3u8'):
                     formats.extend(self._extract_m3u8_formats(
                         format_url, video_id, 'mp4', m3u8_id='hls', fatal=False))