[orf:fm4] Fix extraction (#23599)

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

diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py
index c1fb580ca8d9a20d785d71194d2822e51b2ee1ae..816bd212ee79132342af899a8356c68dd1dc3ae1 100644 (file)
--- a/youtube_dl/extractor/orf.py
+++ b/youtube_dl/extractor/orf.py
@@ -15,6 +15,7 @@ from ..utils import (
     strip_jsonp,
     unescapeHTML,
     unified_strdate,
+    url_or_none,
 )
 
 
@@ -68,26 +69,36 @@ class ORFTVthekIE(InfoExtractor):
                 webpage, 'playlist', group='json'),
             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
 
-        def quality_to_int(s):
-            m = re.search('([0-9]+)', s)
-            if m is None:
-                return -1
-            return int(m.group(1))
-
         entries = []
         for sd in data_jsb:
             video_id, title = sd.get('id'), sd.get('title')
             if not video_id or not title:
                 continue
             video_id = compat_str(video_id)
-            formats = [{
-                'preference': -10 if fd['delivery'] == 'hls' else None,
-                'format_id': '%s-%s-%s' % (
-                    fd['delivery'], fd['quality'], fd['quality_string']),
-                'url': fd['src'],
-                'protocol': fd['protocol'],
-                'quality': quality_to_int(fd['quality']),
-            } for fd in sd['sources']]
+            formats = []
+            for fd in sd['sources']:
+                src = url_or_none(fd.get('src'))
+                if not src:
+                    continue
+                format_id_list = []
+                for key in ('delivery', 'quality', 'quality_string'):
+                    value = fd.get(key)
+                    if value:
+                        format_id_list.append(value)
+                format_id = '-'.join(format_id_list)
+                ext = determine_ext(src)
+                if ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        src, video_id, 'mp4', m3u8_id=format_id, fatal=False))
+                elif ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        src, video_id, f4m_id=format_id, fatal=False))
+                else:
+                    formats.append({
+                        'format_id': format_id,
+                        'url': src,
+                        'protocol': fd.get('protocol'),
+                    })
 
             # Check for geoblocking.
             # There is a property is_geoprotection, but that's always false
@@ -150,9 +161,6 @@ class ORFRadioIE(InfoExtractor):
         show_date = mobj.group('date')
         show_id = mobj.group('show')
 
-        if station == 'fm4':
-            show_id = '4%s' % show_id
-
         data = self._download_json(
             'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s' % (station, show_id, show_date),
             show_id
@@ -166,7 +174,8 @@ class ORFRadioIE(InfoExtractor):
                 'description': subtitle,
                 'duration': (info['end'] - info['start']) / 1000,
                 'timestamp': info['start'] / 1000,
-                'ext': 'mp3'
+                'ext': 'mp3',
+                'series': data.get('programTitle')
             }
 
         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
@@ -183,10 +192,10 @@ class ORFRadioIE(InfoExtractor):
 class ORFFM4IE(ORFRadioIE):
     IE_NAME = 'orf:fm4'
     IE_DESC = 'radio FM4'
-    _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
+    _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>4\w+)'
 
     _TEST = {
-        'url': 'http://fm4.orf.at/player/20170107/CC',
+        'url': 'http://fm4.orf.at/player/20170107/4CC',
         'md5': '2b0be47375432a7ef104453432a19212',
         'info_dict': {
             'id': '2017-01-07_2100_tl_54_7DaysSat18_31295',