Merge pull request #7045 from remitamine/ign

[youtube-dl] / youtube_dl / extractor / prosiebensat1.py

diff --git a/youtube_dl/extractor/prosiebensat1.py b/youtube_dl/extractor/prosiebensat1.py
index 22efa903fad195a71fe800e25c15a8262eb99d08..baa54a3afd10244ef2cda281c0785b24fe19e9cb 100644 (file)
--- a/youtube_dl/extractor/prosiebensat1.py
+++ b/youtube_dl/extractor/prosiebensat1.py
@@ -9,7 +9,9 @@ from ..compat import (
     compat_urllib_parse,
 )
 from ..utils import (
-    fix_xml_ampersands,
+    ExtractorError,
+    determine_ext,
+    float_or_none,
     int_or_none,
     unified_strdate,
 )
@@ -18,10 +20,15 @@ from ..utils import (
 class ProSiebenSat1IE(InfoExtractor):
     IE_NAME = 'prosiebensat1'
     IE_DESC = 'ProSiebenSat.1 Digital'
-    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at)|ran\.de|fem\.com)/(?P<id>.+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|the-voice-of-germany)\.(?:de|at|ch)|ran\.de|fem\.com)/(?P<id>.+)'
 
     _TESTS = [
         {
+            # Tests changes introduced in https://github.com/rg3/youtube-dl/pull/6242
+            # in response to fixing https://github.com/rg3/youtube-dl/issues/6215:
+            # - malformed f4m manifest support
+            # - proper handling of URLs starting with `https?://` in 2.0 manifests
+            # - recursive child f4m manifests extraction
             'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
             'info_dict': {
                 'id': '2104602',
@@ -219,10 +226,13 @@ class ProSiebenSat1IE(InfoExtractor):
             'ids': clip_id,
         })
 
-        videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
+        video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0]
 
-        duration = float(videos[0]['duration'])
-        source_ids = [source['id'] for source in videos[0]['sources']]
+        if video.get('is_protected') is True:
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        duration = float_or_none(video.get('duration'))
+        source_ids = [source['id'] for source in video['sources']]
         source_ids_str = ','.join(map(str, source_ids))
 
         g = '01!8d8F_)r9]4s[qeuXfP%'
@@ -295,15 +305,8 @@ class ProSiebenSat1IE(InfoExtractor):
                     'ext': 'mp4',
                     'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
                 })
-            elif 'f4mgenerator' in source_url:
-                manifest = self._download_xml(
-                    source_url, clip_id, 'Downloading generated f4m manifest',
-                    transform_source=lambda s: fix_xml_ampersands(s).strip())
-                for media in manifest.findall('./{http://ns.adobe.com/f4m/2.0}media'):
-                    manifest_url = media.get('href')
-                    if manifest_url:
-                        formats.extend(self._extract_f4m_formats(
-                            manifest_url, clip_id, f4m_id='hds'))
+            elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
+                formats.extend(self._extract_f4m_formats(source_url, clip_id))
             else:
                 formats.append({
                     'url': source_url,