]> git.bitcoin.ninja Git - youtube-dl/commitdiff
[mediaset] Fix issues and extract all formats (closes #16568)
authorSergey M․ <dstftw@gmail.com>
Fri, 29 Jun 2018 19:16:44 +0000 (02:16 +0700)
committerSergey M․ <dstftw@gmail.com>
Fri, 29 Jun 2018 19:19:02 +0000 (02:19 +0700)
youtube_dl/extractor/mediaset.py

index 76a2ae1254717bbf9f2bfa7cfb883251987c8ed3..9f2b60dcc5b62f83800101822bfb3da79d43badc 100644 (file)
@@ -10,7 +10,6 @@ from ..utils import (
     parse_duration,
     try_get,
     unified_strdate,
-    ExtractorError
 )
 
 
@@ -58,7 +57,7 @@ class MediasetIE(InfoExtractor):
             'series': 'Matrix',
             'categories': ['infotainment'],
         },
-        'expected_warnings': ['is not a supported codec'],
+        'expected_warnings': ['HTTP Error 403: Forbidden'],
     }, {
         # clip
         'url': 'http://www.video.mediaset.it/video/gogglebox/clip/un-grande-classico-della-commedia-sexy_661680.html',
@@ -87,13 +86,14 @@ class MediasetIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        media_info = self._download_json(
+        video = self._download_json(
             'https://www.video.mediaset.it/html/metainfo.sjson',
             video_id, 'Downloading media info', query={
                 'id': video_id
             })['video']
 
-        media_id = try_get(media_info, lambda x: x['guid']) or video_id
+        title = video['title']
+        media_id = video.get('guid') or video_id
 
         video_list = self._download_json(
             'http://cdnsel01.mediaset.net/GetCdn2018.aspx',
@@ -104,12 +104,17 @@ class MediasetIE(InfoExtractor):
 
         formats = []
         for format_url in video_list:
-            if '.ism' in format_url:
-                try:
-                    formats.extend(self._extract_ism_formats(
-                        format_url, video_id, ism_id='mss', fatal=False))
-                except ExtractorError:
-                    pass
+            ext = determine_ext(format_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', fatal=False))
+            elif ext == 'ism' or '.ism' in format_url:
+                formats.extend(self._extract_ism_formats(
+                    format_url, video_id, ism_id='mss', fatal=False))
             else:
                 formats.append({
                     'url': format_url,
@@ -117,24 +122,23 @@ class MediasetIE(InfoExtractor):
                 })
         self._sort_formats(formats)
 
-        title = media_info['title']
-
         creator = try_get(
-            media_info, lambda x: x['brand-info']['publisher'], compat_str)
+            video, lambda x: x['brand-info']['publisher'], compat_str)
         category = try_get(
-            media_info, lambda x: x['brand-info']['category'], compat_str)
+            video, lambda x: x['brand-info']['category'], compat_str)
         categories = [category] if category else None
 
         return {
             'id': video_id,
             'title': title,
-            'description': media_info.get('short-description'),
-            'thumbnail': media_info.get('thumbnail'),
-            'duration': parse_duration(media_info.get('duration')),
+            'description': video.get('short-description'),
+            'thumbnail': video.get('thumbnail'),
+            'duration': parse_duration(video.get('duration')),
             'creator': creator,
-            'upload_date': unified_strdate(media_info.get('production-date')),
-            'webpage_url': media_info.get('url'),
-            'series': media_info.get('brand-value'),
+            'upload_date': unified_strdate(video.get('production-date')),
+            'webpage_url': video.get('url'),
+            'series': video.get('brand-value'),
+            'season': video.get('season'),
             'categories': categories,
             'formats': formats,
         }