[ie/crunchyroll] Fix stream extraction (#10005)

[yt-dlp.git] / yt_dlp / extractor / rutv.py

diff --git a/yt_dlp/extractor/rutv.py b/yt_dlp/extractor/rutv.py
index 66ac32debf4d2877648aca74dc52eed58cc05a15..d7f9a73377de50d89f1b879935d984ce0b84adb0 100644 (file)
--- a/yt_dlp/extractor/rutv.py
+++ b/yt_dlp/extractor/rutv.py
@@ -1,6 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import re
 
 from .common import InfoExtractor
@@ -23,6 +20,10 @@ class RUTVIE(InfoExtractor):
                         )
                         (?P<id>\d+)
                     '''
+    _EMBED_URLS = [
+        r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:test)?player\.(?:rutv\.ru|vgtrk\.com)/(?:iframe/(?:swf|video|live)/id|index/iframe/cast_id)/.+?)\1',
+        r'<meta[^>]+?property=(["\'])og:video\1[^>]+?content=(["\'])(?P<url>https?://(?:test)?player\.(?:rutv\.ru|vgtrk\.com)/flash\d+v/container\.swf\?id=.+?\2)',
+    ]
 
     _TESTS = [
         {
@@ -110,19 +111,6 @@ class RUTVIE(InfoExtractor):
         },
     ]
 
-    @classmethod
-    def _extract_url(cls, webpage):
-        mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:test)?player\.(?:rutv\.ru|vgtrk\.com)/(?:iframe/(?:swf|video|live)/id|index/iframe/cast_id)/.+?)\1', webpage)
-        if mobj:
-            return mobj.group('url')
-
-        mobj = re.search(
-            r'<meta[^>]+?property=(["\'])og:video\1[^>]+?content=(["\'])(?P<url>https?://(?:test)?player\.(?:rutv\.ru|vgtrk\.com)/flash\d+v/container\.swf\?id=.+?\2)',
-            webpage)
-        if mobj:
-            return mobj.group('url')
-
     def _real_extract(self, url):
         mobj = self._match_valid_url(url)
         video_id = mobj.group('id')
@@ -153,7 +141,7 @@ def _real_extract(self, url):
         if media['errors']:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, media['errors']), expected=True)
 
-        view_count = playlist.get('count_views')
+        view_count = int_or_none(playlist.get('count_views'))
         priority_transport = playlist['priority_transport']
 
         thumbnail = media['picture']
@@ -164,6 +152,7 @@ def _real_extract(self, url):
         duration = int_or_none(media.get('duration'))
 
         formats = []
+        subtitles = {}
 
         for transport, links in media['sources'].items():
             for quality, url in links.items():
@@ -181,25 +170,25 @@ def _real_extract(self, url):
                         'rtmp_live': True,
                         'ext': 'flv',
                         'vbr': str_to_int(quality),
-                        'quality': preference,
                     }
                 elif transport == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        url, video_id, 'mp4', quality=preference, m3u8_id='hls'))
+                    fmt, subs = self._extract_m3u8_formats_and_subtitles(
+                        url, video_id, 'mp4', quality=preference, m3u8_id='hls')
+                    formats.extend(fmt)
+                    self._merge_subtitles(subs, target=subtitles)
                     continue
                 else:
                     fmt = {
                         'url': url
                     }
                 fmt.update({
-                    'width': width,
-                    'height': height,
+                    'width': int_or_none(quality, default=height, invscale=width, scale=height),
+                    'height': int_or_none(quality, default=height),
                     'format_id': '%s-%s' % (transport, quality),
+                    'source_preference': preference,
                 })
                 formats.append(fmt)
 
-        self._sort_formats(formats)
-
         return {
             'id': video_id,
             'title': title,
@@ -208,5 +197,7 @@ def _real_extract(self, url):
             'view_count': view_count,
             'duration': duration,
             'formats': formats,
+            'subtitles': subtitles,
             'is_live': is_live,
+            '_format_sort_fields': ('source', ),
         }