[ie/orf:on] Improve extraction (#9677)

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

diff --git a/yt_dlp/extractor/openrec.py b/yt_dlp/extractor/openrec.py
index 7546c12fb4e8bf4a03bbd862b79abf41db4adaf0..82a81c6c261e0f5704e2e1ee7ae14eafa818f3f0 100644 (file)
--- a/yt_dlp/extractor/openrec.py
+++ b/yt_dlp/extractor/openrec.py
@@ -12,6 +12,8 @@
 
 
 class OpenRecBaseIE(InfoExtractor):
+    _M3U8_HEADERS = {'Referer': 'https://www.openrec.tv/'}
+
     def _extract_pagestore(self, webpage, video_id):
         return self._parse_json(
             self._search_regex(r'(?m)window\.pageStore\s*=\s*(\{.+?\});$', webpage, 'window.pageStore'), video_id)
@@ -21,7 +23,7 @@ def _expand_media(self, video_id, media):
             if not m3u8_url:
                 continue
             yield from self._extract_m3u8_formats(
-                m3u8_url, video_id, ext='mp4', m3u8_id=name)
+                m3u8_url, video_id, ext='mp4', m3u8_id=name, headers=self._M3U8_HEADERS)
 
     def _extract_movie(self, webpage, video_id, name, is_live):
         window_stores = self._extract_pagestore(webpage, video_id)
@@ -35,8 +37,8 @@ def _extract_movie(self, webpage, video_id, name, is_live):
             raise ExtractorError(f'Failed to extract {name} info')
 
         formats = list(self._expand_media(video_id, get_first(movie_stores, 'media')))
-        if not formats and is_live:
-            # archived livestreams
+        if not formats:
+            # archived livestreams or subscriber-only videos
             cookies = self._get_cookies('https://www.openrec.tv/')
             detail = self._download_json(
                 f'https://apiv5.openrec.tv/api/v5/movies/{video_id}/detail', video_id,
@@ -50,8 +52,6 @@ def _extract_movie(self, webpage, video_id, name, is_live):
             formats = list(self._expand_media(video_id, new_media))
             is_live = False
 
-        self._sort_formats(formats)
-
         return {
             'id': video_id,
             'title': get_first(movie_stores, 'title'),
@@ -62,6 +62,7 @@ def _extract_movie(self, webpage, video_id, name, is_live):
             'uploader_id': get_first(movie_stores, ('channel', 'user', 'id')),
             'timestamp': int_or_none(get_first(movie_stores, ['publishedAt', 'time']), scale=1000) or unified_timestamp(get_first(movie_stores, 'publishedAt')),
             'is_live': is_live,
+            'http_headers': self._M3U8_HEADERS,
         }
 
 
@@ -112,8 +113,7 @@ def _real_extract(self, url):
             raise ExtractorError('Cannot extract title')
 
         formats = self._extract_m3u8_formats(
-            capture_data.get('source'), video_id, ext='mp4')
-        self._sort_formats(formats)
+            capture_data.get('source'), video_id, ext='mp4', headers=self._M3U8_HEADERS)
 
         return {
             'id': video_id,
@@ -124,6 +124,7 @@ def _real_extract(self, url):
             'uploader': traverse_obj(movie_store, ('channel', 'name'), expected_type=compat_str),
             'uploader_id': traverse_obj(movie_store, ('channel', 'id'), expected_type=compat_str),
             'upload_date': unified_strdate(capture_data.get('createdAt')),
+            'http_headers': self._M3U8_HEADERS,
         }