pull the common formats info outside for second `for` loop

yt_dlp/extractor/xiaohongshu.py

@@ -36,11 +36,7 @@ class XiaoHongShuIE(InfoExtractor):
 
         formats = []
         for info in video_info:
-            for url in traverse_obj(info, (('mediaUrl', ('backupUrls', ...)), {url_or_none})):
-                formats.append({
-                    'url': url,
-                    'ext': 'mp4',
-                    **traverse_obj(info, {
+            format_info = traverse_obj(info, {
                 'fps': 'fps',
                 'width': 'width',
                 'height': 'height',
@@ -54,7 +50,9 @@ class XiaoHongShuIE(InfoExtractor):
                 'filesize': 'size',
                 'duration': ('duration', {lambda x: float_or_none(x, scale=1000)})
             })
-                })
+
+            formats.extend(traverse_obj(info, (('mediaUrl', ('backupUrls', ...)), {
+                lambda url: url_or_none(url) and {'url': url, 'ext': 'mp4', **format_info}})))
 
         thumbnails = []
         for image_info in traverse_obj(note_info, ('imageList', ...)):