[vlc-commits] [Git][videolan/vlc][master] vout_subpictures: keep the last-minute rendered text region

[Steve Lhomme (@robUx4)]

Steve Lhomme pushed to branch master at VideoLAN / VLC


Commits:
e51246cb by Steve Lhomme at 2023-11-16T12:52:08+00:00
vout_subpictures: keep the last-minute rendered text region

It was leaking since fe415420f3ace0b91636884626ff1ddd6ef8c360 and would
have to be rendered again at the next call.
This is the same thing spu_PrerenderText() does.

- - - - -


1 changed file:

- src/video_output/vout_subpictures.c


Changes:

=====================================
src/video_output/vout_subpictures.c
=====================================
@@ -818,7 +818,9 @@ static subpicture_region_t *SpuRenderRegion(spu_t *spu,
         if ( rendered_text  == NULL)
             // not a rendering error for Text-To-Speech
             return NULL;
-        // FIXME notify the caller it is allocated
+        // replace the text region with the rendered region
+        vlc_list_replace(&region->node, &rendered_text->node);
+        subpicture_region_Delete(region);
         region = rendered_text;
     }
 



View it on GitLab: https://code.videolan.org/videolan/vlc/-/commit/e51246cb67ca33731fb7defacfca4254dfa0c6e6

-- 
View it on GitLab: https://code.videolan.org/videolan/vlc/-/commit/e51246cb67ca33731fb7defacfca4254dfa0c6e6
You're receiving this email because of your account on code.videolan.org.


VideoLAN code repository instance