Red Artist

You can try posting a video showing the problem. make sure to print events from https://docs.rs/egui_window_glfw_passthrough/0.8.1/egui_window_glfw_passthrough/struct.GlfwBackend.html#structfield.frame_events every frame to the terminal. It will help if we could see the events...

Sorry for the late reply :) I just got back from a trip too. you can just print to the terminal directly and make sure the video shows the terminal...

As I guessed in https://github.com/coderedart/egui_overlay/issues/37#issuecomment-2128586865 > Wayland doesn't support getting global cursor position (by design), so glfw just returns 0,0 as the global cursor position. This could be the problem....

I just checked, and with latest plasma wayland + xwayland, it works fine. It provides mouse position (and window position) as X11 is supposed to do. re-open the issue if...

pinging after 3 months :) hope this can be moved forward.

@bvssvni I think this issue can be closed, as it is more about opengl than glfw. There's other forums like graphics programming subreddit or its [discord community](https://github.com/graphicsprogramming).

Did you look at the current master branch code? It might already fix your issue. https://github.com/DioxusLabs/taffy/blob/main/src/tree/traits.rs#L169 `LayoutPartialTree` has an associated type bound by the `CoreStyle` trait. Its already implemented for...

This feature might require careful consideration. Anytime a parent node moves, then all its descendant nodes must update their relative_pos_to_root data. This can be expensive in certain cases. ### Scrolling...

I think this issue may be out of scope because glfw only supports calling `glfwSetWindowAttrib` on main-thread, while your example calls it on rendering thread (not main).