Consolatis

> One change I want to suggest however, is to **not** fallback to the default config in the case of a configuration error. I tended to experiment a lot when...

> Honestly while looking through the parser's code I did find the logic somewhat messy. I would not mind doing a larger rewrite, however as you also mentioned this may...

Is this an X11 or a wayland native window? (`[xdg-shell]` or `[xwayland]` when pressing `A-Tab`.)

> It looks like the limit is by all buttons widths summed up, with or without decoration. Correct, the `without decorations` part is still an open issue. This specific minimal...

The usage of the scaled-scene-buffer API looks good to me. Didn't check cairo and menu. Commit message states that docs aren't added yet but just to have it mentioned, IMHO...

I've added the scale check, split out the generic changes on top and rebased on your branch. There might be better options to store the scale along the `lab_data_buffer` /...

I really like that approach! Especially with having the layer 2 cache in the impl and under full control of `wlr-scene-buffer.c`. Will play around with it a bit tonight. Didn't...

Had a chance to play around with your branch a bit: https://github.com/labwc/labwc/compare/master...Consolatis:labwc:sketch/scaled_rect_buffer_reuse_tokyo4j Overall I think its a great design, do you want to open a PR for it? I've removed...

> One thing I'm worried with this approach is that, if we apply scaled-scene-buffer to window decorations, the buffers are released when all the window are closed and recreated every...

Sorry for derailing this PR so much with the cache discussion. Appreciate the new PR and I think it makes it easier to review.