You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Bevy merged a PR adding a repeat boolean to the KeyboardInput struct, corresponding to an equivalent concept in winit. When forwarding crossterm events we should correctly set this flag when reported by the terminal or we can reasonably identify a repeat event as part of our emulation strategy.
Holding off on a PR until events are refactored, just opening this to keep this bevy input change on the radar as I'm not sure if it will be mentioned in 0.15 migration materials.
(Side conversation: how should changes like this and the event refactor, that depend on unreleased changes in bevy, be handled with release-plz? Should a separate 0.15 branch be kept?)
The text was updated successfully, but these errors were encountered:
I'd prefer to keep it simple when possible (and avoid getting into the release engineering game). Try to avoid chasing unreleased features too hard like this unless there's a great need to do so. The problem with having multiple release branches is that it becomes cumbersome to track which one is accurate and keep them up to date, and backport fixes, and ....
Instead I'd generally leave a PR that depends on an upstream fix open until it can be merged after release.
This is not a hard and fast rule, but it's one that leads to less time spent.
Bevy merged a PR adding a
repeat
boolean to the KeyboardInput struct, corresponding to an equivalent concept inwinit
. When forwarding crossterm events we should correctly set this flag when reported by the terminal or we can reasonably identify a repeat event as part of our emulation strategy.Holding off on a PR until events are refactored, just opening this to keep this bevy input change on the radar as I'm not sure if it will be mentioned in 0.15 migration materials.
(Side conversation: how should changes like this and the event refactor, that depend on unreleased changes in bevy, be handled with release-plz? Should a separate 0.15 branch be kept?)
The text was updated successfully, but these errors were encountered: