0.4.0

[salam99823]

0.4.0 - 2025-01-06

Added

• Module prelude which contains all public structs and modules
  (5be223b).
• image dependency (e7fe636).
• Structures
  (e792bbb):
  • AbstractCollidersBuilder
  • AbstractCollider

Changed

• Upgrade dependencies
  (5c03df1)
  (786ab6d)
  (a4cc65e)
  (555756d):
  • edges 0.7
  • bevy 0.15
  • bevy_rapier2d 0.28
  • bevy_avian2d 0.2
• Dependencies bevy_math replaced by bevy
  (555756d).
• ColliderType moved to prelude
  (e792bbb).

Removed

• Modules
  (e792bbb):
  • avian2d
  • rapier2d
• Functions
  (e792bbb):
  • generate_collider
  • generate_colliders
• rayon from dependencies
  (e7fe636).

[shnewto]

oh weird, yeah the boulders just hang there 🤔

[shnewto]

So using convex_polyline colliders where we were using polyline colliders seems to be a workaround but hah I think we should figure out why polyline colliders don't do what we're expecting or understand why they're doing what they're doing if that's expected behavior 🤔

[shnewto]

Oh from the rapier docs

It is discouraged to use a triangle meshes or a polylines for colliders attached to dynamic rigid-bodies.

https://rapier.rs/docs/user_guides/bevy_plugin/colliders

[shnewto]

Ah okay, I think if you're able to make the changes to use convex_polyline for the boulders we're good with these changes. I don't think the behavior is a regression because in the current main branch the boulders are already using convex polyline colliders and if you use simple polyline colliders there it's the same behavior.

[salam99823]

Thanks for helping me find the problem.

[salam99823]

Looks like all the changes. Please do not delete dev branch.

[shnewto]

again @salam99823, thank you for the thoughtful work you put into your PRs and into this crate