This commit introduces the Rotate Euler function node which modifies an input euler rotation. Addresses T91653.
Details
- Reviewers
Jacques Lucke (JacquesLucke) Hans Goudey (HooglyBoogly) - Maniphest Tasks
- T91375: Rotate Euler node
- Commits
- rBc5c94e3eae74: Geometry Nodes: Add Rotate Euler Node
Diff Detail
- Repository
- rB Blender
- Branch
- fn_rotate_euler_node_3 (branched from master)
- Build Status
Buildable 17494 Build 17494: arc lint + arc unit
Event Timeline
| source/blender/nodes/function/nodes/node_fn_rotate_euler.cc | ||
|---|---|---|
| 33 | No need to specify socket identifiers (second input) when the names are different. | |
| 37 | Put everything except register_node_type_fn_rotate_euler in the namespace and remove the unnecessary blender:: prefixes. | |
| 41 | What is the operation? | |
| 43 | This function is not necessary. | |
| 58 | This is dead code. | |
| source/blender/nodes/function/nodes/node_fn_rotate_euler.cc | ||
|---|---|---|
| 32 | The identifier can be removed here as well. Identifiers only need to be unique on one side of the node (e.g. among all inputs). | |
| 44 | You are treating the rotation input as vector, not as a euler. Changing an euler with such a matrix multiplication does not seem to make sense, does it? You'll probably have to convert both eulers to a matrix, then perform a matrix multiplication and then convert back. | |
| source/blender/nodes/function/nodes/node_fn_rotate_euler.cc | ||
|---|---|---|
| 44 | Thanks. I was going to bring this up later today after work (I submitted this patch a bit prematurely--I tried to abort from arc diff but it went through instead). Rotating it did not seem to make any sense, but this explains a lot why. | |
I'm a bit confused on how this is different than the Euler rotation type on the vector rotate node?
I'm a bit confused on how this is different than the Euler rotation type on the vector rotate node?
The difference is that this is changing an euler value, while the Vector Rotate node is changing a vector. The difference would be way more clear when we had a dedicated Euler socket. We will think again about adding that for 3.1, it's probably a bit late for 3.0 now.
