Ce site est optimisé pour être consulté depuis un navigateur moderne dans lequel JavaScript est activé.

Bug in spBone_parentToWorld in spine-c

MintPaw

On line 344: EsotericSoftware/spine-runtimesblob/4.2/spine-c/spine-c/src/spine/Bone.c#L344

The result of the if statement often causes crashes.
Maybe it's been reported already? But I don't see a newer version.

Spinebot

The issue you are experiencing with spBone_parentToWorld in the spine-c runtime has not been explicitly mentioned in the provided contexts. However, there are several instances of bugs and issues being reported and tracked in the Spine runtimes, such as the root scale issue and other runtime errors. It would be advisable to check the Spine runtimes' GitHub repository for any existing issues related to spBone_parentToWorld or to report the issue if it hasn't been documented yet.

Nate

If if (self->parent != NULL) { crashes then self is null.

SilverStraw

So, it doesn't seem like a bug?

Nate

It's not possible to say where the bug is from the information provided. If it's runtime code passing null to spBone_parentToWorld, that'd be a bug. If it's user code that is passing null, then there is a bug in user code.

Mario

@MintPaw without assets and a minimal amount of code that reproduces the issue, we can't tell if that's a bug in spine-c, or a bug in user code. Happy to assist, if you can provide us with that info.

MintPaw

Sorry, I didn't get notified of updates here till now.

If you pass NULL then yes it crashes. But also if you pass in the root bone, because it has no parent.

If parent is NULL then it calls spBone_localToWorld with the NULL, which immediately crashes inside.

I think it probably should be like worldToParent two functions above. if (self->parent == NULL) { instead of if (self->parent != NULL) {

But maybe this is the intention of the API somehow?

Mario

@MintPaw ohhh, yes, that is indeed a bug! I've fixed this in the 4.2 branch. Thank you for reporting!