doggy

The version number should be bumped to v1.4.3 I think.

@TozyZuo Yeah, you're right!! Some code for testing the class is needed after trimming aspects' suffix on that class. I can handle it for several days, I think. Thanks for...

Class statedClass = self.class; Class actualClass = object_getClass(self); It should be named as above I think. P.S. You can take a look at PR #115 if you need KVO coexistence.

New PR created to RAC Take a look and move further discussion to [New Issue Here](https://github.com/ReactiveCocoa/ReactiveCocoa/issues/3280)

Yup, I know the KVO logic that also created a dynamic subclass on the fly. This patch has been tested in our [app project](https://itunes.apple.com/app/id1063183999) (with waxPatch, jsPatch and KVO logic...

> I am not quite sure if the implementation is correct, since [self class] could be a dynamic subclass created by RAC, which results in the newForwardInvocation IMP itself being...

I checked `NSObject+RACDescription.m` again and I doubt that `[self class]` returns the original class rather than 'dynamic subclass'. The logic exists in RACSwizzleMethodSignatureForSelector() already. Both KVO and RAC dynamic subclass...

@andersio Test case committed. Protection logic also committed.

> do all the swizzling at launch time before rac_signalForSelector is used > This scenario you mentioned is fine even without this PR. AFAIK, ObjC IMP swizzling was handled by...

@andersio I have deployed this patch in our develop environment several days.