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
This was seen and worked-around in issue #39 . The problem is that if a mutating method calls another mutating method on one of its item arguments, the latter won't be detected as a mutation if the argument's method is package-private.
Given this is a natural way of enforcing relationship invariants on item that can refer to each other cyclically, we should support this.
The text was updated successfully, but these errors were encountered:
Actually, this is supported already by ByteBuddy - the problem is that an override can't be defined by ByteBuddy unless the proxy class exists in the same package as the method being overridden / defined. Need to think about this - what if a class inherits several interfaces that define package-private methods, but the interfaces are in separate packages?
Even then, I'm unsure as to whether ByteBuddy will forward a call to another object from the package-private override it creates. It seems logical that it can do this, but...
This was seen and worked-around in issue #39 . The problem is that if a mutating method calls another mutating method on one of its item arguments, the latter won't be detected as a mutation if the argument's method is package-private.
Given this is a natural way of enforcing relationship invariants on item that can refer to each other cyclically, we should support this.
The text was updated successfully, but these errors were encountered: