> I'm currently dealing with the fact that it seems that you can no longer get the name
> of a button that was given in the port definition macro for "core" input port types
> unless you have a running machine. Which makes no sense.
In particular, this seems like pretty bad C++ style:
(from device.h, device_t definition):
public:
// getters
running_machine &machine() const { assert(m_machine != NULL); return *m_machine; }
Returning a value as a reference when it is valid for the value to be NULL (as it is before the device has had set_machine() called on it, which is true before a machine has been created and set into the device by set_machine()), but using an assert, is pretty bad.
|
Index 
Flat 
Re: IPT_TILT1 through IPT_TILT4 - is there any reason for these?
Reply
