🚀 go-pugleaf

This problem has been fixed, and should not exist in the release after
next of gcc.

Yes, it could be added, but Richard Stallman does not want the
compiler to do this.  In our comercial product of gcc, we by default
let the user know where in the user source code the code lies that the
compiler dies on.  By default, gcc does not give you a clue what in
the user source code caused the fault.

If you would like to see this functionality added, please him
<rms@prep.ai.mit.edu> know that the behaviour is desirable.

> Message-Id: <9210121639.AA01071@solace.me.uiuc.edu>
> From: Mike Tiller <tiller@solace.me.uiuc.edu>
> To: bug-g++@prep.ai.mit.edu
> Date: Mon, 12 Oct 92 11:39:45 -0500

> The following code although erroneous should trigger an error not a
> segmentation fault.  This brings me to an additional point...would
> it be possible to add a debugging flag to g++ and gcc so you can
> find out what line they are one when they die this most horrible
> kind of death.  Most of the time it is an error that compiler can't
> digest and would be easily recognizable.  In other words, most of
> the time it is something I could fix...If I only knew where it was.

> template<class T>
> class	Array
> {
> 	int	size;
> };

> enum	Source { Scalar, Array, Function };