[ODE] About "malloc" patch

[Jeffrey Smith]

Anton Savin wrote:
> I suppose this is what is called "a patch that enables use of dynamic
> memory". But who is shure that it's bug free? 

I'm fairly sure, but until I can run Purify on the code (soon) I won't
be positive.  If you find any bugs, let me know right away.

> Has actually anyone tested all these code paths?

As far as possible, I've tested all the code paths (the code is
actually pretty simple, structurally).

> Shortly, why not use smart pointers?

That's the plan; this implementation was a just temporary measure.  If
you want to reimplement it using auto_ptr, go for it!

-jeff