On 5/5/07, Andreas Eder <aeder at arcor.de> wrote:
> > I think this is wrong. integration_constant_counter is a documented
> > variable. This used to do something with it. Now it doesn't and you
> > get some random integer appended to integrationconstant.
>
> Yes, you are right. integration_constant_counter is a documented
> variable, but you were not supposed to do something with it. Its
> only use was to generate a fresh, unique symbol each time - just
> what the gentemp does.
Well, GENTEMP is not exactly the same in this case, because
the integration_constant_counter is under control of the integration
code; omitting integration_constant_counter here causes a user-
visible change of behavior.
Andreas, I'm sorry to say it, but you really must be more careful.
Robert