Re: New NUM2INTEGRAL functions - unwanted polymorphism?

[Mikael Djurfeldt]

Rob Browning <address@hidden> writes:

> Michael Livshin <address@hidden> writes:
> 
> > good idea, except I don't know how to deprecate it nicely.  OTOH,
> > you may argue that it's a bugfix, and so the deprecation machinery
> > shouldn't be involved at all.
> 
> I'd tend to agree with that assesment, though if we went that route,
> it might be nice if we made sure the error msg made the problem clear.
> 
>   ERROR: scm_num2long - expected exact, got inexact.
> 
> or similar.

OK, removed.

And I added a message about missing the inexact in the 1.6 release due
to the changed behavior, as you suggested.

For 1.7 I removed the clause completely.  If we later decide to give
this kind of hint about what is expected, we should do it for all
functions and all types.

Best,
Mikael