Re: gh_ interface (Re: guile: going the way of DEATH)

[Russ McManus <mcmanr@eq.gs.com>]

Telford Tendys <telford@triangle.triode.net.au> writes:

> > >    On the opposite, Guile was initially based on one implementation of
> > >    Scheme (SCM) with some pieces borrowed from another one. This led
> > >    to a horrible mix of interfaces and interdependencies in the code.
> > >    Also, nobody even tried to define both Scheme and C API - they are
> > >    still in flux and dirty, lacking logical order.
> > 
> > This is a factual error.  Mark Galassi took considerable care when
> > designing the gh_ API, which is the application writer's C level
> > interface to Guile.  This interface has changed very little since it
> > was defined.  And IMHO the gh_ interface is logical.  Please share a
> > contradictory example.
> 
> You are writing C, you have a variable of type SCM and you want
> to convert it to a double. You use gh_scm2double() and if the SCM
> was a number then it will convert but if it was something else then
> it will error trap and you don't want that.
> 
> So you check the value first with gh_inexact_p() but then you realise
> that someone might send you an integer which you still want to convert
> to a real so you have to check again with gh_exact_p(). Not a huge
> demand I realise but I still use scm_number_p() instead because it is
> one call instead of two <shrug>.

Um, OK.  I agree that scm_number_p() is handy, and should have a
parallel in the gh_ api.

I don't think this makes the gh_ api *illogical*, though.  ;^)

-russ

--
I have found that all ugly things are made by those who strive to make
something beautiful and that all beautiful things are made by those
who strive to make something useful.
             --Oscar Wilde