Doubles on Slate

[Ken Causey]

Lee quickly added the basics of support to ExtLib to at least pass
double arguments.  See the C support added at

http://slate.tunes.org/cgi-bin/darcs.cgi/main/src/mobius/vm/platform/extprim.c?c=annotate

As he has pointed out adding support to return doubles is considerably
more involved.  He is of the opinion and I agree that it is in general a
waste of effort to pursue this at this time.

We all have to keep in mind that all in all Slate is still quite young
and is growing like a butterfly in stages.  Once upon a time Slate was
implemented on top of Common Lisp.  And then like the caterpillar that
it was it was transformed into the butterfly we have now of the pidgin-C
implementation.  Well, this is not the end of the story, it has longed
been planned that another complete metamorphosis would occur and a lot
of work must go into that.  The butterfly analogy breaks down here. ;)

Doing a look of fiddly work to interface with C seems valuable now.  But
in the next stage it will have much less value and a lot of time spent
on it now will be more or less wasted.

So the moral of my story is that if you (the plural 'you' meaning us all
as a community) see something that is needed and you feel that it is
worth your time to implement it, then go write ahead.  But understand
that it will probably have a limited lifetime.  And Lee and Brian need
to be able to spend their time thinking and building for the next stage
and can't afford to spend their time in this way.

And let me make it clear that I myself have spent time begging for
support like this in the past.  But I now resolve to try my best to
either keep quiet or dig in and investigate myself and resist the urge
to ask.

Ken

On Thu, 2005-05-12 at 13:13 -0300, Pupeno wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Je Ĵaŭdo Majo 12 2005 12:31, Pupeno skribis:
> > Do you think that there's some way to get ExtLib to handle Doubles without
> > implementing them on Slate ?
> Ken thought this wasn't clear enough, so, I'm adding my clarifications from 
> IRC:
> is it possible to add support for calling a function that expects a Double 
> (even if only the precistion of Float is available) ? Like the C compiler can 
> automatically cast a Float into a Double when needed.
> 
> - -- 
> Pupeno: pupeno at pupeno.com - http://pupeno.com
> Reading Science Fiction ? http://sfreaders.com.ar
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.1 (GNU/Linux)
> 
> iD8DBQFCg4CifW48a9PWGkURAvWXAJ4+8056dlCLeG76N9r310u7b1bafgCeNno8
> bL7qRf3g5jEMB7hUALyxHP8=
> =PP/H
> -----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
Url : /archives/slate/attachments/20050512/5ab56683/attachment.pgp