Re: FFI crash: what am I missing?!

[Sam S. <sd...@gn...>]

> * In message <9F8582E37B2EE5498E76392AEDDCD3FE05C97186@G8PQD.blf01.telekom.de>
> * On the subject of "FFI crash: what am I missing?!"
> * Sent on Fri, 1 Aug 2003 11:27:17 +0200
> * Honorable "Hoehle, Joerg-Cyril" <Joe...@t-...> writes:
>
> Sam wondered what he missed:
> ... The obvious from http://norsys.com/onLineAPIManual/index.html
> "mesg should be a pointer to a character array
> which is allocated at least MESG_LEN_ns characters long."
> (mesg (c-ptr (c-array character 12345-or-whatever)) :out)

Jorg, thank you very much!!

> BTW2, if you declared the second parameter to new_netica as c-string
> (which is not it's type, it's environ_ns) just so that you can easily
> supply NULL, other people have succesfully used NIL as a type
> declaration, even though impnotes says it's output only.

but I won't be able to pass anything but NIL there, right?

> >1. why can't I pass NIL as a C-POINTER and have CLISP use NULL?!
> It would be possible to add such polymorphism, but I'm not much in
> favour of such things unless the behaviour can be made symmetric.
> 
> So the question becomes: what happens the other way round (e.g. as
> :return-type or as return from FOREIGN-VALUE)?
> 
> 1b. What should be the value of conversion from C-POINTER when the
> foreign pointer happens to be NULL?  Up to now, it is an object equalp
> to #<FOREIGN-ADDRESS 00000000>.

very interesting and enlightening discussion.
what do other FFIs do here?
CMUCL?  ACL?  LW?
what behavior does UFFI expect?

-- 
Sam Steingold (http://www.podval.org/~sds) running w2k
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
<http://www.mideasttruth.com/> <http://www.palestine-central.com/links.html>
The difference between genius and stupidity is that genius has its limits.