Re: [Gcl-devel] Two new tests for (SETF (VALUES ...) ...)

[Camm Maguire]

Greetings, and thanks as always for your work on these tests!

Should I take a look at fixing this, or did you want to try to get the
define-setf-method working here?  By the way, I had tried to use
defsetf to define a values method using the examples at the top of
lsp/setf.lsp.  I didn't spend too much time, but had difficulty in
getting the 'store-form' list to handle multiple values.  This is
still probably the best way to go, but if you'd rather not right now,
I think I can get the evaluation order in what we have working.

Just another note -- the setf implementation goes through a C wrapper
processing certain cases before the lisp is ever called.  I can't
determine whether this was by design or accretion, but in any case, setf
is not uniformly handled at present.

Take care,

"Paul F. Dietz" <address@hidden> writes:

> I added two new tests of (setf (values ...) ...).  They expose two
> bugs in the current implementation: the macro expansion uses setq
> where it should use setf, and the evaluation order of subforms
> is incorrect.
> 
> See SETF-VALUES.[45] in ansi-tests/places.lsp.
> 
>       Paul
> 
> 
> 
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gcl-devel
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah