[r6rs-discuss] "Unspecified"

[Sam Tobin-Hochstadt]

On Mon, 2006-10-02 at 05:38 -0400, Abdulaziz Ghuloum wrote:
> On Oct 2, 2006, at 4:27 AM, Anton van Straaten wrote:
> 
> > Trent Buck wrote:
> >> On Fri, Sep 22, 2006 at 01:41:50AM +0000, John Cowan wrote:
> >>> Summary:  Editors, "the unspecified value" won't fly.  Pick something
> >>> else and insist on it.
> >>> I had proposed "the content-free value" earlier; I would also propose
> >>> "the useless value".
> >> Seriously, my personal preference is for `the meaningless value'.
> >
> > It was pointed out that such a value has meaning
> 
> Seriously though, why are we discussing an unspecified, content-free,
> useless, and meaningless value, when we have the option of not having
> a value at all to begin with?  What's so wrong with having vector-set!
> return no value?

Then we would be unable to use `vector-set!' in many contexts that
require a value, such as the following refactoring:

(for-each (lambda (x) (vector-set! v x)) indexes)

-->

(for-each/logging (lambda (x) (vector-set! v x)) indexes)

where:

(define (for-each/logging f l)
   (for-each (lambda (v) (display v) (newline) (f v)) l))

Now we have an error if `vector-set!' returns no values.

sam th