Re: vecmax issue

["Ruud H.G. van Tol" <rvtol@isolution.nl>]

On 2026-09-15 12:54, Bill Allombert wrote:
> On Tue, Sep 15, 2026 at 12:37:06PM +0200, Ruud H.G. van Tol wrote:

> > ? version()
> > %1 = [2, 17, 4]
> >
> > ? vecmax([[0], [1]])
> >    ***   at top-level: vecmax([[0],[1]])
> >    ***                 ^-----------------
> >    *** vecmax: forbidden comparison t_VEC (1 elts) , t_VEC (1 elts).
> >    ***   Break loop: type 'break' to go back to GP prompt
> > break>
> >
> > Is that as expected?
>
> Yes.
>
> > ? vecsort([[0], [1]],,4)[1]
> > %2 = [1]
>
> ??vecmax
>   'largest refers to the ordinary real ordering ( <= )'
>
> ??vecsort
>   'If  cmpf  is  omitted,   we  use  the standard comparison function lex'
>
> This is not the same ordering.
OK, so how about adding a {cmpf} parameter to vecmax?

-- Ruud

P.S. I might have asked this before.
It just feels weird to me that vecmax and vecsort don't compare the same 
way by default.