Re: bug in simplify_shallow

[Karim Belabas <Karim.Belabas@u-bordeaux.fr>]

* Bill Allombert [2023-09-30 13:27]:
> > P.S. In paridecl.h, comparing listpop:
> > 
> >  void    listpop(GEN L, long index);
> > 
> > shouldn't these now also be void?
> > 
> >  GEN     listinsert(GEN list, GEN object, long index);
> >  GEN     listput(GEN list, GEN object, long index);
> > 
> > (or does GEN cover that already?)
> 
> GP uses listinsert0 and listput0 which returns void.
> I have kept the return value for the C functions.

And in that case, it makes sense: it returns the 'clone' of the input
object, as was inserted in the list, which is not readily available to
the caller. 

- GP can't use this information (without lots of precautions).
- passing to these functions an input value which is alread a safe clone
  is wasteful since we must first validate that 'list' is an actual list
  in which we can insert things (and if not we might have a memory
  leak or have to clean up in the error handler...)

The current situation is fine.

Cheers,

    K.B.
-- 
Pr. Karim Belabas, U. Bordeaux, Vice-président en charge du Numérique
Institut de Mathématiques de Bordeaux UMR 5251 - (+33) 05 40 00 29 77
http://www.math.u-bordeaux.fr/~kbelabas/