Re: Wrong prototype for listcreate()

[Bill Allombert <Bill.Allombert@math.u-bordeaux.fr>]

On Mon, May 25, 2015 at 08:15:41PM +0200, Jeroen Demeyer wrote:
> On 2015-05-24 17:29, Bill Allombert wrote:
> >On Thu, Apr 30, 2015 at 12:21:24PM +0200, Jeroen Demeyer wrote:
> >>Hello,
> >>
> >>I know listcreate() is redundant, obsolete and shouldn't be used,
> >>but at least the prototype should be correct: it still says D0,L,
> >>but the argument was dropped from the libPARI function listcreate().
> >
> >Alas, we cannot change the prototype of listcreate, because its whole purpose
> >is to allow legacy code like L=liscreate(n) to work, and changing the GP
> >prototype would break that.
> Then why did you change the libpari function?

listcreate was not part of the documented PARI API:
the documentation stated "This function is useless in library mode."
so we did not have to maintain the PARI compatibility.

Anyway, it seems like you are confronted to a practical problem that you are
not communicating to this list.

Cheers,
Bill.