Re: [BUG 2.5.0–CURRENT] taille() and taille2() are mixed up in pariold.h

[Ilya Zakharevich <nospam-abuse@ilyaz.org>]

On Wed, Sep 12, 2018 at 03:41:02PM +0200, Karim Belabas wrote:
> >   [BUG 2.5.0–CURRENT] taille() and taille2() are mixed up in pariold.h
> 
> Confirmed; now fixed in 'master'.
> 
> > (Math::Pari still crashes, so something else may be busted too???)
> 
> I guess you're trying to build with pari-2.11.

No, with 2.5.0.

> What is the last version of libpari Math::Pari was known to support  ?

2.3.5.

In 2.4.*, all the hooks allowing use from an external interpreter were
removed.  I have been trying to make 2.5.0 work; I found some places where
the required hooks may be installed.  However, it took many days
(which I do not have!) and still only very primitive functionality works.

Anyway, I cannot test how well 2.5.0 works (even “the part which may
be supported already”): even the hooks in the test suite were removed!

  (One needs «\e» to make the test suite parsable “from outside”, and
   this is missing in all the new test files.)

Yours,
Ilya

P.S.  My conjecture now is that the major missing part is “How to make
      «type=I» and «type=E» arguments”.  (But, as I explained above, I
      cannot check that “the rest” is working already…) 

      Suppose I have a C function
        GEN mysummand(GEN I, void *mydata);
      It is completely unclear how can I make out of it something to
      pass to (e.g.) somme() to make an analogue of
        sum(I=1,3,mysummand(I))