Help needed to understand problems in STK with Octave 6.1.0

[jbect]

Hi all,

There are many failing tests in STK with Octave 6.1, as reported in [1]
for Windows and (downstream) in [2] for Debian.

After some investigations, it seems to me that these failures are all
(or almost all ?) related to OOP-related regressions in Octave 6.1.0.

One of these regressions is described in [3, 4]: the dispatch mechanism
is broken when function handles are involved.

But there are other types of failures (type b and c in Comment #2 of
[1]) that I do not understand at all.  Some help would be much welcome.

@++
Julien


[1] https://savannah.gnu.org/bugs/?59340 
<https://savannah.gnu.org/bugs/?59340>

[2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=976206 
<https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=976206>

[3] https://savannah.gnu.org/bugs/?59631 
<https://savannah.gnu.org/bugs/?59631>

[4] https://savannah.gnu.org/bugs/?59617 
<https://savannah.gnu.org/bugs/?59617>

[Sébastien Villemot-2]

Hi Julien,

Le lundi 07 décembre 2020 à 09:56 +0100, Julien Bect a écrit :

> There are many failing tests in STK with Octave 6.1, as reported in [1]
> for Windows and (downstream) in [2] for Debian.
>
> After some investigations, it seems to me that these failures are all
> (or almost all ?) related to OOP-related regressions in Octave 6.1.0.
>
> One of these regressions is described in [3, 4]: the dispatch mechanism
> is broken when function handles are involved.
>
> But there are other types of failures (type b and c in Comment #2 of
> [1]) that I do not understand at all.  Some help would be much welcome.

FWIW, I’ve reported a classdef-related regression that looks similar to
your “case b”:
https://savannah.gnu.org/bugs/?59661

Best,

--
⢀⣴⠾⠻⢶⣦⠀  Sébastien Villemot
⣾⠁⢠⠒⠀⣿⡁  Debian Developer
⢿⡄⠘⠷⠚⠋⠀  https://sebastien.villemot.name
⠈⠳⣄⠀⠀⠀⠀  https://www.debian.org