On Sun, Jul 8, 2012 at 1:43 PM, Sébastien Villemot <
sebastien.ville...@ens.fr> wrote:
> Jordi Gutiérrez Hermoso <jord...@octave.org> writes:
>
> > On 8 July 2012 08:25, Sébastien Villemot <sebastien.ville...@ens.fr>
> wrote:
> >> What happens is that OpenJDK 7 changes the locale settings and, in
> >> locale environments where the decimal separator is a comma and not a
> >> point, that makes Octave crash when reading a decimal number (as simple
> >> as "0.1").
> >
> > A change in locale should not make Octave crash. Do you have a
> > stripped-down example of how to reproduce this? I tried:
> >
> > jordi@Iris:~$ echo 0.1 > foo
> > jordi@Iris:~$ LANG=fr_FR.UTF-8 octave --eval 'load foo'
> >
> > and I did not see a crash.
>
> Your example does not trigger a crash, of course.
>
> But OpenJDK is ran from an Oct-file, and I guess it bypasses some sanity
> checks and cleanups done by Octave. Then the sscanf() function used from
> the Octave parser becomes unable to parse "0.1".
>
>
That's good that you've found a workaround, still I'd like to understand
what is crashing and why. From the debian bug report, I understand that
starting octave as "LANG=fr_FR.utf-8 octave" is the way to make it crash.
But what commands in octave do I need to use to reproduce the crash?
I will consider integrating the debian patch into main java package source
tree, but I'd like first to understand what's going on.
Michael.
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Octave-dev mailing list
Octave-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/octave-dev
