Re: Gnumeric solver bug?

[Jody Goldberg <jody gnome org>]

On Fri, May 24, 2002 at 05:03:51PM +0200, Michal Kaukic wrote:
Content-Description: Main text
> We attempted to solve folloving one-row LP problem:
> 
> x1+2*x2+3*x3 -> max
> subject to
> x1+x2+3*x3 <= 8
> x1,x2,x3 >= 0.
> 
> Gnumeric crashed with Segmentation fault
> for initial solutions e.g. x=(0,0,0),(8,0,0)
> and also for optimal solution x=(0,8,0).
> 
> We want to use Gnumeric instead of Excel in
> teaching of operation research topics.
> 
> The Gnumeric worksheet we used is attached.
> Maybe we made some mistakes? Or it is a bug
> in Gnumeric? We used Gnumeric 1.0.6 on Debian
> (woody/sid).

Thanks for the test case.  It exposed a couple of simple mistakes in
the solver's setup.  Things will work smoothly for 1.0.7.