[Bug-glpk] Re: Doesn't Solve A Simple Problem

[Andrew Makhorin]

> I tried solving a simple problem using GLPK on a MS Visual Studio 6.
>  I attached the code itself to the e-mail - the lines where I use you
>  library is in the Filter() function.
>  
> The rest of the code is rather sloppy because of the time limits but
>  it forms correct matrices which I put into an LPX structure. In the
>  enclosed example I try solving a simple problem of minimizing
>  
> Z=x3+x4
>  
> With the the variables constrained as follows:
> x1+00+x3+00<=10
> 00+x2+00+x4<=10
> -x1+00+x3+00<=-10
> 00-x2+00+x4<=-10
> 00+00-x3+00<=0
> 00+00+00-x4<=0
>  
> And the output is usually 
>       0:   objval =  0.000000000e+000   infeas =  1.000000000e+000 (0)
> PROBLEM HAS NO FEASIBLE SOLUTION
>  
> Here's as much as I got so far.

I solved your instance (in lp format) with no troubles:

minimize Z: x3+x4
subject to
x1    +x3     <= 10
    x2    +x4 <= 10
-x1   +x3     <=-10
   -x2    +x4 <=-10
      -x3     <=  0
      -x4     <=  0
end

lpx_read_cpxlp: reading problem data from `foo.lp'...
lpx_read_cpxlp: 6 rows, 4 columns, 10 non-zeros
lpx_read_cpxlp: 9 lines were read
lpx_simplex: original LP has 6 rows, 4 columns, 10 non-zeros
lpx_simplex: presolved LP has 4 rows, 4 columns, 8 non-zeros
lpx_adv_basis: size of triangular part = 4
      0:   objval =   0.000000000e+00   infeas =   1.000000000e+00 (0)
      2:   objval =   0.000000000e+00   infeas =   0.000000000e+00 (0)
OPTIMAL SOLUTION FOUND
Time used:   0.0 secs
Memory used: 0.1M (91828 bytes)

Try to check your code more carefully. (Since you are non-member,
your message was rejected and the attachment has been lost. You can
subscribe to the list visiting the page
http://lists.gnu.org/mailman/listinfo/bug-glpk )