[Bug-glpk] generating LP from mathprog models

[Sébastien de Menten]

Hi,

I ran a test for reading a very simple LP model in the new mathprog language 
that looks like:
"""
set t := 1..T;
var x{time};
minimize y:
       sum { t in time } x[t];
data;
end;
"""
where I replaced T by different values and noticed an increase in generation 
time which looks exponential in T:
T       Time for generation (in seconds)
0       0
1000    0.23
2000    0.92
3000    2.24
4000    4.08
5000    7.7
6000    11.99
7000    17.92
8000    24.66
9000    33.17
10000   43.19
11000   53.54
12000   70.89

Is it a "bug" ?
Is there a solution for avoiding this increase in time ?

Sebastien

ps: thank again for this great software :-)

_________________________________________________________________