You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed a regression regarding the handling of set variables and I have two examples to demonstrate the issue.
First example:
var bool: c;
var set of int: x = if c then 1..3 else 1..5 endif;
var set of int: y = if not c then 1..3 else 1..5 endif;
constraint c = true;
constraint x subset y;
MiniZinc 2.7.3 returns the following solution:
c = true
x = 1..3
y = 1..5
----------
MiniZinc 2.7.4, however, produces this error message:
Error: evaluation error: not an integer expression
Second example:
include "alldifferent.mzn";
int: N = 4;
array [1..N] of var set of 1..N: x;
constraint alldifferent(x);
constraint array_union(x) = 1..(N div 2);
MiniZinc 2.7.3 returns the following solution:
x = [{}, 2..2, 1..1, 1..2]
MiniZinc 2.7.4, however, produces the above error message.
The problem does not occur when optimization is disabled (with -O0) or Chuffed is used for solving.
The text was updated successfully, but these errors were encountered:
I noticed a regression regarding the handling of set variables and I have two examples to demonstrate the issue.
First example:
MiniZinc 2.7.3 returns the following solution:
MiniZinc 2.7.4, however, produces this error message:
Second example:
MiniZinc 2.7.3 returns the following solution:
MiniZinc 2.7.4, however, produces the above error message.
The problem does not occur when optimization is disabled (with
-O0
) or Chuffed is used for solving.The text was updated successfully, but these errors were encountered: