Yalmip incorrectly determines equality constraints to be trivially true

[alastair-abbott]

There appears to be a bug in how yalmip determines equality constraints to be trivially true.

For example:

>> a = sdpvar(1); optimize([a,0] == [a,1],0)
Warning: Equality constraint evaluated to trivial true. 
> In constraint (line 45)
In  ==  (line 13) 

When the constraint is clearly trivially false, not true.

The error appears to be on line 44 of constraint.m. It should check all(Z == 0) rather than all(Z) == 0

[johanlofberg]

Right, thanks.