How do I obtain the variables in a JuMP constraint?

Specific Domains Optimization (Mathematical)
,
1

Say I define a constraint with JuMP:

using JuMP
model = Model();
@variable(model, x[1:3]);
constraint = @constraint(model, c1, sum(x) <= 1)

How can I now, only using the variable constraint, obtain an iterable of variables present in this constraint?

2

Hi @BdeKoning, welcome to the forum,

There is no explicit function you can call, but if the constraint is linear (JuMP represents linear constraints as an affine function with type AffExpr), you can use:

julia> using JuMP

julia> model = Model();

julia> @variable(model, x[1:3]);

julia> constraint = @constraint(model, c1, sum(x) <= 1)
c1 : x[1] + x[2] + x[3] ≤ 1

julia> object = constraint_object(constraint)
ScalarConstraint{AffExpr, MathOptInterface.LessThan{Float64}}(x[1] + x[2] + x[3], MathOptInterface.LessThan{Float64}(1.0))

julia> object.func
x[1] + x[2] + x[3]

julia> typeof(object.func)
AffExpr (alias for GenericAffExpr{Float64, GenericVariableRef{Float64}})

julia> variables = [k for k in keys(object.func.terms)]
3-element Vector{VariableRef}:
 x[1]
 x[2]
 x[3]
2 Likes