Update to MOI v0.8 (#83)

Author: odow

REQUIRE

@@ -1,3 +1,3 @@
 julia 0.6
-MathOptInterface 0.7 0.8
+MathOptInterface 0.8 0.9
 Compat 0.59

src/LinQuadOptInterface.jl

@@ -283,7 +283,7 @@ function MOI.is_empty(m::LinQuadOptimizer)
     ret = ret && m.name == ""
     ret = ret && m.obj_type == AffineObjective
     ret = ret && isa(m.single_obj_var, Nothing)
-    ret = ret && m.obj_sense == MOI.MinSense
+    ret = ret && m.obj_sense == MOI.MIN_SENSE
     ret = ret && m.last_variable_reference == 0
     ret = ret && isempty(m.variable_mapping)
     ret = ret && isempty(m.variable_names)
@@ -302,9 +302,9 @@ function MOI.is_empty(m::LinQuadOptimizer)
     ret = ret && isempty(m.constraint_names)
     ret = ret && isempty(m.constraint_names_rev)
     ret = ret && m.objective_constant == 0.0
-    ret = ret && m.termination_status == MOI.OptimizeNotCalled
-    ret = ret && m.primal_status == MOI.NoSolution
-    ret = ret && m.dual_status == MOI.NoSolution
+    ret = ret && m.termination_status == MOI.OPTIMIZE_NOT_CALLED
+    ret = ret && m.primal_status == MOI.NO_SOLUTION
+    ret = ret && m.dual_status == MOI.NO_SOLUTION
     ret = ret && m.primal_result_count == 0
     ret = ret && m.dual_result_count == 0
     ret = ret && m.solvetime == 0.0
@@ -318,7 +318,7 @@ function MOI.empty!(m::M, env = nothing) where M<:LinQuadOptimizer
     m.obj_type = AffineObjective
     m.single_obj_var = nothing
     # we assume the default is minimization
-    m.obj_sense = MOI.MinSense
+    m.obj_sense = MOI.MIN_SENSE
 
     m.last_variable_reference = 0
     m.variable_mapping = Dict{MathOptInterface.VariableIndex, Int}()
@@ -344,9 +344,9 @@ function MOI.empty!(m::M, env = nothing) where M<:LinQuadOptimizer
 
     m.objective_constant = 0.0
 
-    m.termination_status = MathOptInterface.OptimizeNotCalled
-    m.primal_status = MathOptInterface.NoSolution
-    m.dual_status = MathOptInterface.NoSolution
+    m.termination_status = MathOptInterface.OPTIMIZE_NOT_CALLED
+    m.primal_status = MathOptInterface.NO_SOLUTION
+    m.dual_status = MathOptInterface.NO_SOLUTION
     m.primal_result_count = 0
     m.dual_result_count = 0
 

src/mockoptimizer.jl

@@ -63,9 +63,9 @@ mutable struct MockLinQuadModel # <: LinQuadOptInterface.LinQuadOptimizer
 
         m.sos = LinQuadSOS[]
 
-        m.termination_status = MOI.OptimizeNotCalled
-        m.primal_status = MOI.NoSolution
-        m.dual_status = MOI.NoSolution
+        m.termination_status = MOI.OPTIMIZE_NOT_CALLED
+        m.primal_status = MOI.NO_SOLUTION
+        m.dual_status = MOI.NO_SOLUTION
 
         m.variable_primal_solution = zeros(0)
         m.variable_dual_solution = zeros(0)
@@ -288,9 +288,9 @@ function set_solution!(instance::MockLinQuadOptimizer;
     constraint_dual = Float64[],
     quadratic_primal = Float64[],
     quadratic_dual = Float64[],
-    termination_status = MOI.Optimal,
-    primal_status = MOI.FeasiblePoint,
-    dual_status = MOI.FeasiblePoint
+    termination_status = MOI.OPTIMAL,
+    primal_status = MOI.FEASIBLE_POINT,
+    dual_status = MOI.FEASIBLE_POINT
     )
 
     set_variable_primal_solution!(instance, variable_primal)
@@ -670,9 +670,9 @@ end
 function LQOI.get_objectivesense(instance::MockLinQuadOptimizer)
     s = instance.inner.sense
     if s == :maximize
-        return MOI.MaxSense
+        return MOI.MAX_SENSE
     else
-        return MOI.MinSense
+        return MOI.MIN_SENSE
     end
 end
 

src/objective.jl

@@ -7,20 +7,20 @@ MOI.supports(::LinQuadOptimizer, ::MOI.ObjectiveSense) = true
 MOI.get(model::LinQuadOptimizer,::MOI.ObjectiveSense) = model.obj_sense
 function MOI.set(model::LinQuadOptimizer, ::MOI.ObjectiveSense,
                   sense::MOI.OptimizationSense)
-    if sense == MOI.MinSense
+    if sense == MOI.MIN_SENSE
         change_objective_sense!(model, :min)
-        model.obj_sense = MOI.MinSense
-    elseif sense == MOI.MaxSense
+        model.obj_sense = MOI.MIN_SENSE
+    elseif sense == MOI.MAX_SENSE
         change_objective_sense!(model, :max)
-        model.obj_sense = MOI.MaxSense
-    elseif sense == MOI.FeasibilitySense
+        model.obj_sense = MOI.MAX_SENSE
+    elseif sense == MOI.FEASIBILITY_SENSE
         # we set the objective sense to :min, and the objective to 0.0
         change_objective_sense!(model, :min)
         unsafe_set!(model, MOI.ObjectiveFunction{Linear}(),
                     MOI.ScalarAffineFunction(MOI.ScalarAffineTerm{Float64}[],
                     0.0))
         model.obj_type = AffineObjective
-        model.obj_sense = MOI.FeasibilitySense
+        model.obj_sense = MOI.FEASIBILITY_SENSE
     else
         throw(MOI.CannotSetAttribute(MOI.ObjectiveSense,
                                      "ObjectiveSense $(sense) not recognised."))
@@ -33,10 +33,10 @@ end
 
 function __assert_objective__(model::LinQuadOptimizer,
                               attribute::MOI.ObjectiveFunction{F}) where F
-    if MOI.get(model, MOI.ObjectiveSense()) == MOI.FeasibilitySense
+    if MOI.get(model, MOI.ObjectiveSense()) == MOI.FEASIBILITY_SENSE
         # it doesn't make sense to set an objective for a feasibility problem
         throw(MOI.CannotSetAttribute(attribute, "Cannot set $(attribute) when" *
-            " MOI.ObjectiveSense is MOI.FeasibilitySense."))
+            " MOI.ObjectiveSense is MOI.FEASIBILITY_SENSE."))
     elseif !(F in supported_objectives(model))
         throw(MOI.UnsupportedAttribute(attribute))
     end

src/solve.jl

@@ -15,8 +15,8 @@ function MOI.optimize!(model::LinQuadOptimizer)
     fill!(model.variable_dual_solution, NaN)
     fill!(model.constraint_primal_solution, NaN)
     fill!(model.constraint_dual_solution, NaN)
-    model.primal_status = MOI.NoSolution
-    model.dual_status   = MOI.NoSolution
+    model.primal_status = MOI.NO_SOLUTION
+    model.dual_status   = MOI.NO_SOLUTION
     model.primal_result_count = 0
     model.dual_result_count = 0
 
@@ -35,31 +35,31 @@ function MOI.optimize!(model::LinQuadOptimizer)
     model.primal_status = get_primal_status(model)
     model.dual_status = get_dual_status(model)
 
-    if model.primal_status in [MOI.FeasiblePoint, MOI.InfeasiblePoint]
+    if model.primal_status in [MOI.FEASIBLE_POINT, MOI.INFEASIBLE_POINT]
         get_variable_primal_solution!(model, model.variable_primal_solution)
         get_linear_primal_solution!(model, model.constraint_primal_solution)
         if has_quadratic(model)
             get_quadratic_primal_solution!(model, model.qconstraint_primal_solution)
         end
         model.primal_result_count = 1
-    elseif model.primal_status == MOI.InfeasibilityCertificate
+    elseif model.primal_status == MOI.INFEASIBILITY_CERTIFICATE
         get_unbounded_ray!(model, model.variable_primal_solution)
         model.primal_result_count = 1
     end
-    if model.dual_status in [MOI.FeasiblePoint, MOI.InfeasiblePoint]
+    if model.dual_status in [MOI.FEASIBLE_POINT, MOI.INFEASIBLE_POINT]
         get_variable_dual_solution!(model, model.variable_dual_solution)
         get_linear_dual_solution!(model, model.constraint_dual_solution)
         if has_quadratic(model)
             get_quadratic_dual_solution!(model, model.qconstraint_dual_solution)
         end
         model.dual_result_count = 1
-    elseif model.dual_status == MOI.InfeasibilityCertificate
+    elseif model.dual_status == MOI.INFEASIBILITY_CERTIFICATE
         get_farkas_dual!(model, model.constraint_dual_solution)
         get_farkas_dual_bounds!(model, model.variable_dual_solution)
         model.dual_result_count = 1
     end
 
-    if MOI.get(model, MOI.ObjectiveSense()) == MOI.MaxSense
+    if MOI.get(model, MOI.ObjectiveSense()) == MOI.MAX_SENSE
         model.constraint_dual_solution *= -1
         model.variable_dual_solution *= -1
     end
@@ -151,7 +151,7 @@ function MOI.get(model::LinQuadOptimizer, ::MOI.ConstraintDual, index::SVCI{<: L
     column = get_column(model, model[index])
     # the variable reduced cost is only the constraint dual if the bound is active,
     # or it might be a dual ray
-    if model.dual_status == MOI.InfeasibilityCertificate
+    if model.dual_status == MOI.INFEASIBILITY_CERTIFICATE
         return model.variable_dual_solution[column]
     else
         set = MOI.get(model, MOI.ConstraintSet(), index)

test/contconic.jl

@@ -23,9 +23,9 @@ end
 function set_lin3test_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Infeasible,
-        primal_status = MOI.NoSolution,
-        dual_status = MOI.InfeasibilityCertificate,
+        termination_status = MOI.INFEASIBLE,
+        primal_status = MOI.NO_SOLUTION,
+        dual_status = MOI.INFEASIBILITY_CERTIFICATE,
         variable_primal = [NaN],
         constraint_primal = [NaN, NaN],
         variable_dual = [NaN],

test/contlinear.jl

@@ -198,9 +198,9 @@ end
 function set_linear8atest_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Infeasible,
-        primal_status = MOI.NoSolution,
-        dual_status = MOI.InfeasibilityCertificate,
+        termination_status = MOI.INFEASIBLE,
+        primal_status = MOI.NO_SOLUTION,
+        dual_status = MOI.INFEASIBILITY_CERTIFICATE,
         variable_dual = [NaN, NaN],
         constraint_dual = [-0.5]
     )
@@ -210,9 +210,9 @@ end
 function set_linear8btest_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.DualInfeasible,
-        primal_status = MOI.InfeasibilityCertificate,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.DUAL_INFEASIBLE,
+        primal_status = MOI.INFEASIBILITY_CERTIFICATE,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [2.0, 1.0],
         constraint_primal = [NaN]
     )
@@ -222,9 +222,9 @@ end
 function set_linear8ctest_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.DualInfeasible,
-        primal_status = MOI.InfeasibilityCertificate,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.DUAL_INFEASIBLE,
+        primal_status = MOI.INFEASIBILITY_CERTIFICATE,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [1.0, 1.0],
         constraint_primal = [NaN]
     )
@@ -295,9 +295,9 @@ end
 function set_linear12test_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Infeasible,
-        primal_status = MOI.NoSolution,
-        dual_status = MOI.InfeasibilityCertificate,
+        termination_status = MOI.INFEASIBLE,
+        primal_status = MOI.NO_SOLUTION,
+        dual_status = MOI.INFEASIBILITY_CERTIFICATE,
         variable_dual = [2/3, 0.0],
         constraint_dual = [-1/3, -1.0]
     )

test/ex.jl

@@ -33,7 +33,7 @@ for solver in solvers
     @test JuMP.hasresultvalues(m)
 
     @test JuMP.terminationstatus(m) == MOI.Success
-    @test JuMP.primalstatus(m) == MOI.FeasiblePoint
+    @test JuMP.primalstatus(m) == MOI.FEASIBLE_POINT
 
     JuMP.resultvalue(x)
     JuMP.resultvalue(y)

test/intlinear.jl

@@ -1,9 +1,9 @@
 function set_knapsacktest_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Optimal,
-        primal_status = MOI.FeasiblePoint,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.OPTIMAL,
+        primal_status = MOI.FEASIBLE_POINT,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [1.0, 0.0, 0.0, 1.0, 1.0],
         constraint_primal = [9.0]
     )
@@ -13,9 +13,9 @@ end
 function set_int1test_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Optimal,
-        primal_status = MOI.FeasiblePoint,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.OPTIMAL,
+        primal_status = MOI.FEASIBLE_POINT,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [4.0, 5.0, 1.0],
         constraint_primal = [10.0, 15.0]
     )
@@ -25,17 +25,17 @@ end
 function set_int3test_solutions!(solver)
     # SOLVE 1
     LQOI.set_solution!(solver,
-        termination_status = MOI.Optimal,
-        primal_status = MOI.FeasiblePoint,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.OPTIMAL,
+        primal_status = MOI.FEASIBLE_POINT,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [1.0, -0.0, -0.0, -0.0, -0.0, 1.0, -0.0, -0.0, -0.0, -0.0, -0.0],
         constraint_primal = [0.9875]
     )
     # SOLVE 2
     LQOI.set_solution!(solver,
-        termination_status = MOI.Optimal,
-        primal_status = MOI.FeasiblePoint,
-        dual_status = MOI.NoSolution,
+        termination_status = MOI.OPTIMAL,
+        primal_status = MOI.FEASIBLE_POINT,
+        dual_status = MOI.NO_SOLUTION,
         variable_primal = [1.0, -0.0, -0.0, -0.0, -0.0, 1.0, -0.0, -0.0, -0.0, -0.0, -0.0],
         constraint_primal = [0.9875]
     )