Get tests back up running.

src/plans/vectorial_plan.jl

@@ -788,7 +788,7 @@ function get_jacobian!(
     FT,VGF<:AbstractVectorGradientFunction{<:InplaceEvaluation,FT,<:CoordinateVectorialType}
 }
     vgf.jacobian!!(M, JF, p)
-    _change_basis!(JF, vgf.jacobian_type.basis, basis)
+    _change_basis!(M, p, JF, vgf.jacobian_type.basis, basis)
     return JF
 end
 
@@ -1119,7 +1119,7 @@ This function can perform the evalutation inplace of `V`.
 get_value(M::AbstractManifold, vgf::AbstractVectorFunction, p, i)
 function get_value(
     M::AbstractManifold, vgf::AbstractVectorFunction{E,<:FunctionVectorialType}, p, i=:
-) where {E<:AbstractEvaluationType}
+) where {E<:AllocatingEvaluation}
     c = vgf.value!!(M, p)
     if isa(c, Number)
         return c
@@ -1140,9 +1140,25 @@ function get_value(
 ) where {E<:AbstractEvaluationType}
     return [f(M, p) for f in vgf.value!![i]]
 end
+function get_value(
+    M::AbstractManifold,
+    vgf::AbstractVectorFunction{E,<:FunctionVectorialType},
+    p,
+    i=:;
+    value_cache=zeros(vgf.range_dimension),
+) where {E<:InplaceEvaluation}
+    vgf.value!!(M, value_cache, p)
+    return value_cache[i]
+end
+# A ComponentVectorialType Inplace does that make sense, since those would be real - valued functions
+
 function get_value!(
-    M::AbstractManifold, V, vgf::AbstractVectorFunction{E,<:FunctionVectorialType}, p, i=:
-) where {E<:AbstractEvaluationType}
+    M::AbstractManifold,
+    V,
+    vgf::AbstractVectorFunction{AllocatingEvaluation,<:FunctionVectorialType},
+    p,
+    i=:,
+)
     c = vgf.value!!(M, p)
     if isa(c, Number)
         V .= c
@@ -1151,6 +1167,20 @@ function get_value!(
     end
     return V
 end
+
+function get_value!(
+    M::AbstractManifold,
+    V,
+    vgf::AbstractVectorFunction{InplaceEvaluation,<:FunctionVectorialType},
+    p,
+    i=:;
+    value_cache=zeros(vgf.range_dimension),
+)
+    vgf.value!!(M, value_cache, p)
+    V .= value_cache[i]
+    return V
+end
+
 @doc raw"""
     get_value_function(vgf::VectorGradientFunction, recursive=false)
 

test/plans/test_constrained_plan.jl

@@ -12,6 +12,7 @@ include("../utils/dummy_types.jl")
     hess_f!(M, Y, p, X) = (Y .= [2.0, 2.0, 2.0])
     # Inequality constraints
     g(M, p) = [p[1] - 1, -p[2] - 1]
+    g!(M, V, p) = (V .= [p[1] - 1, -p[2] - 1])
     # # Function
     grad_g(M, p) = [[1.0, 0.0, 0.0], [0.0, -1.0, 0.0]]
     grad_gA(M, p) = [1.0 0.0; 0.0 -1.0; 0.0 0.0]
@@ -41,6 +42,7 @@ include("../utils/dummy_types.jl")
     ) == 2
     # Equality Constraints
     h(M, p) = [2 * p[3] - 1]
+    h!(M, V, p) = (V .= [2 * p[3] - 1])
     h1(M, p) = 2 * p[3] - 1
     grad_h(M, p) = [[0.0, 0.0, 2.0]]
     grad_hA(M, p) = [[0.0, 0.0, 2.0];;]
@@ -83,9 +85,9 @@ include("../utils/dummy_types.jl")
     cofm = ConstrainedManifoldObjective(
         f,
         grad_f!,
-        g,
+        g!,
         grad_g!,
-        h,
+        h!,
         grad_h!;
         evaluation=InplaceEvaluation(),
         inequality_constraints=2,
@@ -149,9 +151,9 @@ include("../utils/dummy_types.jl")
     cofhm = ConstrainedManifoldObjective(
         f,
         grad_f!,
-        g,
+        g!,
         grad_g!,
-        h,
+        h!,
         grad_h!;
         hess_f=hess_f!,
         hess_g=hess_g!,

test/plans/test_vectorial_plan.jl

@@ -3,6 +3,7 @@ using Manopt: get_value, get_value_function, get_gradient_function
 @testset "VectorialGradientCost" begin
     M = ManifoldsBase.DefaultManifold(3)
     g(M, p) = [p[1] - 1, -p[2] - 1]
+    g!(M, V, p) = (V .= [p[1] - 1, -p[2] - 1])
     # # Function
     grad_g(M, p) = [[1.0, 0.0, 0.0], [0.0, -1.0, 0.0]]
     hess_g(M, p, X) = [copy(X), -copy(X)]
@@ -37,7 +38,7 @@ using Manopt: get_value, get_value_function, get_gradient_function
         function_type=ComponentVectorialType(),
         jacobian_type=ComponentVectorialType(),
     )
-    vgf_fi = VectorGradientFunction(g, grad_g!, 2; evaluation=InplaceEvaluation())
+    vgf_fi = VectorGradientFunction(g!, grad_g!, 2; evaluation=InplaceEvaluation())
     vgf_vi = VectorGradientFunction(
         [g1, g2],
         [grad_g1!, grad_g2!],
@@ -50,7 +51,7 @@ using Manopt: get_value, get_value_function, get_gradient_function
         g, jac_g, 2; jacobian_type=CoordinateVectorialType(DefaultOrthonormalBasis())
     )
     vgf_ji = VectorGradientFunction(
-        g,
+        g!,
         jac_g!,
         2;
         jacobian_type=CoordinateVectorialType(DefaultOrthonormalBasis()),
@@ -71,7 +72,7 @@ using Manopt: get_value, get_value_function, get_gradient_function
         jacobian_type=ComponentVectorialType(),
         hessian_type=ComponentVectorialType(),
     )
-    vhf_fi = VectorHessianFunction(g, grad_g!, hess_g!, 2; evaluation=InplaceEvaluation())
+    vhf_fi = VectorHessianFunction(g!, grad_g!, hess_g!, 2; evaluation=InplaceEvaluation())
     vhf_vi = VectorHessianFunction(
         [g1, g2],
         [grad_g1!, grad_g2!],