feat(scalar_1D_t): type-bound division operator

This commit adds
1. operator(/) for an integer denominator and
2. a corresponding unit test.

Author: rouson

example/burgers-1D.F90

@@ -64,7 +64,7 @@ program burgers_1D
     double precision dt
     dt = 1D0
     !associate(dt_du22_dx => dt * .d_dx. ((u**2)/2))
-    associate(dt_du22_dx => u**2)
+    associate(dt_du22_dx => (u**2)/2)
     end associate
   end block
 

src/formal/scalar_1D_s.F90

@@ -65,6 +65,15 @@ pure function cell_center_locations(x_min, x_max, cells) result(x)
 
 #endif
 
+  module procedure divide_by_integer
+    ratio%tensor_1D_t = tensor_1D_t( &
+      values = self%values_/2, x_min = self%x_min_, x_max = self%x_max_, cells = self%cells_, order = self%order_ &
+    )
+    ratio%gradient_operator_1D_ = gradient_operator_1D_t( &
+      k = self%order_, dx = (self%x_max_ - self%x_min_)/self%cells_, cells = self%cells_ &
+    )
+  end procedure
+
   module procedure exponentiate
     power%tensor_1D_t = tensor_1D_t( &
       values = self%values_**2, x_min = self%x_min_, x_max = self%x_max_, cells = self%cells_, order = self%order_ &

src/formal/tensors_1D_m.F90

@@ -81,16 +81,18 @@ pure module function construct_1D_tensor_from_components(values, x_min, x_max, c
     private
     type(gradient_operator_1D_t) gradient_operator_1D_
   contains
+    generic :: grid   => scalar_1D_grid
+    generic :: values => scalar_1D_values
+    generic :: operator(/) => divide_by_integer
     generic :: operator(**) => exponentiate
     generic :: operator(.grad.) => grad
     generic :: operator(.laplacian.) => laplacian
-    generic :: grid   => scalar_1D_grid
-    generic :: values => scalar_1D_values
+    procedure, non_overridable, private :: scalar_1D_values
+    procedure, non_overridable, private :: scalar_1D_grid
+    procedure, non_overridable, private :: divide_by_integer 
     procedure, non_overridable, private :: exponentiate
     procedure, non_overridable, private :: grad
     procedure, non_overridable, private :: laplacian
-    procedure, non_overridable, private :: scalar_1D_values
-    procedure, non_overridable, private :: scalar_1D_grid
   end type
 
   interface scalar_1D_t
@@ -334,6 +336,14 @@ pure module function exponentiate(self, exponent) result(power)
       type(scalar_1D_t) power
     end function
 
+    pure module function divide_by_integer(self, denominator) result(ratio)
+      !! Result is scalar_1D_t "self" divided by the integer denominator
+      implicit none
+      class(scalar_1D_t), intent(in) :: self
+      integer, intent(in) :: denominator
+      type(scalar_1D_t) ratio
+    end function
+
     pure module function laplacian(self) result(laplacian_1D)
       !! Result is mimetic Laplacian of the scalar_1D_t "self"
       implicit none

test/scalar_1D_test_m.F90

@@ -41,6 +41,7 @@ function results() result(test_results)
 
    test_results = scalar_1D_test%run([ & 
       test_description_t('raising a 1D scalar field to a power', usher(check_exponentiation)) &
+     ,test_description_t('dividing a 1D scalar field by a constant', usher(check_divison_operator)) &
    ])
   end function
 
@@ -70,4 +71,24 @@ function check_exponentiation() result(test_diagnosis)
 
   end function
 
+  function check_divison_operator() result(test_diagnosis)
+    type(test_diagnosis_t) test_diagnosis
+    procedure(scalar_1D_initializer_i), pointer :: scalar_1D_initializer => null()
+    integer order
+
+    scalar_1D_initializer => line
+    test_diagnosis = passing_test()
+
+    do order = 2, 4, 2
+      associate(scalar_1D => scalar_1D_t(scalar_1D_initializer, order=order, cells=10, x_min=0D0, x_max=10D0) )
+        associate( scalar_1D_squared => scalar_1D/2 )
+          test_diagnosis = test_diagnosis .also. .all. &
+            (scalar_1D_squared%values() .approximates. scalar_1D%values()/2 .within. tolerance) &
+            // string_t(" for order ") // string_t(order)
+        end associate
+      end associate
+    end do
+
+  end function
+
 end module scalar_1D_test_m