added more testing of value_cast conversion accuracy when using integral types

test/runtime/CMakeLists.txt

@@ -26,6 +26,7 @@ find_package(Catch2 3 REQUIRED)
 
 add_executable(
     unit_tests_runtime distribution_test.cpp fixed_string_test.cpp fmt_test.cpp math_test.cpp atomic_test.cpp
+                       truncation_test.cpp
 )
 if(${projectPrefix}BUILD_CXX_MODULES)
     target_compile_definitions(unit_tests_runtime PUBLIC ${projectPrefix}MODULES)

test/runtime/almost_equals.h

@@ -37,18 +37,36 @@ struct AlmostEqualsMatcher : Catch::Matchers::MatcherGenericBase {
   explicit AlmostEqualsMatcher(const T& target) : target_{target} {}
 
   template<std::convertible_to<T> U>
-    requires std::same_as<typename T::rep, typename U::rep> && treat_as_floating_point<typename T::rep>
+    requires std::same_as<typename T::rep, typename U::rep>
   bool match(const U& other) const
   {
     using std::abs;
-    using common = std::common_type_t<T, U>;
+    using rep = typename T::rep;
+    using common = conditional<treat_as_floating_point<rep>, std::common_type_t<T, U>, T>;
     const auto x = common(target_).numerical_value_in(common::unit);
     const auto y = common(other).numerical_value_in(common::unit);
-    const auto maxXYOne = std::max({typename T::rep{1}, abs(x), abs(y)});
-    return abs(x - y) <= std::numeric_limits<typename T::rep>::epsilon() * maxXYOne;
+    if constexpr (treat_as_floating_point<rep>) {
+      const auto maxXYOne = std::max({rep{1}, abs(x), abs(y)});
+      return abs(x - y) <= std::numeric_limits<rep>::epsilon() * maxXYOne;
+    } else {
+      if (x >= 0) {
+        return x - 1 <= y && y - 1 <= x;
+      } else {
+        return x <= y + 1 && y <= x + 1;
+      }
+    }
   }
 
-  std::string describe() const override { return "almost equals: " + MP_UNITS_STD_FMT::format("{}", target_); }
+  std::string describe() const override
+  {
+    if constexpr (treat_as_floating_point<typename T::rep>) {
+      return "almost equals: " + MP_UNITS_STD_FMT::format("{}", target_);
+    } else {
+      return "almost equals: " + MP_UNITS_STD_FMT::format("[ {0} ({0:#x}) +/- 1 ] {1}",
+                                                          target_.numerical_value_is_an_implementation_detail_,
+                                                          target_.unit);
+    }
+  }
 
 private:
   const T& target_;
@@ -60,4 +78,5 @@ AlmostEqualsMatcher<T> AlmostEquals(const T& target)
   return AlmostEqualsMatcher<T>{target};
 }
 
+
 }  // namespace mp_units

test/runtime/truncation_test.cpp

@@ -0,0 +1,126 @@
+// The MIT License (MIT)
+//
+// Copyright (c) 2018 Mateusz Pusz
+//
+// Permission is hereby granted, free of charge, to any person obtaining a copy
+// of this software and associated documentation files (the "Software"), to deal
+// in the Software without restriction, including without limitation the rights
+// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+// copies of the Software, and to permit persons to whom the Software is
+// furnished to do so, subject to the following conditions:
+//
+// The above copyright notice and this permission notice shall be included in all
+// copies or substantial portions of the Software.
+//
+// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+// SOFTWARE.
+
+#include "almost_equals.h"
+#include <catch2/catch_template_test_macros.hpp>
+#include <catch2/catch_test_macros.hpp>
+#include <catch2/matchers/catch_matchers.hpp>
+#include <limits>
+#include <numbers>
+#ifdef MP_UNITS_MODULES
+import mp_units;
+#else
+#include <mp-units/ostream.h>  // IWYU pragma: keep
+#include <mp-units/systems/angular.h>
+#endif
+
+using namespace mp_units;
+using namespace mp_units::angular;
+using namespace mp_units::angular::unit_symbols;
+
+inline constexpr struct half_revolution : named_unit<"hrev", mag_pi * radian> {
+} half_revolution;
+inline constexpr auto hrev = half_revolution;
+
+// constexpr auto revb6 = mag_ratio<1,3> * mag_pi * rad;
+
+TEST_CASE("value_cast should not truncate for valid inputs", "[value_cast]")
+{
+  SECTION("num > den > 1, irr = 1")
+  {
+    REQUIRE_THAT(value_cast<grad>(9 * deg), AlmostEquals(10 * grad));
+    REQUIRE_THAT(value_cast<grad>(360 * deg), AlmostEquals(400 * grad));
+  }
+
+  SECTION("1 < num < den, irr = 1")
+  {
+    REQUIRE_THAT(value_cast<deg>(10 * grad), AlmostEquals(9 * deg));
+    REQUIRE_THAT(value_cast<deg>(400 * grad), AlmostEquals(360 * deg));
+  }
+
+
+  SECTION("num > den = 1, irr > 1")
+  {
+    REQUIRE_THAT(value_cast<rad>(1 * rev), AlmostEquals(6 * rad));
+    REQUIRE_THAT(value_cast<rad>(5 * rev), AlmostEquals(31 * rad));
+    REQUIRE_THAT(value_cast<rad>(10 * rev), AlmostEquals(63 * rad));
+    REQUIRE_THAT(value_cast<rad>(20 * rev), AlmostEquals(126 * rad));
+  }
+
+  SECTION("1 = num < den, irr < 1")
+  {
+    REQUIRE_THAT(value_cast<rev>(6 * rad), AlmostEquals(1 * rev));
+    REQUIRE_THAT(value_cast<rev>(31 * rad), AlmostEquals(5 * rev));
+    REQUIRE_THAT(value_cast<rev>(63 * rad), AlmostEquals(10 * rev));
+    REQUIRE_THAT(value_cast<rev>(126 * rad), AlmostEquals(20 * rev));
+  }
+
+  SECTION("rational = 1, irrational > 1")
+  {
+    REQUIRE_THAT(value_cast<rad>(1 * hrev), AlmostEquals(3 * rad));
+    REQUIRE_THAT(value_cast<rad>(10 * hrev), AlmostEquals(31 * rad));
+    REQUIRE_THAT(value_cast<rad>(20 * hrev), AlmostEquals(63 * rad));
+    REQUIRE_THAT(value_cast<rad>(40 * hrev), AlmostEquals(126 * rad));
+  }
+
+  SECTION("rational = 1, irrational < 1")
+  {
+    REQUIRE_THAT(value_cast<hrev>(3 * rad), AlmostEquals(1 * hrev));
+    REQUIRE_THAT(value_cast<hrev>(31 * rad), AlmostEquals(10 * hrev));
+    REQUIRE_THAT(value_cast<hrev>(63 * rad), AlmostEquals(20 * hrev));
+    REQUIRE_THAT(value_cast<hrev>(126 * rad), AlmostEquals(40 * hrev));
+  }
+}
+
+
+TEMPLATE_TEST_CASE("value_cast should not overflow internally for valid inputs", "[value_cast]", std::int8_t,
+                   std::uint8_t, std::int16_t, std::uint16_t, std::int32_t, std::uint32_t)
+{
+  // max()/20: small enough so that none of the tested factors are likely to cause overflow, but still be nonzero;
+  // the "easy" test to verify the test itself is good.
+  std::vector<TestType> test_values = {std::numeric_limits<TestType>::max() / 20,
+                                       std::numeric_limits<TestType>::max() - 1};
+  if (std::is_signed_v<TestType>) {
+    test_values.push_back(std::numeric_limits<TestType>::min() + 1);
+  }
+
+  for (TestType tv : test_values) {
+    SECTION("grad <-> deg")
+    {
+      auto deg_number = static_cast<TestType>(std::trunc(0.9 * tv));
+      auto grad_number = static_cast<TestType>(std::round(deg_number / 0.9));
+      INFO(MP_UNITS_STD_FMT::format("{} deg ~ {} grad", deg_number, grad_number));
+      REQUIRE_THAT(value_cast<grad>(deg_number * deg), AlmostEquals(grad_number * grad));
+      REQUIRE_THAT(value_cast<deg>(grad_number * grad), AlmostEquals(deg_number * deg));
+    }
+
+
+    SECTION("rad <-> rev")
+    {
+      auto rev_number = static_cast<TestType>(std::trunc(0.5 * std::numbers::inv_pi * tv));
+      auto rad_number = static_cast<TestType>(std::round(2 * std::numbers::pi * rev_number));
+      INFO(MP_UNITS_STD_FMT::format("{} rev ~ {} rad", rev_number, rad_number));
+      REQUIRE_THAT(value_cast<rad>(rev_number * rev), AlmostEquals(rad_number * rad));
+      REQUIRE_THAT(value_cast<rev>(rad_number * rad), AlmostEquals(rev_number * rev));
+    }
+  }
+}