Upgrade to C++17

This commit is contained in:
Chris Thrasher
2025-09-21 13:23:25 -06:00
parent dc3a4ea41a
commit 9c089788ab
92 changed files with 428 additions and 913 deletions

View File

@@ -214,7 +214,7 @@ TEMPLATE_TEST_CASE( "uniform_integer_distribution can handle full width ranges",
constexpr auto lowest = std::numeric_limits<TestType>::min();
constexpr auto highest = std::numeric_limits<TestType>::max();
Catch::uniform_integer_distribution<TestType> dist( lowest, highest );
STATIC_REQUIRE( std::is_same<TestType, decltype( dist( pcg ) )>::value );
STATIC_REQUIRE( std::is_same_v<TestType, decltype( dist( pcg ) )> );
// We need to do bit operations on the results, so we will have to
// cast them to unsigned type.
@@ -451,21 +451,6 @@ namespace {
381264073 };
};
// We need these definitions for C++14 and earlier, but
// GCC will complain about them in newer C++ standards
#if __cplusplus <= 201402L
constexpr bool uniform_integer_test_params<bool>::expected[];
constexpr char uniform_integer_test_params<char>::expected[];
constexpr uint8_t uniform_integer_test_params<uint8_t>::expected[];
constexpr int8_t uniform_integer_test_params<int8_t>::expected[];
constexpr uint16_t uniform_integer_test_params<uint16_t>::expected[];
constexpr int16_t uniform_integer_test_params<int16_t>::expected[];
constexpr uint32_t uniform_integer_test_params<uint32_t>::expected[];
constexpr int32_t uniform_integer_test_params<int32_t>::expected[];
constexpr uint64_t uniform_integer_test_params<uint64_t>::expected[];
constexpr int64_t uniform_integer_test_params<int64_t>::expected[];
#endif
}
TEMPLATE_TEST_CASE( "uniform_integer_distribution is reproducible",
@@ -559,13 +544,6 @@ namespace {
60912.7615841752,
-149060.05936760217 };
};
// We need these definitions for C++14 and earlier, but
// GCC will complain about them in newer C++ standards
#if __cplusplus <= 201402L
constexpr float uniform_fp_test_params<float>::expected[];
constexpr double uniform_fp_test_params<double>::expected[];
#endif
} // namespace
TEMPLATE_TEST_CASE( "uniform_floating_point_distribution is reproducible",