wgtmac commented on code in PR #196: URL: https://github.com/apache/iceberg-cpp/pull/196#discussion_r2306077067
########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value Review Comment: ```suggestion ``` The comment is redundant. ########## src/iceberg/util/endian.h: ########## @@ -0,0 +1,94 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#pragma once + +#include <bit> +#include <concepts> +#include <cstdint> + +/// \file iceberg/util/endian.h +/// \brief Endianness conversion utilities + +namespace iceberg { + +/// \brief Concept for values that can be converted to/from another endian format. +template <typename T> +concept EndianConvertible = std::is_arithmetic_v<T> && !std::same_as<T, bool>; + +/// \brief Byte-swap a value. For floating-point types, only support 32-bit and 64-bit +/// floats. +template <EndianConvertible T> +constexpr T ByteSwap(T value) { + if constexpr (sizeof(T) <= 1) { + return value; + } else if constexpr (std::is_integral_v<T>) { + return std::byteswap(value); + } else if constexpr (std::is_floating_point_v<T>) { + if constexpr (sizeof(T) == sizeof(uint32_t)) { Review Comment: Should we add `if constexpr (sizeof(T) == sizeof(uint16_t))` to be future-proof? There are std::float16_t and std::bfloat16_t. ########## src/iceberg/util/endian.h: ########## @@ -0,0 +1,94 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#pragma once + +#include <bit> +#include <concepts> +#include <cstdint> + +/// \file iceberg/util/endian.h +/// \brief Endianness conversion utilities + +namespace iceberg { + +/// \brief Concept for values that can be converted to/from another endian format. +template <typename T> +concept EndianConvertible = std::is_arithmetic_v<T> && !std::same_as<T, bool>; + +/// \brief Byte-swap a value. For floating-point types, only support 32-bit and 64-bit +/// floats. +template <EndianConvertible T> +constexpr T ByteSwap(T value) { + if constexpr (sizeof(T) <= 1) { Review Comment: It seems that you have already supported bool type here. ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0xDEADBEEF)), 0xDEADBEEF); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<uint64_t>::max())), + std::numeric_limits<uint64_t>::max()); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0)), 0); + + EXPECT_EQ(FromBigEndian(ToBigEndian<int16_t>(-1)), -1); + EXPECT_EQ(FromLittleEndian(ToLittleEndian<int32_t>(-0x12345678)), -0x12345678); + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<int64_t>::min())), + std::numeric_limits<int64_t>::min()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<int16_t>::max())), + std::numeric_limits<int16_t>::max()); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(3.14f)), 3.14f); + EXPECT_EQ(FromBigEndian(ToBigEndian(2.718281828459045)), 2.718281828459045); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::infinity())), + std::numeric_limits<float>::infinity()); + EXPECT_EQ(FromBigEndian(ToBigEndian(-std::numeric_limits<float>::infinity())), + -std::numeric_limits<float>::infinity()); + EXPECT_TRUE(std::isnan( + FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::quiet_NaN())))); + EXPECT_EQ(FromBigEndian(ToBigEndian(0.0f)), 0.0f); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-0.0f)), -0.0f); + + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<double>::infinity())), + std::numeric_limits<double>::infinity()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-std::numeric_limits<double>::infinity())), + -std::numeric_limits<double>::infinity()); + EXPECT_TRUE( + std::isnan(FromBigEndian(ToBigEndian(std::numeric_limits<double>::quiet_NaN())))); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(0.0)), 0.0); + EXPECT_EQ(FromBigEndian(ToBigEndian(-0.0)), -0.0); +} + +// test constexpr evaluation +TEST(EndianTest, ConstexprEvaluation) { + static_assert(FromBigEndian(ToBigEndian<uint16_t>(0x1234)) == 0x1234); + static_assert(FromLittleEndian(ToLittleEndian<uint32_t>(0x12345678)) == 0x12345678); + static_assert(FromBigEndian(ToBigEndian<int64_t>(-1)) == -1); + + static_assert(ToBigEndian<uint8_t>(0xFF) == 0xFF); + static_assert(FromLittleEndian<int8_t>(-1) == -1); + + static_assert(FromLittleEndian(ToLittleEndian(3.14f)) == 3.14f); + static_assert(FromBigEndian(ToBigEndian(2.71)) == 2.71); +} + +// test platform dependent behavior +TEST(EndianTest, PlatformDependentBehavior) { Review Comment: I don't think we need this case. ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); Review Comment: There are a lot of duplicate lines here. Can we use a simple macro for this? ``` #define EXPECT_ROUNDTRIP(T value) \ EXPECT_EQ(FromLittleEndian(ToLittleEndian(value)), value); \ EXPECT_EQ(FromBigEndian(ToBigEndian(value)), value); ``` ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0xDEADBEEF)), 0xDEADBEEF); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<uint64_t>::max())), + std::numeric_limits<uint64_t>::max()); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0)), 0); + + EXPECT_EQ(FromBigEndian(ToBigEndian<int16_t>(-1)), -1); + EXPECT_EQ(FromLittleEndian(ToLittleEndian<int32_t>(-0x12345678)), -0x12345678); + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<int64_t>::min())), + std::numeric_limits<int64_t>::min()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<int16_t>::max())), + std::numeric_limits<int16_t>::max()); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(3.14f)), 3.14f); + EXPECT_EQ(FromBigEndian(ToBigEndian(2.718281828459045)), 2.718281828459045); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::infinity())), + std::numeric_limits<float>::infinity()); + EXPECT_EQ(FromBigEndian(ToBigEndian(-std::numeric_limits<float>::infinity())), + -std::numeric_limits<float>::infinity()); + EXPECT_TRUE(std::isnan( + FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::quiet_NaN())))); + EXPECT_EQ(FromBigEndian(ToBigEndian(0.0f)), 0.0f); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-0.0f)), -0.0f); + + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<double>::infinity())), + std::numeric_limits<double>::infinity()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-std::numeric_limits<double>::infinity())), + -std::numeric_limits<double>::infinity()); + EXPECT_TRUE( + std::isnan(FromBigEndian(ToBigEndian(std::numeric_limits<double>::quiet_NaN())))); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(0.0)), 0.0); + EXPECT_EQ(FromBigEndian(ToBigEndian(-0.0)), -0.0); +} + +// test constexpr evaluation +TEST(EndianTest, ConstexprEvaluation) { + static_assert(FromBigEndian(ToBigEndian<uint16_t>(0x1234)) == 0x1234); + static_assert(FromLittleEndian(ToLittleEndian<uint32_t>(0x12345678)) == 0x12345678); + static_assert(FromBigEndian(ToBigEndian<int64_t>(-1)) == -1); + + static_assert(ToBigEndian<uint8_t>(0xFF) == 0xFF); + static_assert(FromLittleEndian<int8_t>(-1) == -1); + + static_assert(FromLittleEndian(ToLittleEndian(3.14f)) == 3.14f); + static_assert(FromBigEndian(ToBigEndian(2.71)) == 2.71); +} + +// test platform dependent behavior +TEST(EndianTest, PlatformDependentBehavior) { + uint32_t test_value = 0x12345678; + + if constexpr (std::endian::native == std::endian::little) { + EXPECT_EQ(ToLittleEndian(test_value), test_value); + EXPECT_EQ(FromLittleEndian(test_value), test_value); + EXPECT_NE(ToBigEndian(test_value), test_value); + } else if constexpr (std::endian::native == std::endian::big) { + EXPECT_EQ(ToBigEndian(test_value), test_value); + EXPECT_EQ(FromBigEndian(test_value), test_value); + EXPECT_NE(ToLittleEndian(test_value), test_value); + } + + EXPECT_EQ(ToLittleEndian<uint8_t>(0xAB), 0xAB); + EXPECT_EQ(ToBigEndian<uint8_t>(0xAB), 0xAB); +} + +// test specific byte pattern validation +TEST(EndianTest, SpecificBytePatternValidation) { + uint32_t original_int = 0x12345678; + uint32_t little_endian_int = ToLittleEndian(original_int); + uint32_t big_endian_int = ToBigEndian(original_int); + + auto little_int_bytes = std::bit_cast<std::array<uint8_t, 4>>(little_endian_int); + auto big_int_bytes = std::bit_cast<std::array<uint8_t, 4>>(big_endian_int); + + EXPECT_EQ(little_int_bytes, (std::array<uint8_t, 4>{0x78, 0x56, 0x34, 0x12})); + EXPECT_EQ(big_int_bytes, (std::array<uint8_t, 4>{0x12, 0x34, 0x56, 0x78})); + + float original_float = 3.14f; + float little_endian_float = ToLittleEndian(original_float); + float big_endian_float = ToBigEndian(original_float); + + auto little_float_bytes = std::bit_cast<std::array<uint8_t, 4>>(little_endian_float); + auto big_float_bytes = std::bit_cast<std::array<uint8_t, 4>>(big_endian_float); + + EXPECT_EQ(little_float_bytes, (std::array<uint8_t, 4>{0xC3, 0xF5, 0x48, 0x40})); + EXPECT_EQ(big_float_bytes, (std::array<uint8_t, 4>{0x40, 0x48, 0xF5, 0xC3})); Review Comment: ```suggestion EXPECT_EQ(little_float_bytes, std::array<uint8_t, 4>{0xC3, 0xF5, 0x48, 0x40}); EXPECT_EQ(big_float_bytes, std::array<uint8_t, 4>{0x40, 0x48, 0xF5, 0xC3}); ``` ########## src/iceberg/util/endian.h: ########## @@ -0,0 +1,94 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#pragma once + +#include <bit> +#include <concepts> +#include <cstdint> + +/// \file iceberg/util/endian.h +/// \brief Endianness conversion utilities + +namespace iceberg { + +/// \brief Concept for values that can be converted to/from another endian format. +template <typename T> +concept EndianConvertible = std::is_arithmetic_v<T> && !std::same_as<T, bool>; + +/// \brief Byte-swap a value. For floating-point types, only support 32-bit and 64-bit +/// floats. +template <EndianConvertible T> +constexpr T ByteSwap(T value) { + if constexpr (sizeof(T) <= 1) { + return value; + } else if constexpr (std::is_integral_v<T>) { + return std::byteswap(value); + } else if constexpr (std::is_floating_point_v<T>) { + if constexpr (sizeof(T) == sizeof(uint32_t)) { + return std::bit_cast<T>(std::byteswap(std::bit_cast<uint32_t>(value))); + } else if constexpr (sizeof(T) == sizeof(uint64_t)) { + return std::bit_cast<T>(std::byteswap(std::bit_cast<uint64_t>(value))); + } else { + static_assert(false, "Unsupported floating-point size for endian conversion."); + } + } +} + +/// \brief Convert a value to little-endian format. +template <EndianConvertible T> +constexpr T ToLittleEndian(T value) { + if constexpr (std::endian::native == std::endian::little || sizeof(T) <= 1) { + return value; + } else { + return ByteSwap(value); + } Review Comment: ```suggestion if constexpr (std::endian::native == std::endian::little) { return value; } else { return ByteSwap(value); } ``` Doesn't it look better? ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0xDEADBEEF)), 0xDEADBEEF); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<uint64_t>::max())), + std::numeric_limits<uint64_t>::max()); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0)), 0); + + EXPECT_EQ(FromBigEndian(ToBigEndian<int16_t>(-1)), -1); + EXPECT_EQ(FromLittleEndian(ToLittleEndian<int32_t>(-0x12345678)), -0x12345678); + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<int64_t>::min())), + std::numeric_limits<int64_t>::min()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<int16_t>::max())), + std::numeric_limits<int16_t>::max()); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(3.14f)), 3.14f); + EXPECT_EQ(FromBigEndian(ToBigEndian(2.718281828459045)), 2.718281828459045); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::infinity())), + std::numeric_limits<float>::infinity()); + EXPECT_EQ(FromBigEndian(ToBigEndian(-std::numeric_limits<float>::infinity())), + -std::numeric_limits<float>::infinity()); + EXPECT_TRUE(std::isnan( + FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::quiet_NaN())))); + EXPECT_EQ(FromBigEndian(ToBigEndian(0.0f)), 0.0f); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-0.0f)), -0.0f); + + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<double>::infinity())), + std::numeric_limits<double>::infinity()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-std::numeric_limits<double>::infinity())), + -std::numeric_limits<double>::infinity()); + EXPECT_TRUE( + std::isnan(FromBigEndian(ToBigEndian(std::numeric_limits<double>::quiet_NaN())))); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(0.0)), 0.0); + EXPECT_EQ(FromBigEndian(ToBigEndian(-0.0)), -0.0); +} + +// test constexpr evaluation +TEST(EndianTest, ConstexprEvaluation) { + static_assert(FromBigEndian(ToBigEndian<uint16_t>(0x1234)) == 0x1234); + static_assert(FromLittleEndian(ToLittleEndian<uint32_t>(0x12345678)) == 0x12345678); + static_assert(FromBigEndian(ToBigEndian<int64_t>(-1)) == -1); + + static_assert(ToBigEndian<uint8_t>(0xFF) == 0xFF); + static_assert(FromLittleEndian<int8_t>(-1) == -1); + + static_assert(FromLittleEndian(ToLittleEndian(3.14f)) == 3.14f); + static_assert(FromBigEndian(ToBigEndian(2.71)) == 2.71); +} + +// test platform dependent behavior +TEST(EndianTest, PlatformDependentBehavior) { + uint32_t test_value = 0x12345678; + + if constexpr (std::endian::native == std::endian::little) { + EXPECT_EQ(ToLittleEndian(test_value), test_value); + EXPECT_EQ(FromLittleEndian(test_value), test_value); + EXPECT_NE(ToBigEndian(test_value), test_value); + } else if constexpr (std::endian::native == std::endian::big) { + EXPECT_EQ(ToBigEndian(test_value), test_value); + EXPECT_EQ(FromBigEndian(test_value), test_value); + EXPECT_NE(ToLittleEndian(test_value), test_value); + } + + EXPECT_EQ(ToLittleEndian<uint8_t>(0xAB), 0xAB); + EXPECT_EQ(ToBigEndian<uint8_t>(0xAB), 0xAB); +} + +// test specific byte pattern validation +TEST(EndianTest, SpecificBytePatternValidation) { Review Comment: ```suggestion TEST(EndianTest, ByteWiseValidation) { ``` ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0xDEADBEEF)), 0xDEADBEEF); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<uint64_t>::max())), + std::numeric_limits<uint64_t>::max()); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0)), 0); + + EXPECT_EQ(FromBigEndian(ToBigEndian<int16_t>(-1)), -1); + EXPECT_EQ(FromLittleEndian(ToLittleEndian<int32_t>(-0x12345678)), -0x12345678); + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<int64_t>::min())), + std::numeric_limits<int64_t>::min()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<int16_t>::max())), + std::numeric_limits<int16_t>::max()); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(3.14f)), 3.14f); + EXPECT_EQ(FromBigEndian(ToBigEndian(2.718281828459045)), 2.718281828459045); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::infinity())), + std::numeric_limits<float>::infinity()); + EXPECT_EQ(FromBigEndian(ToBigEndian(-std::numeric_limits<float>::infinity())), + -std::numeric_limits<float>::infinity()); + EXPECT_TRUE(std::isnan( + FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::quiet_NaN())))); + EXPECT_EQ(FromBigEndian(ToBigEndian(0.0f)), 0.0f); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-0.0f)), -0.0f); + + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<double>::infinity())), + std::numeric_limits<double>::infinity()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-std::numeric_limits<double>::infinity())), + -std::numeric_limits<double>::infinity()); + EXPECT_TRUE( + std::isnan(FromBigEndian(ToBigEndian(std::numeric_limits<double>::quiet_NaN())))); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(0.0)), 0.0); + EXPECT_EQ(FromBigEndian(ToBigEndian(-0.0)), -0.0); +} + +// test constexpr evaluation +TEST(EndianTest, ConstexprEvaluation) { Review Comment: Is there any difference with the above case? ########## src/iceberg/util/endian.h: ########## @@ -0,0 +1,124 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#pragma once + +#include <bit> +#include <concepts> + +/// \file iceberg/util/endian.h +/// \brief Endianness conversion utilities + +namespace iceberg { + +/// \brief Concept for values that can be converted to/from another endian format. +template <typename T> +concept EndianConvertible = std::is_arithmetic_v<T> && !std::same_as<T, bool>; Review Comment: If we don't support bool here, users may need to special case bool to avoid passing it to `ByteSwap`. For simplicity, should we support bool as well? ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); Review Comment: Or we can use template function to leverage parameterized test to improve readability. ########## test/endian_test.cc: ########## @@ -0,0 +1,121 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +#include "iceberg/util/endian.h" + +#include <array> +#include <cmath> +#include <limits> + +#include <gtest/gtest.h> + +namespace iceberg { + +// test round trip preserves value +TEST(EndianTest, RoundTripPreservesValue) { + EXPECT_EQ(FromLittleEndian(ToLittleEndian<uint16_t>(0x1234)), 0x1234); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0xDEADBEEF)), 0xDEADBEEF); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<uint64_t>::max())), + std::numeric_limits<uint64_t>::max()); + EXPECT_EQ(FromBigEndian(ToBigEndian<uint32_t>(0)), 0); + + EXPECT_EQ(FromBigEndian(ToBigEndian<int16_t>(-1)), -1); + EXPECT_EQ(FromLittleEndian(ToLittleEndian<int32_t>(-0x12345678)), -0x12345678); + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<int64_t>::min())), + std::numeric_limits<int64_t>::min()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<int16_t>::max())), + std::numeric_limits<int16_t>::max()); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(3.14f)), 3.14f); + EXPECT_EQ(FromBigEndian(ToBigEndian(2.718281828459045)), 2.718281828459045); + + EXPECT_EQ(FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::infinity())), + std::numeric_limits<float>::infinity()); + EXPECT_EQ(FromBigEndian(ToBigEndian(-std::numeric_limits<float>::infinity())), + -std::numeric_limits<float>::infinity()); + EXPECT_TRUE(std::isnan( + FromLittleEndian(ToLittleEndian(std::numeric_limits<float>::quiet_NaN())))); + EXPECT_EQ(FromBigEndian(ToBigEndian(0.0f)), 0.0f); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-0.0f)), -0.0f); + + EXPECT_EQ(FromBigEndian(ToBigEndian(std::numeric_limits<double>::infinity())), + std::numeric_limits<double>::infinity()); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(-std::numeric_limits<double>::infinity())), + -std::numeric_limits<double>::infinity()); + EXPECT_TRUE( + std::isnan(FromBigEndian(ToBigEndian(std::numeric_limits<double>::quiet_NaN())))); + EXPECT_EQ(FromLittleEndian(ToLittleEndian(0.0)), 0.0); + EXPECT_EQ(FromBigEndian(ToBigEndian(-0.0)), -0.0); +} + +// test constexpr evaluation +TEST(EndianTest, ConstexprEvaluation) { + static_assert(FromBigEndian(ToBigEndian<uint16_t>(0x1234)) == 0x1234); + static_assert(FromLittleEndian(ToLittleEndian<uint32_t>(0x12345678)) == 0x12345678); + static_assert(FromBigEndian(ToBigEndian<int64_t>(-1)) == -1); + + static_assert(ToBigEndian<uint8_t>(0xFF) == 0xFF); + static_assert(FromLittleEndian<int8_t>(-1) == -1); + + static_assert(FromLittleEndian(ToLittleEndian(3.14f)) == 3.14f); + static_assert(FromBigEndian(ToBigEndian(2.71)) == 2.71); +} + +// test platform dependent behavior +TEST(EndianTest, PlatformDependentBehavior) { + uint32_t test_value = 0x12345678; + + if constexpr (std::endian::native == std::endian::little) { + EXPECT_EQ(ToLittleEndian(test_value), test_value); + EXPECT_EQ(FromLittleEndian(test_value), test_value); + EXPECT_NE(ToBigEndian(test_value), test_value); + } else if constexpr (std::endian::native == std::endian::big) { + EXPECT_EQ(ToBigEndian(test_value), test_value); + EXPECT_EQ(FromBigEndian(test_value), test_value); + EXPECT_NE(ToLittleEndian(test_value), test_value); + } + + EXPECT_EQ(ToLittleEndian<uint8_t>(0xAB), 0xAB); + EXPECT_EQ(ToBigEndian<uint8_t>(0xAB), 0xAB); +} + +// test specific byte pattern validation +TEST(EndianTest, SpecificBytePatternValidation) { + uint32_t original_int = 0x12345678; + uint32_t little_endian_int = ToLittleEndian(original_int); + uint32_t big_endian_int = ToBigEndian(original_int); + + auto little_int_bytes = std::bit_cast<std::array<uint8_t, 4>>(little_endian_int); + auto big_int_bytes = std::bit_cast<std::array<uint8_t, 4>>(big_endian_int); + + EXPECT_EQ(little_int_bytes, (std::array<uint8_t, 4>{0x78, 0x56, 0x34, 0x12})); + EXPECT_EQ(big_int_bytes, (std::array<uint8_t, 4>{0x12, 0x34, 0x56, 0x78})); Review Comment: ```suggestion EXPECT_EQ(little_int_bytes, std::array<uint8_t, 4>{0x78, 0x56, 0x34, 0x12}); EXPECT_EQ(big_int_bytes, std::array<uint8_t, 4>{0x12, 0x34, 0x56, 0x78}); ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
