koffi 0.9.3 → 0.9.6

package/vendor/dragonbox/subproject/common/include/random_float.h

@@ -1,182 +1,182 @@
-// Copyright 2020-2021 Junekey Jeon
-//
-// The contents of this file may be used under the terms of
-// the Apache License v2.0 with LLVM Exceptions.
-//
-//    (See accompanying file LICENSE-Apache or copy at
-//     https://llvm.org/foundation/relicensing/LICENSE.txt)
-//
-// Alternatively, the contents of this file may be used under the terms of
-// the Boost Software License, Version 1.0.
-//    (See accompanying file LICENSE-Boost or copy at
-//     https://www.boost.org/LICENSE_1_0.txt)
-//
-// Unless required by applicable law or agreed to in writing, this software
-// is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-// KIND, either express or implied.
-
-#ifndef JKJ_HEADER_RANDOM_FLOAT
-#define JKJ_HEADER_RANDOM_FLOAT
-
-#include "dragonbox/dragonbox.h"
-#include <algorithm>
-#include <cstring>
-#include <random>
-#include <stdexcept>
-#include <string>
-#include <vector>
-
-// For correct seeding
-class repeating_seed_seq {
-public:
-    using result_type = std::uint32_t;
-
-    repeating_seed_seq() : stored_values{0} {}
-    template <class InputIterator>
-    repeating_seed_seq(InputIterator first, InputIterator last) : stored_values(first, last) {}
-    template <class T>
-    repeating_seed_seq(std::initializer_list<T> list) : stored_values(list) {}
-
-    repeating_seed_seq(std::random_device&& rd, std::size_t count) {
-        stored_values.resize(count);
-        for (auto& elem : stored_values)
-            elem = rd();
-    }
-
-    template <class RandomAccessIterator>
-    void generate(RandomAccessIterator first, RandomAccessIterator last) {
-        auto count = last - first;
-        auto q = count / stored_values.size();
-        for (std::size_t i = 0; i < q; ++i) {
-            std::copy_n(stored_values.cbegin(), stored_values.size(), first);
-            first += stored_values.size();
-        }
-        count -= q * stored_values.size();
-        std::copy_n(stored_values.cbegin(), count, first);
-    }
-
-    std::size_t size() const noexcept { return stored_values.size(); }
-
-    template <class OutputIterator>
-    void param(OutputIterator first) const {
-        std::copy(stored_values.begin(), stored_values.end(), first);
-    }
-
-private:
-    std::vector<std::uint32_t> stored_values;
-};
-
-inline std::mt19937_64 generate_correctly_seeded_mt19937_64() {
-    repeating_seed_seq seed_seq{std::random_device{}, std::mt19937_64::state_size *
-                                                          std::mt19937_64::word_size /
-                                                          (sizeof(std::uint32_t) * 8)};
-    return std::mt19937_64{seed_seq};
-}
-
-template <class Float, class RandGen>
-Float uniformly_randomly_generate_finite_float(RandGen& rg) {
-    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
-    using ieee754_format_info = typename ieee754_traits::format;
-    using carrier_uint = typename ieee754_traits::carrier_uint;
-    using uniform_distribution = std::uniform_int_distribution<carrier_uint>;
-
-    // Generate sign bit
-    auto sign_bit = uniform_distribution{0, 1}(rg);
-
-    // Generate exponent bits
-    auto exponent_bits =
-        uniform_distribution{0, (carrier_uint(1) << ieee754_format_info::exponent_bits) - 2}(rg);
-
-    // Generate significand bits
-    auto significand_bits =
-        uniform_distribution{0, (carrier_uint(1) << ieee754_format_info::significand_bits) - 1}(rg);
-
-    auto bit_representation = (sign_bit << (ieee754_traits::carrier_bits - 1)) |
-                              (exponent_bits << (ieee754_format_info::significand_bits)) |
-                              significand_bits;
-
-    return ieee754_traits::carrier_to_float(bit_representation);
-}
-
-template <class Float, class RandGen>
-Float uniformly_randomly_generate_general_float(RandGen& rg) {
-    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
-    using carrier_uint = typename ieee754_traits::carrier_uint;
-    using uniform_distribution = std::uniform_int_distribution<carrier_uint>;
-
-    // Generate sign bit
-    auto bit_representation = uniform_distribution{0, std::numeric_limits<carrier_uint>::max()}(rg);
-    return ieee754_traits::carrier_to_float(bit_representation);
-}
-
-// This function tries to uniformly randomly generate a float number with the
-// given number of decimal digits, and the end-result is not perfectly bias-free.
-// However, I don't think there is an easy way to do it correctly.
-template <class Float, class RandGen>
-Float randomly_generate_float_with_given_digits(unsigned int digits, RandGen& rg) {
-    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
-    using carrier_uint = typename ieee754_traits::carrier_uint;
-    using signed_int_t = std::make_signed_t<carrier_uint>;
-
-    assert(digits >= 1);
-    assert(digits <= ieee754_traits::format::decimal_digits);
-
-    // Generate sign uniformly randomly
-    signed_int_t sign = std::uniform_int_distribution<signed_int_t>{0, 1}(rg) == 0 ? 1 : -1;
-
-
-    // Try to generate significand uniformly randomly
-    Float result;
-    signed_int_t from = 0, to = 9;
-    if (digits > 1) {
-        from = 1;
-        for (unsigned int e = 1; e < digits - 1; ++e) {
-            from *= 10;
-        }
-        to = from * 10 - 1;
-    }
-
-    while (true) {
-        auto significand = std::uniform_int_distribution<signed_int_t>{from, to}(rg);
-        if (digits > 1) {
-            significand *= 10;
-            significand += std::uniform_int_distribution<signed_int_t>{1, 9}(rg);
-        }
-
-        // Generate exponent uniformly randomly
-        auto exp = std::uniform_int_distribution<int>{
-            std::numeric_limits<Float>::min_exponent10 - (int(digits) - 1),
-            std::numeric_limits<Float>::max_exponent10 - (int(digits) - 1)}(rg);
-
-        // Cook up
-        auto str = std::to_string(sign * significand) + 'e' + std::to_string(exp);
-
-        if constexpr (std::is_same_v<Float, float>) {
-            result = std::strtof(str.c_str(), nullptr);
-        }
-        else {
-            result = std::strtod(str.c_str(), nullptr);
-        }
-
-        if (!std::isfinite(result)) {
-            continue;
-        }
-
-        // Discard if a shorter representation exists
-        // We don't need to care about sign and correct rounding here
-        if (from != 0) {
-            auto roundtrip = jkj::dragonbox::to_decimal(
-                result, jkj::dragonbox::policy::sign::ignore,
-                jkj::dragonbox::policy::decimal_to_binary_rounding::nearest_to_even,
-                jkj::dragonbox::policy::binary_to_decimal_rounding::do_not_care);
-            if (roundtrip.significand <= carrier_uint(from * 10)) {
-                continue;
-            }
-        }
-        break;
-    }
-
-    return result;
-}
-
-#endif
+// Copyright 2020-2021 Junekey Jeon
+//
+// The contents of this file may be used under the terms of
+// the Apache License v2.0 with LLVM Exceptions.
+//
+//    (See accompanying file LICENSE-Apache or copy at
+//     https://llvm.org/foundation/relicensing/LICENSE.txt)
+//
+// Alternatively, the contents of this file may be used under the terms of
+// the Boost Software License, Version 1.0.
+//    (See accompanying file LICENSE-Boost or copy at
+//     https://www.boost.org/LICENSE_1_0.txt)
+//
+// Unless required by applicable law or agreed to in writing, this software
+// is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied.
+
+#ifndef JKJ_HEADER_RANDOM_FLOAT
+#define JKJ_HEADER_RANDOM_FLOAT
+
+#include "dragonbox/dragonbox.h"
+#include <algorithm>
+#include <cstring>
+#include <random>
+#include <stdexcept>
+#include <string>
+#include <vector>
+
+// For correct seeding
+class repeating_seed_seq {
+public:
+    using result_type = std::uint32_t;
+
+    repeating_seed_seq() : stored_values{0} {}
+    template <class InputIterator>
+    repeating_seed_seq(InputIterator first, InputIterator last) : stored_values(first, last) {}
+    template <class T>
+    repeating_seed_seq(std::initializer_list<T> list) : stored_values(list) {}
+
+    repeating_seed_seq(std::random_device&& rd, std::size_t count) {
+        stored_values.resize(count);
+        for (auto& elem : stored_values)
+            elem = rd();
+    }
+
+    template <class RandomAccessIterator>
+    void generate(RandomAccessIterator first, RandomAccessIterator last) {
+        auto count = last - first;
+        auto q = count / stored_values.size();
+        for (std::size_t i = 0; i < q; ++i) {
+            std::copy_n(stored_values.cbegin(), stored_values.size(), first);
+            first += stored_values.size();
+        }
+        count -= q * stored_values.size();
+        std::copy_n(stored_values.cbegin(), count, first);
+    }
+
+    std::size_t size() const noexcept { return stored_values.size(); }
+
+    template <class OutputIterator>
+    void param(OutputIterator first) const {
+        std::copy(stored_values.begin(), stored_values.end(), first);
+    }
+
+private:
+    std::vector<std::uint32_t> stored_values;
+};
+
+inline std::mt19937_64 generate_correctly_seeded_mt19937_64() {
+    repeating_seed_seq seed_seq{std::random_device{}, std::mt19937_64::state_size *
+                                                          std::mt19937_64::word_size /
+                                                          (sizeof(std::uint32_t) * 8)};
+    return std::mt19937_64{seed_seq};
+}
+
+template <class Float, class RandGen>
+Float uniformly_randomly_generate_finite_float(RandGen& rg) {
+    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
+    using ieee754_format_info = typename ieee754_traits::format;
+    using carrier_uint = typename ieee754_traits::carrier_uint;
+    using uniform_distribution = std::uniform_int_distribution<carrier_uint>;
+
+    // Generate sign bit
+    auto sign_bit = uniform_distribution{0, 1}(rg);
+
+    // Generate exponent bits
+    auto exponent_bits =
+        uniform_distribution{0, (carrier_uint(1) << ieee754_format_info::exponent_bits) - 2}(rg);
+
+    // Generate significand bits
+    auto significand_bits =
+        uniform_distribution{0, (carrier_uint(1) << ieee754_format_info::significand_bits) - 1}(rg);
+
+    auto bit_representation = (sign_bit << (ieee754_traits::carrier_bits - 1)) |
+                              (exponent_bits << (ieee754_format_info::significand_bits)) |
+                              significand_bits;
+
+    return ieee754_traits::carrier_to_float(bit_representation);
+}
+
+template <class Float, class RandGen>
+Float uniformly_randomly_generate_general_float(RandGen& rg) {
+    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
+    using carrier_uint = typename ieee754_traits::carrier_uint;
+    using uniform_distribution = std::uniform_int_distribution<carrier_uint>;
+
+    // Generate sign bit
+    auto bit_representation = uniform_distribution{0, std::numeric_limits<carrier_uint>::max()}(rg);
+    return ieee754_traits::carrier_to_float(bit_representation);
+}
+
+// This function tries to uniformly randomly generate a float number with the
+// given number of decimal digits, and the end-result is not perfectly bias-free.
+// However, I don't think there is an easy way to do it correctly.
+template <class Float, class RandGen>
+Float randomly_generate_float_with_given_digits(unsigned int digits, RandGen& rg) {
+    using ieee754_traits = jkj::dragonbox::default_float_traits<Float>;
+    using carrier_uint = typename ieee754_traits::carrier_uint;
+    using signed_int_t = std::make_signed_t<carrier_uint>;
+
+    assert(digits >= 1);
+    assert(digits <= ieee754_traits::format::decimal_digits);
+
+    // Generate sign uniformly randomly
+    signed_int_t sign = std::uniform_int_distribution<signed_int_t>{0, 1}(rg) == 0 ? 1 : -1;
+
+
+    // Try to generate significand uniformly randomly
+    Float result;
+    signed_int_t from = 0, to = 9;
+    if (digits > 1) {
+        from = 1;
+        for (unsigned int e = 1; e < digits - 1; ++e) {
+            from *= 10;
+        }
+        to = from * 10 - 1;
+    }
+
+    while (true) {
+        auto significand = std::uniform_int_distribution<signed_int_t>{from, to}(rg);
+        if (digits > 1) {
+            significand *= 10;
+            significand += std::uniform_int_distribution<signed_int_t>{1, 9}(rg);
+        }
+
+        // Generate exponent uniformly randomly
+        auto exp = std::uniform_int_distribution<int>{
+            std::numeric_limits<Float>::min_exponent10 - (int(digits) - 1),
+            std::numeric_limits<Float>::max_exponent10 - (int(digits) - 1)}(rg);
+
+        // Cook up
+        auto str = std::to_string(sign * significand) + 'e' + std::to_string(exp);
+
+        if constexpr (std::is_same_v<Float, float>) {
+            result = std::strtof(str.c_str(), nullptr);
+        }
+        else {
+            result = std::strtod(str.c_str(), nullptr);
+        }
+
+        if (!std::isfinite(result)) {
+            continue;
+        }
+
+        // Discard if a shorter representation exists
+        // We don't need to care about sign and correct rounding here
+        if (from != 0) {
+            auto roundtrip = jkj::dragonbox::to_decimal(
+                result, jkj::dragonbox::policy::sign::ignore,
+                jkj::dragonbox::policy::decimal_to_binary_rounding::nearest_to_even,
+                jkj::dragonbox::policy::binary_to_decimal_rounding::do_not_care);
+            if (roundtrip.significand <= carrier_uint(from * 10)) {
+                continue;
+            }
+        }
+        break;
+    }
+
+    return result;
+}
+
+#endif

package/vendor/dragonbox/subproject/common/include/rational_continued_fractions.h

@@ -1,57 +1,57 @@
-// Copyright 2022 Junekey Jeon
-//
-// The contents of this file may be used under the terms of
-// the Apache License v2.0 with LLVM Exceptions.
-//
-//    (See accompanying file LICENSE-Apache or copy at
-//     https://llvm.org/foundation/relicensing/LICENSE.txt)
-//
-// Alternatively, the contents of this file may be used under the terms of
-// the Boost Software License, Version 1.0.
-//    (See accompanying file LICENSE-Boost or copy at
-//     https://www.boost.org/LICENSE_1_0.txt)
-//
-// Unless required by applicable law or agreed to in writing, this software
-// is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
-// KIND, either express or implied.
-
-#ifndef JKJ_HEADER_RATIONAL_CONTINUED_FRACTIONS
-#define JKJ_HEADER_RATIONAL_CONTINUED_FRACTIONS
-
-#include "continued_fractions.h"
-#include <cassert>
-#include <cstdlib>
-
-namespace jkj {
-    template <class UInt>
-    class rational_continued_fractions
-        : public continued_fractions<rational_continued_fractions<UInt>, UInt> {
-        using crtp_base = continued_fractions<rational_continued_fractions<UInt>, UInt>;
-        friend crtp_base;
-
-        UInt prev_error_;
-        UInt curr_error_;
-
-        UInt compute_next_coefficient() {
-            using std::div;
-            auto div_result = div(prev_error_, curr_error_);
-            prev_error_ = static_cast<UInt&&>(curr_error_);
-            curr_error_ = static_cast<UInt&&>(div_result.rem);
-
-            if (curr_error_ == 0) {
-                crtp_base::set_terminate_flag();
-            }
-
-            return static_cast<UInt&&>(div_result.quot);
-        }
-
-    public:
-        rational_continued_fractions(unsigned_rational<UInt> r)
-            : prev_error_{static_cast<UInt&&>(r.numerator)}, curr_error_{static_cast<UInt&&>(
-                                                                 r.denominator)} {
-            assert(curr_error_ != 0);
-        }
-    };
-}
-
+// Copyright 2022 Junekey Jeon
+//
+// The contents of this file may be used under the terms of
+// the Apache License v2.0 with LLVM Exceptions.
+//
+//    (See accompanying file LICENSE-Apache or copy at
+//     https://llvm.org/foundation/relicensing/LICENSE.txt)
+//
+// Alternatively, the contents of this file may be used under the terms of
+// the Boost Software License, Version 1.0.
+//    (See accompanying file LICENSE-Boost or copy at
+//     https://www.boost.org/LICENSE_1_0.txt)
+//
+// Unless required by applicable law or agreed to in writing, this software
+// is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied.
+
+#ifndef JKJ_HEADER_RATIONAL_CONTINUED_FRACTIONS
+#define JKJ_HEADER_RATIONAL_CONTINUED_FRACTIONS
+
+#include "continued_fractions.h"
+#include <cassert>
+#include <cstdlib>
+
+namespace jkj {
+    template <class UInt>
+    class rational_continued_fractions
+        : public continued_fractions<rational_continued_fractions<UInt>, UInt> {
+        using crtp_base = continued_fractions<rational_continued_fractions<UInt>, UInt>;
+        friend crtp_base;
+
+        UInt prev_error_;
+        UInt curr_error_;
+
+        UInt compute_next_coefficient() {
+            using std::div;
+            auto div_result = div(prev_error_, curr_error_);
+            prev_error_ = static_cast<UInt&&>(curr_error_);
+            curr_error_ = static_cast<UInt&&>(div_result.rem);
+
+            if (curr_error_ == 0) {
+                crtp_base::set_terminate_flag();
+            }
+
+            return static_cast<UInt&&>(div_result.quot);
+        }
+
+    public:
+        rational_continued_fractions(unsigned_rational<UInt> r)
+            : prev_error_{static_cast<UInt&&>(r.numerator)}, curr_error_{static_cast<UInt&&>(
+                                                                 r.denominator)} {
+            assert(curr_error_ != 0);
+        }
+    };
+}
+
 #endif