Release 1.17.3

.forgejo/workflows/do-nothing.yml

@@ -0,0 +1,10 @@
+# Copyright (c) 2026 Morwenn
+# SPDX-License-Identifier: MIT
+
+# Codeberg seems to run the repository's GitHub actions when
+# there is no Forgejo workflow: the goa of this file is to
+# skip that callback
+name: Do Nothing
+
+on:
+  workflow_dispatch:

.github/workflows/build-macos.yml

@@ -40,7 +40,7 @@ jobs:
             sanitize: address,undefined
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v7
 
       - name: Configure CMake
         working-directory: ${{runner.workspace}}

.github/workflows/build-mingw.yml

@@ -1,4 +1,4 @@
-# Copyright (c) 2021-2025 Morwenn
+# Copyright (c) 2021-2026 Morwenn
 # SPDX-License-Identifier: MIT
 
 name: MinGW-w64 Builds
@@ -31,7 +31,7 @@ jobs:
         build_type: [Debug, Release]
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v7
 
       - name: Configure CMake
         shell: pwsh

.github/workflows/build-msvc.yml

@@ -1,4 +1,4 @@
-# Copyright (c) 2021-2025 Morwenn
+# Copyright (c) 2021-2026 Morwenn
 # SPDX-License-Identifier: MIT
 
 name: MSVC Builds
@@ -36,15 +36,14 @@ jobs:
           build_tools: '-T ClangCL'
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v7
 
       - name: Configure CMake
         shell: pwsh
         working-directory: ${{runner.workspace}}
         run: |
           cmake -H${{github.event.repository.name}} -Bbuild `
                 -DCMAKE_CONFIGURATION_TYPES=${{matrix.config.build_type}} `
-                -DCMAKE_BUILD_TYPE=${{matrix.config.build_type}} `
                 -G"Visual Studio 17 2022" -A x64 ${{matrix.config.build_tools}} `
                 -DCPPSORT_BUILD_EXAMPLES=ON
 

.github/workflows/build-ubuntu.yml

@@ -1,4 +1,4 @@
-# Copyright (c) 2021-2025 Morwenn
+# Copyright (c) 2021-2026 Morwenn
 # SPDX-License-Identifier: MIT
 
 name: Ubuntu Builds
@@ -41,7 +41,7 @@ jobs:
             sanitize: address,undefined
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v7
 
       - name: Install GCC
         if: ${{matrix.cxx == 'g++-9'}}

.github/workflows/code-coverage.yml

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2025 Morwenn
+# Copyright (c) 2020-2026 Morwenn
 # SPDX-License-Identifier: MIT
 
 name: Coverage Upload to Codecov
@@ -23,7 +23,7 @@ jobs:
 
     steps:
       - name: Checkout project
-        uses: actions/checkout@v4
+        uses: actions/checkout@v7
 
       - name: Install LCOV
         run: sudo apt-get install -y lcov

.github/workflows/mirror-to-codeberg.yml

@@ -1,4 +1,4 @@
-# Copyright (c) 2025 Morwenn
+# Copyright (c) 2025-2026 Morwenn
 # SPDX-License-Identifier: MIT
 
 name: Mirror Commits to Codeberg
@@ -13,7 +13,7 @@ jobs:
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v5
+        uses: actions/checkout@v7
         with:
           fetch-depth: 0
 

CMakeLists.txt

@@ -5,7 +5,7 @@ cmake_minimum_required(VERSION 3.8.0)
 
 list(APPEND CMAKE_MODULE_PATH ${CMAKE_CURRENT_SOURCE_DIR}/cmake)
 
-project(cpp-sort VERSION 1.17.2 LANGUAGES CXX)
+project(cpp-sort VERSION 1.17.3 LANGUAGES CXX)
 
 include(CMakePackageConfigHelpers)
 include(cpp-sort-utils)

README.md

@@ -1,7 +1,7 @@
 ![cpp-sort logo](docs/images/cpp-sort-logo.svg)
 
-[![Latest Release](https://img.shields.io/badge/release-1.17.2-blue.svg)](https://github.com/Morwenn/cpp-sort/releases/tag/1.17.2)
-[![Conan Package](https://img.shields.io/badge/conan-cpp--sort%2F1.17.2-blue.svg)](https://conan.io/center/recipes/cpp-sort?version=1.17.2)
+[![Latest Release](https://img.shields.io/badge/release-1.17.3-blue.svg)](https://github.com/Morwenn/cpp-sort/releases/tag/1.17.3)
+[![Conan Package](https://img.shields.io/badge/conan-cpp--sort%2F1.17.3-blue.svg)](https://conan.io/center/recipes/cpp-sort?version=1.17.3)
 [![Code Coverage](https://codecov.io/gh/Morwenn/cpp-sort/branch/develop/graph/badge.svg)](https://codecov.io/gh/Morwenn/cpp-sort)
 [![Pitchfork Layout](https://img.shields.io/badge/standard-PFL-orange.svg)](https://github.com/vector-of-bool/pitchfork)
 

conanfile.py

@@ -16,7 +16,7 @@
 
 class CppSortConan(ConanFile):
     name = "cpp-sort"
-    version = "1.17.2"
+    version = "1.17.3"
     description = "Sorting algorithms & related tools"
     license = "MIT"
     url = "https://github.com/Morwenn/cpp-sort"

docs/Changelog.md

@@ -68,6 +68,11 @@ When compiled with C++17, **cpp-sort** might gain a few additional features depe
 
 When compiled with C++20, **cpp-sort** might gain a few additional features depending on the level of C++20 support provided by the compiler. The availability of those features depends on the presence of corresponding [feature-testing macros][feature-test-macros] when possible, even though some checks are more granular. Don't hesitate to open an issue if your compiler and standard library supports one of those features but it doesn't seem to work in **cpp-sort**.
 
+**Correctness improvements:**
+* When possible, the total order predicates [`total_less` and `total_greater`][total-greater-less] use [`std::strong_order`][std-strong-order] to compare floating-point numbers, leading to a supposedly correct implementation of the IEEE *totalOrder* operator. In C++17 and earlier, those comparators don't actually implement a proper total order: they are unable to differentiate between quiet and signaling NaNs, and connot read NaN payloads.
+
+    The feature-test macro `__cpp_lib_three_way_comparison` can be used to check whether `std::strong_order` is available.
+
 **New features:**
 * When available, [`std::identity`][std-identity] benefits from dedicated support wherever [`utility::identity`][cpp-sort-function-objects] is supported, with equivalent semantics.
 
@@ -110,5 +115,7 @@ When compiled with C++20, **cpp-sort** might gain a few additional features depe
   [std-ranges-greater]: https://en.cppreference.com/w/cpp/utility/functional/ranges/greater
   [std-ranges-less]: https://en.cppreference.com/w/cpp/utility/functional/ranges/less
   [std-string-view]: https://en.cppreference.com/w/cpp/string/basic_string_view
+  [std-strong-order]: https://en.cppreference.com/w/cpp/utility/compare/strong_order
   [std-swap]: https://en.cppreference.com/w/cpp/algorithm/swap
+  [total-greater-less]: Comparators.md#total-order-comparators
   [utility-iter-move]: Miscellaneous-utilities.md#iter_move-and-iter_swap

docs/Comparators.md

@@ -11,7 +11,7 @@ Every non-refined comparator described below is also a [transparent comparator][
 #include <cpp-sort/comparators/total_less.h>
 ```
 
-The comparators `total_less` and `total_order` are [customization points][custom-point] implementing a [total order][total-order], inspired by the similar functions described in [P0100][P0100]. The provided functions handle built-in integer out of the box (using the built-in relational operators) and implement IEEE 754's `totalOrder` for floating point numbers (from lesser to greater):
+The comparators `total_less` and `total_order` are [customization points][custom-point] implementing a [total order][total-order], inspired by the similar functions described in [P0100][P0100]. The provided functions handle built-in integer types out of the box (using the built-in relational operators) and attempt to implement IEEE 754's `totalOrder` for floating point numbers (from lesser to greater):
 * positive quiet NaNs
 * positive signaling NaNs
 * positive infinity
@@ -23,14 +23,18 @@ The comparators `total_less` and `total_order` are [customization points][custom
 * negative signaling NaNs
 * negative quiet NaNs
 
-That said, the comparators are currently unable to discriminate between quiet and signaling NaNs, so they are considered to be *equivalent*. When it doesn't handle a type natively and ADL doesn't find any suitable `total_less` function in a class namespace, `cppsort::total_less` does *not* fall back to `operator<`; see [P0100][P0100] for the rationale (it applies to the whole `total_*` family of customization points).
+That said, the comparators are unable to discriminate between quiet and signaling NaNs, and have no way to inspect the bit patterns of NaNs to compare them. All NaNs that share a same sign are therefore considered to be *equivalent* in C++17. In C++20 however, the comparators rely on [`std::strong_order`][std-strong-order] when available, which correctly implements `totalOrder` for IEEE 754 floating point numbers.
+
+When it doesn't handle a type natively and ADL doesn't find any suitable `total_less` function in a class namespace, `cppsort::total_less` does *not* fall back to `operator<`; see [P0100][P0100] for the rationale (it applies to the whole `total_*` family of customization points).
 
 Total order comparators are considered as [generating branchless code][branchless-traits] when comparing instances of a type that satisfies [`std::is_integral`][std-is-integral].
 
 *Changed in version 1.5.0:* `total_greater` and `total_less` are respectively of type `total_greater_t` and `total_less_t`.
 
 *Changed in version 1.13.1:* support for `[un]signed __int128`.
 
+*Changed in version 1.17.3:* `total_less` and `total_greater` now correctly implement IEEE 754 `totalOrder` in C++20 mode, when the feature-test macro `__cpp_lib_three_way_comparison` is defined.
+
 ### Weak order comparators
 
 ```cpp

docs/Fixed-size-sorters.md

@@ -153,7 +153,7 @@ Size | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16
 :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-:
 **CEs** | 0 | 1 | 3 | 5 | 9 | 12 | 16 | 19 | 25 | 29 | 35 | 39 | 45 | 51 | 56 | 60
 **Size** | **17** | **18** | **19** | **20** | **21** | **22** | **23** | **24** | **25** | **26** | **27** | **28** | **29** | **30** | **31** | **32**
-**CEs** | 71 | 77 | 85 | 91 | 99 | 106 | 114 | 120 | 130 | 139 | 147 | 155 | 164 | 172 | 180 | 185
+**CEs** | 71 | 77 | 85 | 91 | 99 | 106 | 114 | 120 | 130 | 138 | 147 | 155 | 164 | 172 | 180 | 185
 **Size** | **33** | **34** | **35** | **36** | **37** | **38** | **39** | **40** | **41** | **42** | **43** | **44** | **45** | **46** | **47** | **48**
 **CEs** | 199 | 209 | 220 | 227 | 240 | 250 | 259 | 265 | 282 | 291 | 303 | 309 | 324 | 332 | 340 | 346
 **Size** | **49** | **50** | **51** | **52** | **53** | **54** | **55** | **56** | **57** | **58** | **59** | **60** | **61** | **62** | **63** | **64**

docs/Home.md

@@ -1,6 +1,6 @@
 ![cpp-sort logo](images/cpp-sort-logo.svg)
 
-Welcome to the **cpp-sort 1.17.2** documentation!
+Welcome to the **cpp-sort 1.17.3** documentation!
 
 This wiki contains documentation about the library: basic documentation about the many sorting tools and how to use them, documentation about the additional utilities provided by the library, as well as a few tutorials about writing your own sorters or sorter adapters. This main page explains a few general things that didn't quite fit in other parts of the documentation.
 

docs/Tooling.md

@@ -63,10 +63,10 @@ Note: when `CPPSORT_ENABLE_AUDITS` is `ON`, assertions in the library are enable
 conan search cpp-sort --remote=conancenter
 ```
 
-And then install any version to your local cache as follows (here with version 1.17.2):
+And then install any version to your local cache as follows (here with version 1.17.3):
 
 ```sh
-conan install --requires=cpp-sort/1.17.2
+conan install --requires=cpp-sort/1.17.3
 ```
 
 The packages downloaded from conan-center are minimal and only contain the files required to use **cpp-sort** as a library: the headers, CMake files and licensing information. If you need anything else you have to create your own package with the `conanfile.py` available in this repository.

include/cpp-sort/comparators/partial_greater.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2021 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_PARTIAL_GREATER_H_
@@ -33,9 +33,12 @@ namespace cppsort
         // Generic overload: a weak order is also a partial order
 
         template<typename T>
-        auto partial_greater(const T& lhs, const T& rhs)
+        constexpr auto partial_greater(const T& lhs, const T& rhs)
             noexcept(noexcept(cppsort::weak_greater(lhs, rhs)))
-            -> decltype(cppsort::weak_greater(lhs, rhs))
+            -> detail::enable_if_t<
+                not std::is_floating_point<T>::value,
+                decltype(cppsort::weak_greater(lhs, rhs))
+            >
         {
             return cppsort::weak_greater(lhs, rhs);
         }

include/cpp-sort/comparators/partial_less.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2021 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_PARTIAL_LESS_H_
@@ -33,9 +33,12 @@ namespace cppsort
         // Generic overload: a weak order is also a partial order
 
         template<typename T>
-        auto partial_less(const T& lhs, const T& rhs)
+        constexpr auto partial_less(const T& lhs, const T& rhs)
             noexcept(noexcept(cppsort::weak_less(lhs, rhs)))
-            -> decltype(cppsort::weak_less(lhs, rhs))
+            -> detail::enable_if_t<
+                not std::is_floating_point<T>::value,
+                decltype(cppsort::weak_less(lhs, rhs))
+            >
         {
             return cppsort::weak_less(lhs, rhs);
         }

include/cpp-sort/comparators/total_greater.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2022 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_TOTAL_GREATER_H_
@@ -8,14 +8,19 @@
 ////////////////////////////////////////////////////////////
 // Headers
 ////////////////////////////////////////////////////////////
-#include <cmath>
 #include <type_traits>
 #include <utility>
 #include <cpp-sort/utility/branchless_traits.h>
 #include <cpp-sort/utility/static_const.h>
-#include "../detail/floating_point_weight.h"
 #include "../detail/type_traits.h"
 
+#if defined(__cpp_lib_three_way_comparison)
+#   include <compare>
+#else
+#   include <cmath>
+#   include "../detail/floating_point_weight.h"
+#endif
+
 namespace cppsort
 {
     namespace detail
@@ -37,6 +42,11 @@ namespace cppsort
         auto total_greater(T lhs, T rhs)
             -> detail::enable_if_t<std::is_floating_point<T>::value, bool>
         {
+#if defined(__cpp_lib_three_way_comparison)
+            // Full implementation of totalOrder
+            return std::is_gt(std::strong_order(lhs, rhs));
+#else
+            // Approximation, unable to differentiate NaNs
             if (std::isfinite(lhs) && std::isfinite(rhs)) {
                 if (lhs == 0 && rhs == 0) {
                     return std::signbit(rhs) && not std::signbit(lhs);
@@ -47,6 +57,7 @@ namespace cppsort
             int lhs_weight = total_weight(lhs);
             int rhs_weight = total_weight(rhs);
             return lhs_weight > rhs_weight;
+#endif
         }
 
         ////////////////////////////////////////////////////////////

include/cpp-sort/comparators/total_less.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2022 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_TOTAL_LESS_H_
@@ -8,14 +8,20 @@
 ////////////////////////////////////////////////////////////
 // Headers
 ////////////////////////////////////////////////////////////
-#include <cmath>
 #include <type_traits>
 #include <utility>
+#include <version>
 #include <cpp-sort/utility/branchless_traits.h>
 #include <cpp-sort/utility/static_const.h>
-#include "../detail/floating_point_weight.h"
 #include "../detail/type_traits.h"
 
+#if defined(__cpp_lib_three_way_comparison)
+#   include <compare>
+#else
+#   include <cmath>
+#   include "../detail/floating_point_weight.h"
+#endif
+
 namespace cppsort
 {
     namespace detail
@@ -37,6 +43,11 @@ namespace cppsort
         auto total_less(T lhs, T rhs)
             -> detail::enable_if_t<std::is_floating_point<T>::value, bool>
         {
+#if defined(__cpp_lib_three_way_comparison)
+            // Full implementation of totalOrder
+            return std::is_lt(std::strong_order(lhs, rhs));
+#else
+            // Approximation, unable to differentiate NaNs
             if (std::isfinite(lhs) && std::isfinite(rhs)) {
                 if (lhs == 0 && rhs == 0) {
                     return std::signbit(lhs) && not std::signbit(rhs);
@@ -47,6 +58,7 @@ namespace cppsort
             int lhs_weight = total_weight(lhs);
             int rhs_weight = total_weight(rhs);
             return lhs_weight < rhs_weight;
+#endif
         }
 
         ////////////////////////////////////////////////////////////

include/cpp-sort/comparators/weak_greater.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2021 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_WEAK_GREATER_H_
@@ -41,9 +41,12 @@ namespace cppsort
         // Generic overload: a total order is also a weak order
 
         template<typename T>
-        auto weak_greater(const T& lhs, const T& rhs)
+        constexpr auto weak_greater(const T& lhs, const T& rhs)
             noexcept(noexcept(cppsort::total_greater(lhs, rhs)))
-            -> decltype(cppsort::total_greater(lhs, rhs))
+            -> detail::enable_if_t<
+                not std::is_floating_point<T>::value,
+                decltype(cppsort::total_greater(lhs, rhs))
+            >
         {
             return cppsort::total_greater(lhs, rhs);
         }

include/cpp-sort/comparators/weak_less.h

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016-2021 Morwenn
+ * Copyright (c) 2016-2026 Morwenn
  * SPDX-License-Identifier: MIT
  */
 #ifndef CPPSORT_COMPARATORS_WEAK_LESS_H_
@@ -41,9 +41,12 @@ namespace cppsort
         // Generic overload: a total order is also a weak order
 
         template<typename T>
-        auto weak_less(const T& lhs, const T& rhs)
+        constexpr auto weak_less(const T& lhs, const T& rhs)
             noexcept(noexcept(cppsort::total_less(lhs, rhs)))
-            -> decltype(cppsort::total_less(lhs, rhs))
+            -> detail::enable_if_t<
+                not std::is_floating_point<T>::value,
+                decltype(cppsort::total_less(lhs, rhs))
+            >
         {
             return cppsort::total_less(lhs, rhs);
         }