[libc++][test] Add test coverage for codecvt<char(16|32)_t, char8_t, …

…mbstate_t>

This change adds test coverage for the `codecvt<char16_t, char8_t, mbstate_t>` and `codecvt<char32_t, char8_t, mbstate_t>` ctype facets added to the C++20 WD by [P0482R6](https://wg21.link/P0428R6). Note that libc++ does not implement these facets despite implementing the remainder of P0482, presumably for ABI reasons, so these tests are marked `UNSUPPORTED: libc++`.

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/ctor_char16_t_char8_t.pass.cpp

@@ -0,0 +1,53 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// explicit codecvt(size_t refs = 0);
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+
+struct my_facet : F {
+  static int count;
+
+  explicit my_facet(std::size_t refs = 0) : F(refs) { ++count; }
+
+  ~my_facet() { --count; }
+};
+
+int my_facet::count = 0;
+
+int main(int, char**) {
+  {
+    std::locale l(std::locale::classic(), new my_facet);
+    assert(my_facet::count == 1);
+  }
+  assert(my_facet::count == 0);
+  {
+    my_facet f(1);
+    assert(my_facet::count == 1);
+    {
+      std::locale l(std::locale::classic(), &f);
+      assert(my_facet::count == 1);
+    }
+    assert(my_facet::count == 1);
+  }
+  assert(my_facet::count == 0);
+
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/ctor_char32_t_char8_t.pass.cpp

@@ -0,0 +1,53 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char32_t, char8_t, mbstate_t>
+
+// explicit codecvt(size_t refs = 0);
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+using F = std::codecvt<char32_t, char8_t, std::mbstate_t>;
+
+struct my_facet : F {
+  static int count;
+
+  explicit my_facet(std::size_t refs = 0) : F(refs) { ++count; }
+
+  ~my_facet() { --count; }
+};
+
+int my_facet::count = 0;
+
+int main(int, char**) {
+  {
+    std::locale l(std::locale::classic(), new my_facet);
+    assert(my_facet::count == 1);
+  }
+  assert(my_facet::count == 0);
+  {
+    my_facet f(1);
+    assert(my_facet::count == 1);
+    {
+      std::locale l(std::locale::classic(), &f);
+      assert(my_facet::count == 1);
+    }
+    assert(my_facet::count == 1);
+  }
+  assert(my_facet::count == 0);
+
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_always_noconv.pass.cpp

@@ -0,0 +1,29 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// bool always_noconv() const noexcept;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  assert(!f.always_noconv());
+  static_assert(noexcept(f.always_noconv()));
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_encoding.pass.cpp

@@ -0,0 +1,29 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// int encoding() const noexcept;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  assert(f.encoding() == 0);
+  static_assert(noexcept(f.encoding()));
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_in.pass.cpp

@@ -0,0 +1,39 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// result in(stateT& state,
+//           const externT* from, const externT* from_end, const externT*& from_next,
+//           internT* to, internT* to_end, internT*& to_next) const;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F::extern_type from[] = u8"some text";
+  F::intern_type to[9];
+  const F& f = std::use_facet<F>(std::locale::classic());
+  std::mbstate_t mbs = {};
+  const F::extern_type* from_next = nullptr;
+  F::intern_type* to_next = nullptr;
+  assert(f.in(mbs, from, from + 9, from_next, to, to + 9, to_next) == F::ok);
+  assert(from_next - from == 9);
+  assert(to_next - to == 9);
+  for (unsigned i = 0; i < 9; ++i)
+    assert(to[i] == from[i]);
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_length.pass.cpp

@@ -0,0 +1,34 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// int length(stateT& state, const externT* from, const externT* from_end, size_t max) const;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  std::mbstate_t mbs = {};
+  const char8_t from[] = u8"some text";
+  assert(f.length(mbs, from, from + 10, 0) == 0);
+  assert(f.length(mbs, from, from + 10, 8) == 8);
+  assert(f.length(mbs, from, from + 10, 9) == 9);
+  assert(f.length(mbs, from, from + 10, 10) == 10);
+  assert(f.length(mbs, from, from + 10, 100) == 10);
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_max_length.pass.cpp

@@ -0,0 +1,29 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// int max_length() const noexcept;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  assert(f.max_length() == 4);
+  static_assert(noexcept(f.max_length()));
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_out.pass.cpp

@@ -0,0 +1,40 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// result out(stateT& state,
+//            const internT* from, const internT* from_end, const internT*& from_next,
+//            externT* to, externT* to_end, externT*& to_next) const;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  F::intern_type from[9] = {'s', 'o', 'm', 'e', ' ', 't', 'e', 'x', 't'};
+  F::extern_type to[9] = {0};
+  std::mbstate_t mbs = {};
+  const F::intern_type* from_next = nullptr;
+  F::extern_type* to_next = nullptr;
+  F::result r = f.out(mbs, from, from + 9, from_next, to, to + 9, to_next);
+  assert(r == F::ok);
+  assert(from_next - from == 9);
+  assert(to_next - to == 9);
+  for (unsigned i = 0; i < 9; ++i)
+    assert(to[i] == from[i]);
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char16_t_char8_t_unshift.pass.cpp

@@ -0,0 +1,33 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char16_t, char8_t, mbstate_t>
+
+// result unshift(stateT& state,
+//                externT* to, externT* to_end, externT*& to_next) const;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char16_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  std::mbstate_t mbs = {};
+  char8_t to[3];
+  char8_t* to_next = nullptr;
+  assert(f.unshift(mbs, to, to + 3, to_next) == F::noconv);
+  assert(to_next == to);
+  return 0;
+}

libcxx/test/std/localization/locale.categories/category.ctype/locale.codecvt/locale.codecvt.members/char32_t_char8_t_always_noconv.pass.cpp

@@ -0,0 +1,29 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// <locale>
+
+// template <> class codecvt<char32_t, char8_t, mbstate_t>
+
+// bool always_noconv() const noexcept;
+
+// UNSUPPORTED: c++98, c++03, c++11, c++14, c++17
+
+// C++20 codecvt specializations for char8_t are not yet implemented:
+// UNSUPPORTED: libc++
+
+#include <cassert>
+#include <locale>
+
+int main(int, char**) {
+  using F = std::codecvt<char32_t, char8_t, std::mbstate_t>;
+  const F& f = std::use_facet<F>(std::locale::classic());
+  assert(!f.always_noconv());
+  static_assert(noexcept(f.always_noconv()));
+  return 0;
+}