Implement ranges::ref_view

libcudacxx/include/cuda/std/__ranges/dangling.h

@@ -38,7 +38,7 @@ struct dangling
 #if !defined(_CCCL_NO_CONCEPTS)
 template <range _Rp>
 using borrowed_iterator_t = _If<borrowed_range<_Rp>, iterator_t<_Rp>, dangling>;
-#else // ^^^ C++20 ^^^ / vvv C++17 vvv
+#else // ^^^ !_CCCL_NO_CONCEPTS ^^^ / vvv _CCCL_NO_CONCEPTS vvv
 template <class _Rp>
 using borrowed_iterator_t = enable_if_t<range<_Rp>, _If<borrowed_range<_Rp>, iterator_t<_Rp>, dangling>>;
 #endif // _CCCL_NO_CONCEPTS

libcudacxx/include/cuda/std/__ranges/ref_view.h

@@ -0,0 +1,121 @@
+// -*- C++ -*-
+//===----------------------------------------------------------------------===//
+//
+// 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
+// SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES
+//
+//===----------------------------------------------------------------------===//
+#ifndef _LIBCUDACXX___RANGES_REF_VIEW_H
+#define _LIBCUDACXX___RANGES_REF_VIEW_H
+
+#include <cuda/std/detail/__config>
+
+#if defined(_CCCL_IMPLICIT_SYSTEM_HEADER_GCC)
+#  pragma GCC system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_CLANG)
+#  pragma clang system_header
+#elif defined(_CCCL_IMPLICIT_SYSTEM_HEADER_MSVC)
+#  pragma system_header
+#endif // no system header
+
+#include <cuda/std/__concepts/convertible_to.h>
+#include <cuda/std/__concepts/different_from.h>
+#include <cuda/std/__iterator/concepts.h>
+#include <cuda/std/__iterator/incrementable_traits.h>
+#include <cuda/std/__iterator/iterator_traits.h>
+#include <cuda/std/__memory/addressof.h>
+#include <cuda/std/__ranges/access.h>
+#include <cuda/std/__ranges/concepts.h>
+#include <cuda/std/__ranges/data.h>
+#include <cuda/std/__ranges/empty.h>
+#include <cuda/std/__ranges/enable_borrowed_range.h>
+#include <cuda/std/__ranges/size.h>
+#include <cuda/std/__ranges/view_interface.h>
+#include <cuda/std/__type_traits/enable_if.h>
+#include <cuda/std/__type_traits/is_object.h>
+#include <cuda/std/__utility/forward.h>
+
+_LIBCUDACXX_BEGIN_NAMESPACE_RANGES
+
+_LIBCUDACXX_BEGIN_NAMESPACE_RANGES_ABI
+
+template <class _Range>
+struct __conversion_tester
+{
+  _LIBCUDACXX_HIDE_FROM_ABI static void __fun(_Range&);
+  static void __fun(_Range&&) = delete;
+};
+
+template <class _Tp, class _Range>
+_CCCL_CONCEPT __convertible_to_lvalue =
+  _CCCL_REQUIRES_EXPR((_Tp, _Range))((__conversion_tester<_Range>::__fun(declval<_Tp>())));
+
+#if !defined(_CCCL_NO_CONCEPTS)
+
+template <range _Range>
+  requires is_object_v<_Range>
+#else // ^^^ !_CCCL_NO_CONCEPTS ^^^ / vvv _CCCL_NO_CONCEPTS vvv
+template <class _Range, enable_if_t<range<_Range>, int> = 0, enable_if_t<is_object_v<_Range>, int> = 0>
+#endif // _CCCL_NO_CONCEPTS
+class ref_view : public view_interface<ref_view<_Range>>
+{
+  _Range* __range_;
+
+public:
+  _CCCL_TEMPLATE(class _Tp)
+  _CCCL_REQUIRES(__different_from<_Tp, ref_view> _CCCL_AND convertible_to<_Tp, _Range&> _CCCL_AND
+                   __convertible_to_lvalue<_Tp, _Range>)
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr ref_view(_Tp&& __t)
+      : view_interface<ref_view<_Range>>()
+      , __range_(_CUDA_VSTD::addressof(static_cast<_Range&>(_CUDA_VSTD::forward<_Tp>(__t))))
+  {}
+
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr _Range& base() const
+  {
+    return *__range_;
+  }
+
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr iterator_t<_Range> begin() const
+  {
+    return _CUDA_VRANGES::begin(*__range_);
+  }
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr sentinel_t<_Range> end() const
+  {
+    return _CUDA_VRANGES::end(*__range_);
+  }
+
+  _CCCL_TEMPLATE(class _Range2 = _Range)
+  _CCCL_REQUIRES(invocable<_CUDA_VRANGES::__empty::__fn, const _Range2&>)
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr bool empty() const
+  {
+    return _CUDA_VRANGES::empty(*__range_);
+  }
+
+  _CCCL_TEMPLATE(class _Range2 = _Range)
+  _CCCL_REQUIRES(sized_range<_Range2>)
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr auto size() const
+  {
+    return _CUDA_VRANGES::size(*__range_);
+  }
+
+  _CCCL_TEMPLATE(class _Range2 = _Range)
+  _CCCL_REQUIRES(contiguous_range<_Range2>)
+  _LIBCUDACXX_HIDE_FROM_ABI constexpr auto data() const
+  {
+    return _CUDA_VRANGES::data(*__range_);
+  }
+};
+
+template <class _Range>
+_CCCL_HOST_DEVICE ref_view(_Range&) -> ref_view<_Range>;
+
+_LIBCUDACXX_END_NAMESPACE_RANGES_ABI
+
+template <class _Tp>
+_CCCL_INLINE_VAR constexpr bool enable_borrowed_range<ref_view<_Tp>> = true;
+
+_LIBCUDACXX_END_NAMESPACE_RANGES
+
+#endif // _LIBCUDACXX___RANGES_REF_VIEW_H

libcudacxx/include/cuda/std/__ranges/subrange.h

@@ -418,7 +418,7 @@ _LIBCUDACXX_END_NAMESPACE_RANGES_ABI
 #if !defined(_CCCL_NO_CONCEPTS)
 template <size_t _Index, class _Iter, class _Sent, subrange_kind _Kind>
   requires((_Index == 0) && copyable<_Iter>) || (_Index == 1)
-#else // ^^^ C++20 ^^^ / vvv C++17 vvv
+#else // ^^^ !_CCCL_NO_CONCEPTS ^^^ / vvv _CCCL_NO_CONCEPTS vvv
 template <size_t _Index,
           class _Iter,
           class _Sent,
@@ -441,7 +441,7 @@ _LIBCUDACXX_HIDE_FROM_ABI constexpr auto get(const subrange<_Iter, _Sent, _Kind>
 #if !defined(_CCCL_NO_CONCEPTS)
 template <size_t _Index, class _Iter, class _Sent, subrange_kind _Kind>
   requires(_Index < 2)
-#else // ^^^ C++20 ^^^ / vvv C++17 vvv
+#else // ^^^ !_CCCL_NO_CONCEPTS ^^^ / vvv _CCCL_NO_CONCEPTS vvv
 template <
   size_t _Index,
   class _Iter,

libcudacxx/include/cuda/std/ranges

@@ -34,6 +34,7 @@ _CCCL_DIAG_SUPPRESS_MSVC(4848)
 #include <cuda/std/__ranges/enable_borrowed_range.h>
 #include <cuda/std/__ranges/enable_view.h>
 #include <cuda/std/__ranges/rbegin.h>
+#include <cuda/std/__ranges/ref_view.h>
 #include <cuda/std/__ranges/rend.h>
 #include <cuda/std/__ranges/size.h>
 #include <cuda/std/__ranges/subrange.h>

libcudacxx/test/libcudacxx/std/ranges/range.adaptors/range.all/range.ref.view/borrowing.compile.pass.cpp

@@ -0,0 +1,27 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of libcu++, the C++ Standard Library for your entire system,
+// 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
+// SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES.
+//
+//===----------------------------------------------------------------------===//
+
+// template<class T>
+//   inline constexpr bool enable_borrowed_range<ref_view<T>> = true;
+
+#include <cuda/std/ranges>
+
+#include "test_range.h"
+
+static_assert(cuda::std::ranges::borrowed_range<cuda::std::ranges::ref_view<BorrowedRange>>);
+static_assert(cuda::std::ranges::borrowed_range<cuda::std::ranges::ref_view<BorrowedView>>);
+#if _LIBCUDACXX_HAS_RANGES
+static_assert(cuda::std::ranges::borrowed_range<cuda::std::ranges::ref_view<NonBorrowedView>>);
+#endif
+
+int main(int, char**)
+{
+  return 0;
+}