//===----------------------------------------------------------------------===//
//
// 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) 2023 NVIDIA CORPORATION & AFFILIATES.
//
//===----------------------------------------------------------------------===//

// <cuda/std/complex>

// test cases

#ifndef CASES_H
#define CASES_H

#include <cuda/std/cassert>
#include <cuda/std/complex>
#include <cuda/std/limits>

template <class T>
using testcases_t = cuda::std::complex<T>[152];

template <class T>
struct _testcases
{
  testcases_t<T> _cases;

  static constexpr size_t count = sizeof(testcases_t<T>) / sizeof(cuda::std::complex<T>);

  __host__ __device__ const cuda::std::complex<T>* begin() const
  {
    return &_cases[0];
  }
  __host__ __device__ const cuda::std::complex<T>* cbegin() const
  {
    return &_cases[0];
  }
  __host__ __device__ cuda::std::complex<T>* begin()
  {
    return &_cases[0];
  }

  __host__ __device__ const cuda::std::complex<T>* end() const
  {
    return &_cases[count];
  }
  __host__ __device__ const cuda::std::complex<T>* cend() const
  {
    return &_cases[count];
  }
  __host__ __device__ cuda::std::complex<T>* end()
  {
    return &_cases[count];
  }

  __host__ __device__ cuda::std::complex<T>& operator[](size_t n)
  {
    return _cases[n];
  }

  __host__ __device__ const cuda::std::complex<T>& operator[](size_t n) const
  {
    return _cases[n];
  }
};

template <class T>
__host__ __device__ _testcases<T> get_testcases()
{
  _testcases<T> tc{
    cuda::std::complex<T>(1.e-2, 1.e-2),
    cuda::std::complex<T>(-1.e-2, 1.e-2),
    cuda::std::complex<T>(-1.e-2, -1.e-2),
    cuda::std::complex<T>(1.e-2, -1.e-2),

    cuda::std::complex<T>(1.e+2, 1.e-2),
    cuda::std::complex<T>(-1.e+2, 1.e-2),
    cuda::std::complex<T>(-1.e+2, -1.e-2),
    cuda::std::complex<T>(1.e+2, -1.e-2),

    cuda::std::complex<T>(1.e-2, 1.e+2),
    cuda::std::complex<T>(-1.e-2, 1.e+2),
    cuda::std::complex<T>(-1.e-2, -1.e+2),
    cuda::std::complex<T>(1.e-2, -1.e+2),

    cuda::std::complex<T>(1.e+2, 1.e+2),
    cuda::std::complex<T>(-1.e+2, 1.e+2),
    cuda::std::complex<T>(-1.e+2, -1.e+2),
    cuda::std::complex<T>(1.e+2, -1.e+2),

    cuda::std::complex<T>(-0, -1.e-2),
    cuda::std::complex<T>(-0, 1.e-2),
    cuda::std::complex<T>(-0, 1.e+2),
    cuda::std::complex<T>(-0, -1.e+2),
    cuda::std::complex<T>(0, -1.e-2),
    cuda::std::complex<T>(0, 1.e-2),
    cuda::std::complex<T>(0, 1.e+2),
    cuda::std::complex<T>(0, -1.e+2),

    cuda::std::complex<T>(-1.e-2, -0),
    cuda::std::complex<T>(1.e-2, -0),
    cuda::std::complex<T>(1.e+2, -0),
    cuda::std::complex<T>(-1.e+2, -0),
    cuda::std::complex<T>(-1.e-2, 0),
    cuda::std::complex<T>(1.e-2, 0),
    cuda::std::complex<T>(1.e+2, 0),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(-2, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(-1, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(-0.5, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(-0., cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(+0., cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(0.5, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(1, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(2, cuda::std::numeric_limits<T>::quiet_NaN()),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-2, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-1, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-0.5, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-0., -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(+0., -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(0.5, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(1, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(2, -cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), -cuda::std::numeric_limits<T>::infinity()),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), -2),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), -2),
    cuda::std::complex<T>(-2, -2),
    cuda::std::complex<T>(-1, -2),
    cuda::std::complex<T>(-0.5, -2),
    cuda::std::complex<T>(-0., -2),
    cuda::std::complex<T>(+0., -2),
    cuda::std::complex<T>(0.5, -2),
    cuda::std::complex<T>(1, -2),
    cuda::std::complex<T>(2, -2),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), -2),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), -1),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), -1),
    cuda::std::complex<T>(-2, -1),
    cuda::std::complex<T>(-1, -1),
    cuda::std::complex<T>(-0.5, -1),
    cuda::std::complex<T>(-0., -1),
    cuda::std::complex<T>(+0., -1),
    cuda::std::complex<T>(0.5, -1),
    cuda::std::complex<T>(1, -1),
    cuda::std::complex<T>(2, -1),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), -1),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), -0.5),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), -0.5),
    cuda::std::complex<T>(-2, -0.5),
    cuda::std::complex<T>(-1, -0.5),
    cuda::std::complex<T>(-0.5, -0.5),
    cuda::std::complex<T>(-0., -0.5),
    cuda::std::complex<T>(+0., -0.5),
    cuda::std::complex<T>(0.5, -0.5),
    cuda::std::complex<T>(1, -0.5),
    cuda::std::complex<T>(2, -0.5),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), -0.5),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), -0.),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), -0.),
    cuda::std::complex<T>(-2, -0.),
    cuda::std::complex<T>(-1, -0.),
    cuda::std::complex<T>(-0.5, -0.),
    cuda::std::complex<T>(-0., -0.),
    cuda::std::complex<T>(+0., -0.),
    cuda::std::complex<T>(0.5, -0.),
    cuda::std::complex<T>(1, -0.),
    cuda::std::complex<T>(2, -0.),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), -0.),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), +0.),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), +0.),
    cuda::std::complex<T>(-2, +0.),
    cuda::std::complex<T>(-1, +0.),
    cuda::std::complex<T>(-0.5, +0.),
    cuda::std::complex<T>(-0., +0.),
    cuda::std::complex<T>(+0., +0.),
    cuda::std::complex<T>(0.5, +0.),
    cuda::std::complex<T>(1, +0.),
    cuda::std::complex<T>(2, +0.),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), +0.),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), 0.5),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), 0.5),
    cuda::std::complex<T>(-2, 0.5),
    cuda::std::complex<T>(-1, 0.5),
    cuda::std::complex<T>(-0.5, 0.5),
    cuda::std::complex<T>(-0., 0.5),
    cuda::std::complex<T>(+0., 0.5),
    cuda::std::complex<T>(0.5, 0.5),
    cuda::std::complex<T>(1, 0.5),
    cuda::std::complex<T>(2, 0.5),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), 0.5),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), 1),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), 1),
    cuda::std::complex<T>(-2, 1),
    cuda::std::complex<T>(-1, 1),
    cuda::std::complex<T>(-0.5, 1),
    cuda::std::complex<T>(-0., 1),
    cuda::std::complex<T>(+0., 1),
    cuda::std::complex<T>(0.5, 1),
    cuda::std::complex<T>(1, 1),
    cuda::std::complex<T>(2, 1),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), 1),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), 2),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), 2),
    cuda::std::complex<T>(-2, 2),
    cuda::std::complex<T>(-1, 2),
    cuda::std::complex<T>(-0.5, 2),
    cuda::std::complex<T>(-0., 2),
    cuda::std::complex<T>(+0., 2),
    cuda::std::complex<T>(0.5, 2),
    cuda::std::complex<T>(1, 2),
    cuda::std::complex<T>(2, 2),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), 2),

    cuda::std::complex<T>(cuda::std::numeric_limits<T>::quiet_NaN(), cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-2, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-1, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-0.5, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(-0., cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(+0., cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(0.5, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(1, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(2, cuda::std::numeric_limits<T>::infinity()),
    cuda::std::complex<T>(cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::infinity())};

  return tc;
}

enum
{
  zero,
  non_zero,
  inf,
  NaN,
  non_zero_nan
};

template <class T>
__host__ __device__ int classify(const cuda::std::complex<T>& x)
{
  if (x == cuda::std::complex<T>())
  {
    return zero;
  }
  if (cuda::std::isinf(x.real()) || cuda::std::isinf(x.imag()))
  {
    return inf;
  }
  if (cuda::std::isnan(x.real()) && cuda::std::isnan(x.imag()))
  {
    return NaN;
  }
  if (cuda::std::isnan(x.real()))
  {
    if (x.imag() == T(0))
    {
      return NaN;
    }
    return non_zero_nan;
  }
  if (cuda::std::isnan(x.imag()))
  {
    if (x.real() == T(0))
    {
      return NaN;
    }
    return non_zero_nan;
  }
  return non_zero;
}

template <class T>
inline __host__ __device__ int classify(T x)
{
  if (x == T(0))
  {
    return zero;
  }
  if (cuda::std::isinf(x))
  {
    return inf;
  }
  if (cuda::std::isnan(x))
  {
    return NaN;
  }
  return non_zero;
}

__host__ __device__ void is_about(float x, float y)
{
  assert(cuda::std::abs((x - y) / (x + y)) < 1.e-6);
}

__host__ __device__ void is_about(double x, double y)
{
  assert(cuda::std::abs((x - y) / (x + y)) < 1.e-14);
}

#if _CCCL_HAS_LONG_DOUBLE()
__host__ __device__ void is_about(long double x, long double y)
{
  assert(cuda::std::abs((x - y) / (x + y)) < 1.e-14);
}
#endif // _CCCL_HAS_LONG_DOUBLE()

#if _LIBCUDACXX_HAS_NVFP16()
__host__ __device__ void is_about(__half x, __half y)
{
  assert(cuda::std::fabs((x - y) / (x + y)) <= __half(1e-3));
}
#endif // _LIBCUDACXX_HAS_NVFP16()

#if _LIBCUDACXX_HAS_NVBF16()
__host__ __device__ void is_about(__nv_bfloat16 x, __nv_bfloat16 y)
{
  assert(cuda::std::fabs((x - y) / (x + y)) <= __nv_bfloat16(5e-3));
}
#endif // _LIBCUDACXX_HAS_NVBF16()

#endif // CASES_H