/*M///////////////////////////////////////////////////////////////////////////////////////
//
//  IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
//
//  By downloading, copying, installing or using the software you agree to this license.
//  If you do not agree to this license, do not download, install,
//  copy or use the software.
//
//
//                           License Agreement
//                For Open Source Computer Vision Library
//
// Copyright (C) 2000-2008, Intel Corporation, all rights reserved.
// Copyright (C) 2009, Willow Garage Inc., all rights reserved.
// Third party copyrights are property of their respective owners.
//
// Redistribution and use in source and binary forms, with or without modification,
// are permitted provided that the following conditions are met:
//
//   * Redistribution's of source code must retain the above copyright notice,
//     this list of conditions and the following disclaimer.
//
//   * Redistribution's in binary form must reproduce the above copyright notice,
//     this list of conditions and the following disclaimer in the documentation
//     and/or other materials provided with the distribution.
//
//   * The name of the copyright holders may not be used to endorse or promote products
//     derived from this software without specific prior written permission.
//
// This software is provided by the copyright holders and contributors "as is" and
// any express or implied warranties, including, but not limited to, the implied
// warranties of merchantability and fitness for a particular purpose are disclaimed.
// In no event shall the Intel Corporation or contributors be liable for any direct,
// indirect, incidental, special, exemplary, or consequential damages
// (including, but not limited to, procurement of substitute goods or services;
// loss of use, data, or profits; or business interruption) however caused
// and on any theory of liability, whether in contract, strict liability,
// or tort (including negligence or otherwise) arising in any way out of
// the use of this software, even if advised of the possibility of such damage.
//
//M*/

#include "precomp.hpp"

using namespace cv;
using namespace cv::cuda;

#if !defined (HAVE_CUDA) || defined (CUDA_DISABLER)

void cv::cuda::merge(const GpuMat*, size_t, OutputArray, Stream&) { throw_no_cuda(); }
void cv::cuda::merge(const std::vector<GpuMat>&, OutputArray, Stream&) { throw_no_cuda(); }

void cv::cuda::split(InputArray, GpuMat*, Stream&) { throw_no_cuda(); }
void cv::cuda::split(InputArray, std::vector<GpuMat>&, Stream&) { throw_no_cuda(); }

void cv::cuda::transpose(InputArray, OutputArray, Stream&) { throw_no_cuda(); }

void cv::cuda::flip(InputArray, OutputArray, int, Stream&) { throw_no_cuda(); }

Ptr<LookUpTable> cv::cuda::createLookUpTable(InputArray) { throw_no_cuda(); return Ptr<LookUpTable>(); }

void cv::cuda::copyMakeBorder(InputArray, OutputArray, int, int, int, int, int, Scalar, Stream&) { throw_no_cuda(); }

#else /* !defined (HAVE_CUDA) */

////////////////////////////////////////////////////////////////////////
// flip

namespace
{
    template<int DEPTH> struct NppTypeTraits;
    template<> struct NppTypeTraits<CV_8U>  { typedef Npp8u npp_t; };
    template<> struct NppTypeTraits<CV_8S>  { typedef Npp8s npp_t; };
    template<> struct NppTypeTraits<CV_16U> { typedef Npp16u npp_t; };
    template<> struct NppTypeTraits<CV_16S> { typedef Npp16s npp_t; };
    template<> struct NppTypeTraits<CV_32S> { typedef Npp32s npp_t; };
    template<> struct NppTypeTraits<CV_32F> { typedef Npp32f npp_t; };
    template<> struct NppTypeTraits<CV_64F> { typedef Npp64f npp_t; };

    template <int DEPTH> struct NppMirrorFunc
    {
        typedef typename NppTypeTraits<DEPTH>::npp_t npp_t;

        typedef NppStatus (*func_t)(const npp_t* pSrc, int nSrcStep, npp_t* pDst, int nDstStep, NppiSize oROI, NppiAxis flip);
    };

    template <int DEPTH, typename NppMirrorFunc<DEPTH>::func_t func> struct NppMirror
    {
        typedef typename NppMirrorFunc<DEPTH>::npp_t npp_t;

        static void call(const GpuMat& src, GpuMat& dst, int flipCode, cudaStream_t stream)
        {
            NppStreamHandler h(stream);

            NppiSize sz;
            sz.width  = src.cols;
            sz.height = src.rows;

            nppSafeCall( func(src.ptr<npp_t>(), static_cast<int>(src.step),
                dst.ptr<npp_t>(), static_cast<int>(dst.step), sz,
                (flipCode == 0 ? NPP_HORIZONTAL_AXIS : (flipCode > 0 ? NPP_VERTICAL_AXIS : NPP_BOTH_AXIS))) );

            if (stream == 0)
                cudaSafeCall( cudaDeviceSynchronize() );
        }
    };
}

void cv::cuda::flip(InputArray _src, OutputArray _dst, int flipCode, Stream& stream)
{
    typedef void (*func_t)(const GpuMat& src, GpuMat& dst, int flipCode, cudaStream_t stream);
    static const func_t funcs[6][4] =
    {
        {NppMirror<CV_8U, nppiMirror_8u_C1R>::call, 0, NppMirror<CV_8U, nppiMirror_8u_C3R>::call, NppMirror<CV_8U, nppiMirror_8u_C4R>::call},
        {0,0,0,0},
        {NppMirror<CV_16U, nppiMirror_16u_C1R>::call, 0, NppMirror<CV_16U, nppiMirror_16u_C3R>::call, NppMirror<CV_16U, nppiMirror_16u_C4R>::call},
        {0,0,0,0},
        {NppMirror<CV_32S, nppiMirror_32s_C1R>::call, 0, NppMirror<CV_32S, nppiMirror_32s_C3R>::call, NppMirror<CV_32S, nppiMirror_32s_C4R>::call},
        {NppMirror<CV_32F, nppiMirror_32f_C1R>::call, 0, NppMirror<CV_32F, nppiMirror_32f_C3R>::call, NppMirror<CV_32F, nppiMirror_32f_C4R>::call}
    };

    GpuMat src = _src.getGpuMat();

    CV_Assert(src.depth() == CV_8U || src.depth() == CV_16U || src.depth() == CV_32S || src.depth() == CV_32F);
    CV_Assert(src.channels() == 1 || src.channels() == 3 || src.channels() == 4);

    _dst.create(src.size(), src.type());
    GpuMat dst = _dst.getGpuMat();

    funcs[src.depth()][src.channels() - 1](src, dst, flipCode, StreamAccessor::getStream(stream));
}

////////////////////////////////////////////////////////////////////////
// LUT

#if (CUDA_VERSION >= 5000)

namespace
{
    class LookUpTableImpl : public LookUpTable
    {
    public:
        LookUpTableImpl(InputArray lut);

        void transform(InputArray src, OutputArray dst, Stream& stream = Stream::Null());

    private:
        int lut_cn;

        int nValues3[3];
        const Npp32s* pValues3[3];
        const Npp32s* pLevels3[3];

        GpuMat d_pLevels;
        GpuMat d_nppLut;
        GpuMat d_nppLut3[3];
    };

    LookUpTableImpl::LookUpTableImpl(InputArray _lut)
    {
        nValues3[0] = nValues3[1] = nValues3[2] = 256;

        Npp32s pLevels[256];
        for (int i = 0; i < 256; ++i)
            pLevels[i] = i;

        d_pLevels.upload(Mat(1, 256, CV_32S, pLevels));
        pLevels3[0] = pLevels3[1] = pLevels3[2] = d_pLevels.ptr<Npp32s>();

        GpuMat lut;
        if (_lut.kind() == _InputArray::GPU_MAT)
        {
            lut = _lut.getGpuMat();
        }
        else
        {
            Mat hLut = _lut.getMat();
            CV_Assert( hLut.total() == 256 && hLut.isContinuous() );
            lut.upload(Mat(1, 256, hLut.type(), hLut.data));
        }

        lut_cn = lut.channels();

        CV_Assert( lut.depth() == CV_8U );
        CV_Assert( lut.rows == 1 && lut.cols == 256 );

        lut.convertTo(d_nppLut, CV_32S);

        if (lut_cn == 1)
        {
            pValues3[0] = pValues3[1] = pValues3[2] = d_nppLut.ptr<Npp32s>();
        }
        else
        {
            cuda::split(d_nppLut, d_nppLut3);

            pValues3[0] = d_nppLut3[0].ptr<Npp32s>();
            pValues3[1] = d_nppLut3[1].ptr<Npp32s>();
            pValues3[2] = d_nppLut3[2].ptr<Npp32s>();
        }
    }

    void LookUpTableImpl::transform(InputArray _src, OutputArray _dst, Stream& _stream)
    {
        GpuMat src = _src.getGpuMat();

        const int cn = src.channels();

        CV_Assert( src.type() == CV_8UC1 || src.type() == CV_8UC3 );
        CV_Assert( lut_cn == 1 || lut_cn == cn );

        _dst.create(src.size(), src.type());
        GpuMat dst = _dst.getGpuMat();

        cudaStream_t stream = StreamAccessor::getStream(_stream);

        NppStreamHandler h(stream);

        NppiSize sz;
        sz.height = src.rows;
        sz.width = src.cols;

        if (src.type() == CV_8UC1)
        {
            nppSafeCall( nppiLUT_Linear_8u_C1R(src.ptr<Npp8u>(), static_cast<int>(src.step),
                dst.ptr<Npp8u>(), static_cast<int>(dst.step), sz, d_nppLut.ptr<Npp32s>(), d_pLevels.ptr<Npp32s>(), 256) );
        }
        else
        {
            nppSafeCall( nppiLUT_Linear_8u_C3R(src.ptr<Npp8u>(), static_cast<int>(src.step),
                dst.ptr<Npp8u>(), static_cast<int>(dst.step), sz, pValues3, pLevels3, nValues3) );
        }

        if (stream == 0)
            cudaSafeCall( cudaDeviceSynchronize() );
    }
}

#else //  (CUDA_VERSION >= 5000)

namespace
{
    class LookUpTableImpl : public LookUpTable
    {
    public:
        LookUpTableImpl(InputArray lut);

        void transform(InputArray src, OutputArray dst, Stream& stream = Stream::Null());

    private:
        int lut_cn;

        Npp32s pLevels[256];
        int nValues3[3];
        const Npp32s* pValues3[3];
        const Npp32s* pLevels3[3];

        Mat nppLut;
        Mat nppLut3[3];
    };

    LookUpTableImpl::LookUpTableImpl(InputArray _lut)
    {
        nValues3[0] = nValues3[1] = nValues3[2] = 256;

        for (int i = 0; i < 256; ++i)
            pLevels[i] = i;
        pLevels3[0] = pLevels3[1] = pLevels3[2] = pLevels;

        Mat lut;
        if (_lut.kind() == _InputArray::GPU_MAT)
        {
            lut = Mat(_lut.getGpuMat());
        }
        else
        {
            Mat hLut = _lut.getMat();
            CV_Assert( hLut.total() == 256 && hLut.isContinuous() );
            lut = hLut;
        }

        lut_cn = lut.channels();

        CV_Assert( lut.depth() == CV_8U );
        CV_Assert( lut.rows == 1 && lut.cols == 256 );

        lut.convertTo(nppLut, CV_32S);

        if (lut_cn == 1)
        {
            pValues3[0] = pValues3[1] = pValues3[2] = nppLut.ptr<Npp32s>();
        }
        else
        {
            cv::split(nppLut, nppLut3);

            pValues3[0] = nppLut3[0].ptr<Npp32s>();
            pValues3[1] = nppLut3[1].ptr<Npp32s>();
            pValues3[2] = nppLut3[2].ptr<Npp32s>();
        }
    }

    void LookUpTableImpl::transform(InputArray _src, OutputArray _dst, Stream& _stream)
    {
        GpuMat src = _src.getGpuMat();

        const int cn = src.channels();

        CV_Assert( src.type() == CV_8UC1 || src.type() == CV_8UC3 );
        CV_Assert( lut_cn == 1 || lut_cn == cn );

        _dst.create(src.size(), src.type());
        GpuMat dst = _dst.getGpuMat();

        cudaStream_t stream = StreamAccessor::getStream(_stream);

        NppStreamHandler h(stream);

        NppiSize sz;
        sz.height = src.rows;
        sz.width = src.cols;

        if (src.type() == CV_8UC1)
        {
            nppSafeCall( nppiLUT_Linear_8u_C1R(src.ptr<Npp8u>(), static_cast<int>(src.step),
                dst.ptr<Npp8u>(), static_cast<int>(dst.step), sz, nppLut.ptr<Npp32s>(), pLevels, 256) );
        }
        else
        {
            nppSafeCall( nppiLUT_Linear_8u_C3R(src.ptr<Npp8u>(), static_cast<int>(src.step),
                dst.ptr<Npp8u>(), static_cast<int>(dst.step), sz, pValues3, pLevels3, nValues3) );
        }

        if (stream == 0)
            cudaSafeCall( cudaDeviceSynchronize() );
    }
}

#endif //  (CUDA_VERSION >= 5000)

Ptr<LookUpTable> cv::cuda::createLookUpTable(InputArray lut)
{
    return makePtr<LookUpTableImpl>(lut);
}

#endif /* !defined (HAVE_CUDA) */