// ======================================================================== //
// Copyright 2009-2017 Intel Corporation                                    //
//                                                                          //
// Licensed under the Apache License, Version 2.0 (the "License");          //
// you may not use this file except in compliance with the License.         //
// You may obtain a copy of the License at                                  //
//                                                                          //
//     http://www.apache.org/licenses/LICENSE-2.0                           //
//                                                                          //
// Unless required by applicable law or agreed to in writing, software      //
// distributed under the License is distributed on an "AS IS" BASIS,        //
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. //
// See the License for the specific language governing permissions and      //
// limitations under the License.                                           //
// ======================================================================== //

#include "scene_bezier_curves.h"
#include "scene.h"

namespace embree
{
  NativeCurves::NativeCurves (Scene* parent, SubType subtype, Basis basis, RTCGeometryFlags flags, size_t numPrimitives, size_t numVertices, size_t numTimeSteps) 
    : Geometry(parent,BEZIER_CURVES,numPrimitives,numTimeSteps,flags), subtype(subtype), basis(basis), tessellationRate(4)
  {
    curves.init(parent->device,numPrimitives,sizeof(int));
    vertices.resize(numTimeSteps);
    for (size_t i=0; i<numTimeSteps; i++) {
      vertices[i].init(parent->device,numVertices,sizeof(Vec3fa));
    }
    enabling();
  }

  void NativeCurves::enabling() 
  { 
    if (numTimeSteps == 1) parent->world.numBezierCurves += numPrimitives; 
    else                   parent->worldMB.numBezierCurves += numPrimitives; 
  }
  
  void NativeCurves::disabling() 
  { 
    if (numTimeSteps == 1) parent->world.numBezierCurves -= numPrimitives; 
    else                   parent->worldMB.numBezierCurves -= numPrimitives;
  }
  
  void NativeCurves::setMask (unsigned mask) 
  {
    if (parent->isStatic() && parent->isBuild())
      throw_RTCError(RTC_INVALID_OPERATION,"static geometries cannot get modified");

    this->mask = mask; 
    Geometry::update();
  }

  void NativeCurves::setBuffer(RTCBufferType type, void* ptr, size_t offset, size_t stride, size_t size) 
  { 
    if (parent->isStatic() && parent->isBuild())
      throw_RTCError(RTC_INVALID_OPERATION,"static geometries cannot get modified");

    /* verify that all accesses are 4 bytes aligned */
    if (((size_t(ptr) + offset) & 0x3) || (stride & 0x3)) 
      throw_RTCError(RTC_INVALID_OPERATION,"data must be 4 bytes aligned");

    unsigned bid = type & 0xFFFF;
    if (type >= RTC_VERTEX_BUFFER0 && type < RTCBufferType(RTC_VERTEX_BUFFER0 + numTimeSteps)) 
    {
      size_t t = type - RTC_VERTEX_BUFFER0;
      vertices[t].set(ptr,offset,stride,size); 
      vertices[t].checkPadding16();
    } 
    else if (type >= RTC_USER_VERTEX_BUFFER0 && type < RTC_USER_VERTEX_BUFFER0+RTC_MAX_USER_VERTEX_BUFFERS)
    {
      if (bid >= userbuffers.size()) userbuffers.resize(bid+1);
      userbuffers[bid] = APIBuffer<char>(parent->device,numVertices(),stride);
      userbuffers[bid].set(ptr,offset,stride,size);  
      userbuffers[bid].checkPadding16();
    }
    else if (type == RTC_INDEX_BUFFER) 
    {
      if (size != (size_t)-1) disabling();
      curves.set(ptr,offset,stride,size); 
      setNumPrimitives(size);
      if (size != (size_t)-1) enabling();
    }
    else 
        throw_RTCError(RTC_INVALID_ARGUMENT,"unknown buffer type"); 
  }

  void* NativeCurves::map(RTCBufferType type) 
  {
    if (parent->isStatic() && parent->isBuild()) {
      throw_RTCError(RTC_INVALID_OPERATION,"static geometries cannot get modified");
      return nullptr;
    }

    if (type == RTC_INDEX_BUFFER) {
      return curves.map(parent->numMappedBuffers);
    }
    else if (type >= RTC_VERTEX_BUFFER0 && type < RTCBufferType(RTC_VERTEX_BUFFER0 + numTimeSteps)) {
      return vertices[type - RTC_VERTEX_BUFFER0].map(parent->numMappedBuffers);
    }
    else {
      throw_RTCError(RTC_INVALID_ARGUMENT,"unknown buffer type"); 
      return nullptr;
    }
  }

  void NativeCurves::unmap(RTCBufferType type) 
  {
    if (parent->isStatic() && parent->isBuild()) 
      throw_RTCError(RTC_INVALID_OPERATION,"static geometries cannot get modified");

    if (type == RTC_INDEX_BUFFER) {
      curves.unmap(parent->numMappedBuffers);
    }
    else if (type >= RTC_VERTEX_BUFFER0 && type < RTCBufferType(RTC_VERTEX_BUFFER0 + numTimeSteps)) {
      vertices[type - RTC_VERTEX_BUFFER0].unmap(parent->numMappedBuffers);
    }
    else {
      throw_RTCError(RTC_INVALID_ARGUMENT,"unknown buffer type"); 
    }
  }
  
  void NativeCurves::setTessellationRate(float N)
  {
    if (parent->isStatic() && parent->isBuild()) 
      throw_RTCError(RTC_INVALID_OPERATION,"static geometries cannot get modified");

    tessellationRate = clamp((int)N,1,16);
  }

  void NativeCurves::immutable () 
  {
    const bool freeIndices = !parent->needBezierIndices;
    const bool freeVertices  = !parent->needBezierVertices;
    if (freeIndices) curves.free();
    if (freeVertices )
      for (auto& buffer : vertices)
        buffer.free();
  }

  bool NativeCurves::verify () 
  {
    /*! verify consistent size of vertex arrays */
    if (vertices.size() == 0) return false;
    for (const auto& buffer : vertices)
      if (vertices[0].size() != buffer.size())
        return false;

    /*! verify indices */
    for (size_t i=0; i<numPrimitives; i++) {
      if (curves[i]+3 >= numVertices()) return false;
    }
    
    /*! verify vertices */
    for (const auto& buffer : vertices) {
      for (size_t i=0; i<buffer.size(); i++) {
	if (!isvalid(buffer[i].x)) return false;
        if (!isvalid(buffer[i].y)) return false;
        if (!isvalid(buffer[i].z)) return false;
        if (!isvalid(buffer[i].w)) return false;
      }
    }
    return true;
  }

  void NativeCurves::preCommit()
  {
    if (!isEnabled()) return;

    native_curves = (BufferRefT<unsigned>) curves;
    if (native_vertices.size() != vertices.size())
      native_vertices.resize(vertices.size());

    native_vertices0 = vertices[0];
    for (size_t i=0; i<vertices.size(); i++)
      native_vertices[i] = (BufferRefT<Vec3fa>) vertices[i];
  }

  template<typename Curve>
    void NativeCurves::interpolate_helper(unsigned primID, float u, float v, RTCBufferType buffer, 
                                          float* P, float* dPdu, float* dPdv, float* ddPdudu, float* ddPdvdv, float* ddPdudv, size_t numFloats) 
  {
  /* test if interpolation is enabled */
#if defined(DEBUG) 
    if ((parent->aflags & RTC_INTERPOLATE) == 0) 
      throw_RTCError(RTC_INVALID_OPERATION,"rtcInterpolate can only get called when RTC_INTERPOLATE is enabled for the scene");
#endif

    /* calculate base pointer and stride */
    assert((buffer >= RTC_VERTEX_BUFFER0 && buffer < RTCBufferType(RTC_VERTEX_BUFFER0 + numTimeSteps)) ||
           (buffer >= RTC_USER_VERTEX_BUFFER0 && buffer <= RTC_USER_VERTEX_BUFFER1));
    const char* src = nullptr; 
    size_t stride = 0;
    if (buffer >= RTC_USER_VERTEX_BUFFER0) {
      src    = userbuffers[buffer&0xFFFF].getPtr();
      stride = userbuffers[buffer&0xFFFF].getStride();
    } else {
      src    = vertices[buffer&0xFFFF].getPtr();
      stride = vertices[buffer&0xFFFF].getStride();
    }

    for (size_t i=0; i<numFloats; i+=VSIZEX)
    {
      size_t ofs = i*sizeof(float);
      const size_t curve = curves[primID];
      const vboolx valid = vintx((int)i)+vintx(step) < vintx(numFloats);
      const vfloatx p0 = vfloatx::loadu(valid,(float*)&src[(curve+0)*stride+ofs]);
      const vfloatx p1 = vfloatx::loadu(valid,(float*)&src[(curve+1)*stride+ofs]);
      const vfloatx p2 = vfloatx::loadu(valid,(float*)&src[(curve+2)*stride+ofs]);
      const vfloatx p3 = vfloatx::loadu(valid,(float*)&src[(curve+3)*stride+ofs]);
      
      const Curve bezier(p0,p1,p2,p3);
      if (P      ) vfloatx::storeu(valid,P+i,      bezier.eval(u));
      if (dPdu   ) vfloatx::storeu(valid,dPdu+i,   bezier.eval_du(u));
      if (ddPdudu) vfloatx::storeu(valid,ddPdudu+i,bezier.eval_dudu(u));
    }
  }

  void NativeCurves::interpolate(unsigned primID, float u, float v, RTCBufferType buffer, 
                                 float* P, float* dPdu, float* dPdv, float* ddPdudu, float* ddPdvdv, float* ddPdudv, size_t numFloats) 
  {
    interpolate_helper<CurveT<vfloatx>>(primID,u,v,buffer,P,dPdu,dPdv,ddPdudu,ddPdvdv,ddPdudv,numFloats);
  }

  template<typename InputCurve3fa>
    void NativeCurves::commit_helper()
  {
    if (native_curves.size() != curves.size()) 
    {
      native_curves = APIBuffer<unsigned>(parent->device,curves.size(),sizeof(unsigned int),true);
      parallel_for(size_t(0), curves.size(), size_t(1024), [&] ( const range<size_t> r) {
          for (size_t i=r.begin(); i<r.end(); i++) {
            if (curves[i]+3 >= numVertices()) native_curves[i] = 0xFFFFFFF0; // invalid curves stay invalid this way
            else                              native_curves[i] = unsigned(4*i);
          }
        });
    }

    if (native_vertices.size() != vertices.size())
      native_vertices.resize(vertices.size());

    parallel_for(vertices.size(), [&] ( const size_t i ) {
        
        if (native_vertices[i].size() != 4*curves.size())
          native_vertices[i] = APIBuffer<Vec3fa>(parent->device,4*curves.size(),sizeof(Vec3fa),true);
        
        parallel_for(size_t(0), curves.size(), size_t(1024), [&] ( const range<size_t> rj ) {

            for (size_t j=rj.begin(); j<rj.end(); j++)
            {
              const unsigned id = curves[j];
              if (id+3 >= numVertices()) continue; // ignore invalid curves
              const Vec3fa v0 = vertices[i][id+0];
              const Vec3fa v1 = vertices[i][id+1];
              const Vec3fa v2 = vertices[i][id+2];
              const Vec3fa v3 = vertices[i][id+3];
              const InputCurve3fa icurve(v0,v1,v2,v3);
              Curve3fa ocurve; convert<Vec3fa>(icurve,ocurve);
              native_vertices[i].store(4*j+0,ocurve.v0);
              native_vertices[i].store(4*j+1,ocurve.v1);
              native_vertices[i].store(4*j+2,ocurve.v2);
              native_vertices[i].store(4*j+3,ocurve.v3);
            }
        });
      });
    native_vertices0 = native_vertices[0];
  }

  CurvesBezier::CurvesBezier (Scene* parent, SubType subtype, Basis basis, RTCGeometryFlags flags, size_t numPrimitives, size_t numVertices, size_t numTimeSteps)
    : NativeCurves(parent,subtype,basis,flags,numPrimitives,numVertices,numTimeSteps) {}
  
  void CurvesBezier::preCommit() {
    if (isEnabled()) commit_helper<BezierCurve3fa>();
  }

  void CurvesBezier::interpolate(unsigned primID, float u, float v, RTCBufferType buffer, 
                                 float* P, float* dPdu, float* dPdv, float* ddPdudu, float* ddPdvdv, float* ddPdudv, size_t numFloats) 
  {
    interpolate_helper<BezierCurveT<vfloatx>>(primID,u,v,buffer,P,dPdu,dPdv,ddPdudu,ddPdvdv,ddPdudv,numFloats);
  }

  CurvesBSpline::CurvesBSpline (Scene* parent, SubType subtype, Basis basis, RTCGeometryFlags flags, size_t numPrimitives, size_t numVertices, size_t numTimeSteps)
    : NativeCurves(parent,subtype,basis,flags,numPrimitives,numVertices,numTimeSteps) {}
  
  void CurvesBSpline::preCommit() {
    if (isEnabled()) commit_helper<BSplineCurve3fa>();
  }

  void CurvesBSpline::interpolate(unsigned primID, float u, float v, RTCBufferType buffer, 
                                 float* P, float* dPdu, float* dPdv, float* ddPdudu, float* ddPdvdv, float* ddPdudv, size_t numFloats) 
  {
    interpolate_helper<BSplineCurveT<vfloatx>>(primID,u,v,buffer,P,dPdu,dPdv,ddPdudu,ddPdvdv,ddPdudv,numFloats);
  }
}