virtualx-engine/thirdparty/bullet/BulletCollision/CollisionShapes/btBvhTriangleMeshShape.cpp

453 lines
14 KiB
C++
Raw Normal View History

/*
Bullet Continuous Collision Detection and Physics Library
Copyright (c) 2003-2009 Erwin Coumans http://bulletphysics.org
This software is provided 'as-is', without any express or implied warranty.
In no event will the authors be held liable for any damages arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it freely,
subject to the following restrictions:
1. The origin of this software must not be misrepresented; you must not claim that you wrote the original software. If you use this software in a product, an acknowledgment in the product documentation would be appreciated but is not required.
2. Altered source versions must be plainly marked as such, and must not be misrepresented as being the original software.
3. This notice may not be removed or altered from any source distribution.
*/
//#define DISABLE_BVH
#include "BulletCollision/CollisionShapes/btBvhTriangleMeshShape.h"
#include "BulletCollision/CollisionShapes/btOptimizedBvh.h"
#include "LinearMath/btSerializer.h"
///Bvh Concave triangle mesh is a static-triangle mesh shape with Bounding Volume Hierarchy optimization.
///Uses an interface to access the triangles to allow for sharing graphics/physics triangles.
btBvhTriangleMeshShape::btBvhTriangleMeshShape(btStridingMeshInterface* meshInterface, bool useQuantizedAabbCompression, bool buildBvh)
: btTriangleMeshShape(meshInterface),
m_bvh(0),
m_triangleInfoMap(0),
m_useQuantizedAabbCompression(useQuantizedAabbCompression),
m_ownsBvh(false)
{
m_shapeType = TRIANGLE_MESH_SHAPE_PROXYTYPE;
//construct bvh from meshInterface
#ifndef DISABLE_BVH
if (buildBvh)
{
buildOptimizedBvh();
}
#endif //DISABLE_BVH
}
btBvhTriangleMeshShape::btBvhTriangleMeshShape(btStridingMeshInterface* meshInterface, bool useQuantizedAabbCompression, const btVector3& bvhAabbMin, const btVector3& bvhAabbMax, bool buildBvh)
: btTriangleMeshShape(meshInterface),
m_bvh(0),
m_triangleInfoMap(0),
m_useQuantizedAabbCompression(useQuantizedAabbCompression),
m_ownsBvh(false)
{
m_shapeType = TRIANGLE_MESH_SHAPE_PROXYTYPE;
//construct bvh from meshInterface
#ifndef DISABLE_BVH
if (buildBvh)
{
void* mem = btAlignedAlloc(sizeof(btOptimizedBvh), 16);
m_bvh = new (mem) btOptimizedBvh();
m_bvh->build(meshInterface, m_useQuantizedAabbCompression, bvhAabbMin, bvhAabbMax);
m_ownsBvh = true;
}
#endif //DISABLE_BVH
}
void btBvhTriangleMeshShape::partialRefitTree(const btVector3& aabbMin, const btVector3& aabbMax)
{
m_bvh->refitPartial(m_meshInterface, aabbMin, aabbMax);
m_localAabbMin.setMin(aabbMin);
m_localAabbMax.setMax(aabbMax);
}
void btBvhTriangleMeshShape::refitTree(const btVector3& aabbMin, const btVector3& aabbMax)
{
m_bvh->refit(m_meshInterface, aabbMin, aabbMax);
recalcLocalAabb();
}
btBvhTriangleMeshShape::~btBvhTriangleMeshShape()
{
if (m_ownsBvh)
{
m_bvh->~btOptimizedBvh();
btAlignedFree(m_bvh);
}
}
void btBvhTriangleMeshShape::performRaycast(btTriangleCallback* callback, const btVector3& raySource, const btVector3& rayTarget)
{
struct MyNodeOverlapCallback : public btNodeOverlapCallback
{
btStridingMeshInterface* m_meshInterface;
btTriangleCallback* m_callback;
MyNodeOverlapCallback(btTriangleCallback* callback, btStridingMeshInterface* meshInterface)
: m_meshInterface(meshInterface),
m_callback(callback)
{
}
virtual void processNode(int nodeSubPart, int nodeTriangleIndex)
{
btVector3 m_triangle[3];
const unsigned char* vertexbase;
int numverts;
PHY_ScalarType type;
int stride;
const unsigned char* indexbase;
int indexstride;
int numfaces;
PHY_ScalarType indicestype;
m_meshInterface->getLockedReadOnlyVertexIndexBase(
&vertexbase,
numverts,
type,
stride,
&indexbase,
indexstride,
numfaces,
indicestype,
nodeSubPart);
unsigned int* gfxbase = (unsigned int*)(indexbase + nodeTriangleIndex * indexstride);
btAssert(indicestype == PHY_INTEGER || indicestype == PHY_SHORT);
const btVector3& meshScaling = m_meshInterface->getScaling();
for (int j = 2; j >= 0; j--)
{
int graphicsindex = indicestype == PHY_SHORT ? ((unsigned short*)gfxbase)[j] : gfxbase[j];
if (type == PHY_FLOAT)
{
float* graphicsbase = (float*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(graphicsbase[0] * meshScaling.getX(), graphicsbase[1] * meshScaling.getY(), graphicsbase[2] * meshScaling.getZ());
}
else
{
double* graphicsbase = (double*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(btScalar(graphicsbase[0]) * meshScaling.getX(), btScalar(graphicsbase[1]) * meshScaling.getY(), btScalar(graphicsbase[2]) * meshScaling.getZ());
}
}
/* Perform ray vs. triangle collision here */
m_callback->processTriangle(m_triangle, nodeSubPart, nodeTriangleIndex);
m_meshInterface->unLockReadOnlyVertexBase(nodeSubPart);
}
};
MyNodeOverlapCallback myNodeCallback(callback, m_meshInterface);
m_bvh->reportRayOverlappingNodex(&myNodeCallback, raySource, rayTarget);
}
void btBvhTriangleMeshShape::performConvexcast(btTriangleCallback* callback, const btVector3& raySource, const btVector3& rayTarget, const btVector3& aabbMin, const btVector3& aabbMax)
{
struct MyNodeOverlapCallback : public btNodeOverlapCallback
{
btStridingMeshInterface* m_meshInterface;
btTriangleCallback* m_callback;
MyNodeOverlapCallback(btTriangleCallback* callback, btStridingMeshInterface* meshInterface)
: m_meshInterface(meshInterface),
m_callback(callback)
{
}
virtual void processNode(int nodeSubPart, int nodeTriangleIndex)
{
btVector3 m_triangle[3];
const unsigned char* vertexbase;
int numverts;
PHY_ScalarType type;
int stride;
const unsigned char* indexbase;
int indexstride;
int numfaces;
PHY_ScalarType indicestype;
m_meshInterface->getLockedReadOnlyVertexIndexBase(
&vertexbase,
numverts,
type,
stride,
&indexbase,
indexstride,
numfaces,
indicestype,
nodeSubPart);
unsigned int* gfxbase = (unsigned int*)(indexbase + nodeTriangleIndex * indexstride);
btAssert(indicestype == PHY_INTEGER || indicestype == PHY_SHORT);
const btVector3& meshScaling = m_meshInterface->getScaling();
for (int j = 2; j >= 0; j--)
{
int graphicsindex = indicestype == PHY_SHORT ? ((unsigned short*)gfxbase)[j] : gfxbase[j];
if (type == PHY_FLOAT)
{
float* graphicsbase = (float*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(graphicsbase[0] * meshScaling.getX(), graphicsbase[1] * meshScaling.getY(), graphicsbase[2] * meshScaling.getZ());
}
else
{
double* graphicsbase = (double*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(btScalar(graphicsbase[0]) * meshScaling.getX(), btScalar(graphicsbase[1]) * meshScaling.getY(), btScalar(graphicsbase[2]) * meshScaling.getZ());
}
}
/* Perform ray vs. triangle collision here */
m_callback->processTriangle(m_triangle, nodeSubPart, nodeTriangleIndex);
m_meshInterface->unLockReadOnlyVertexBase(nodeSubPart);
}
};
MyNodeOverlapCallback myNodeCallback(callback, m_meshInterface);
m_bvh->reportBoxCastOverlappingNodex(&myNodeCallback, raySource, rayTarget, aabbMin, aabbMax);
}
//perform bvh tree traversal and report overlapping triangles to 'callback'
void btBvhTriangleMeshShape::processAllTriangles(btTriangleCallback* callback, const btVector3& aabbMin, const btVector3& aabbMax) const
{
#ifdef DISABLE_BVH
//brute force traverse all triangles
btTriangleMeshShape::processAllTriangles(callback, aabbMin, aabbMax);
#else
//first get all the nodes
struct MyNodeOverlapCallback : public btNodeOverlapCallback
{
btStridingMeshInterface* m_meshInterface;
btTriangleCallback* m_callback;
btVector3 m_triangle[3];
int m_numOverlap;
MyNodeOverlapCallback(btTriangleCallback* callback, btStridingMeshInterface* meshInterface)
: m_meshInterface(meshInterface),
m_callback(callback),
m_numOverlap(0)
{
}
virtual void processNode(int nodeSubPart, int nodeTriangleIndex)
{
m_numOverlap++;
const unsigned char* vertexbase;
int numverts;
PHY_ScalarType type;
int stride;
const unsigned char* indexbase;
int indexstride;
int numfaces;
PHY_ScalarType indicestype;
m_meshInterface->getLockedReadOnlyVertexIndexBase(
&vertexbase,
numverts,
type,
stride,
&indexbase,
indexstride,
numfaces,
indicestype,
nodeSubPart);
unsigned int* gfxbase = (unsigned int*)(indexbase + nodeTriangleIndex * indexstride);
btAssert(indicestype == PHY_INTEGER || indicestype == PHY_SHORT || indicestype == PHY_UCHAR);
const btVector3& meshScaling = m_meshInterface->getScaling();
for (int j = 2; j >= 0; j--)
{
int graphicsindex = indicestype == PHY_SHORT ? ((unsigned short*)gfxbase)[j] : indicestype == PHY_INTEGER ? gfxbase[j] : ((unsigned char*)gfxbase)[j];
#ifdef DEBUG_TRIANGLE_MESH
printf("%d ,", graphicsindex);
#endif //DEBUG_TRIANGLE_MESH
if (type == PHY_FLOAT)
{
float* graphicsbase = (float*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(
graphicsbase[0] * meshScaling.getX(),
graphicsbase[1] * meshScaling.getY(),
graphicsbase[2] * meshScaling.getZ());
}
else
{
double* graphicsbase = (double*)(vertexbase + graphicsindex * stride);
m_triangle[j] = btVector3(
btScalar(graphicsbase[0]) * meshScaling.getX(),
btScalar(graphicsbase[1]) * meshScaling.getY(),
btScalar(graphicsbase[2]) * meshScaling.getZ());
}
#ifdef DEBUG_TRIANGLE_MESH
printf("triangle vertices:%f,%f,%f\n", triangle[j].x(), triangle[j].y(), triangle[j].z());
#endif //DEBUG_TRIANGLE_MESH
}
m_callback->processTriangle(m_triangle, nodeSubPart, nodeTriangleIndex);
m_meshInterface->unLockReadOnlyVertexBase(nodeSubPart);
}
};
MyNodeOverlapCallback myNodeCallback(callback, m_meshInterface);
m_bvh->reportAabbOverlappingNodex(&myNodeCallback, aabbMin, aabbMax);
#endif //DISABLE_BVH
}
void btBvhTriangleMeshShape::setLocalScaling(const btVector3& scaling)
{
if ((getLocalScaling() - scaling).length2() > SIMD_EPSILON)
{
btTriangleMeshShape::setLocalScaling(scaling);
buildOptimizedBvh();
}
}
void btBvhTriangleMeshShape::buildOptimizedBvh()
{
if (m_ownsBvh)
{
m_bvh->~btOptimizedBvh();
btAlignedFree(m_bvh);
}
///m_localAabbMin/m_localAabbMax is already re-calculated in btTriangleMeshShape. We could just scale aabb, but this needs some more work
void* mem = btAlignedAlloc(sizeof(btOptimizedBvh), 16);
m_bvh = new (mem) btOptimizedBvh();
//rebuild the bvh...
m_bvh->build(m_meshInterface, m_useQuantizedAabbCompression, m_localAabbMin, m_localAabbMax);
m_ownsBvh = true;
}
void btBvhTriangleMeshShape::setOptimizedBvh(btOptimizedBvh* bvh, const btVector3& scaling)
{
btAssert(!m_bvh);
btAssert(!m_ownsBvh);
m_bvh = bvh;
m_ownsBvh = false;
// update the scaling without rebuilding the bvh
if ((getLocalScaling() - scaling).length2() > SIMD_EPSILON)
{
btTriangleMeshShape::setLocalScaling(scaling);
}
}
///fills the dataBuffer and returns the struct name (and 0 on failure)
const char* btBvhTriangleMeshShape::serialize(void* dataBuffer, btSerializer* serializer) const
{
btTriangleMeshShapeData* trimeshData = (btTriangleMeshShapeData*)dataBuffer;
btCollisionShape::serialize(&trimeshData->m_collisionShapeData, serializer);
m_meshInterface->serialize(&trimeshData->m_meshInterface, serializer);
trimeshData->m_collisionMargin = float(m_collisionMargin);
if (m_bvh && !(serializer->getSerializationFlags() & BT_SERIALIZE_NO_BVH))
{
void* chunk = serializer->findPointer(m_bvh);
if (chunk)
{
#ifdef BT_USE_DOUBLE_PRECISION
trimeshData->m_quantizedDoubleBvh = (btQuantizedBvhData*)chunk;
trimeshData->m_quantizedFloatBvh = 0;
#else
trimeshData->m_quantizedFloatBvh = (btQuantizedBvhData*)chunk;
trimeshData->m_quantizedDoubleBvh = 0;
#endif //BT_USE_DOUBLE_PRECISION
}
else
{
#ifdef BT_USE_DOUBLE_PRECISION
trimeshData->m_quantizedDoubleBvh = (btQuantizedBvhData*)serializer->getUniquePointer(m_bvh);
trimeshData->m_quantizedFloatBvh = 0;
#else
trimeshData->m_quantizedFloatBvh = (btQuantizedBvhData*)serializer->getUniquePointer(m_bvh);
trimeshData->m_quantizedDoubleBvh = 0;
#endif //BT_USE_DOUBLE_PRECISION
int sz = m_bvh->calculateSerializeBufferSizeNew();
btChunk* chunk = serializer->allocate(sz, 1);
const char* structType = m_bvh->serialize(chunk->m_oldPtr, serializer);
serializer->finalizeChunk(chunk, structType, BT_QUANTIZED_BVH_CODE, m_bvh);
}
}
else
{
trimeshData->m_quantizedFloatBvh = 0;
trimeshData->m_quantizedDoubleBvh = 0;
}
if (m_triangleInfoMap && !(serializer->getSerializationFlags() & BT_SERIALIZE_NO_TRIANGLEINFOMAP))
{
void* chunk = serializer->findPointer(m_triangleInfoMap);
if (chunk)
{
trimeshData->m_triangleInfoMap = (btTriangleInfoMapData*)chunk;
}
else
{
trimeshData->m_triangleInfoMap = (btTriangleInfoMapData*)serializer->getUniquePointer(m_triangleInfoMap);
int sz = m_triangleInfoMap->calculateSerializeBufferSize();
btChunk* chunk = serializer->allocate(sz, 1);
const char* structType = m_triangleInfoMap->serialize(chunk->m_oldPtr, serializer);
serializer->finalizeChunk(chunk, structType, BT_TRIANLGE_INFO_MAP, m_triangleInfoMap);
}
}
else
{
trimeshData->m_triangleInfoMap = 0;
}
// Fill padding with zeros to appease msan.
memset(trimeshData->m_pad3, 0, sizeof(trimeshData->m_pad3));
return "btTriangleMeshShapeData";
}
void btBvhTriangleMeshShape::serializeSingleBvh(btSerializer* serializer) const
{
if (m_bvh)
{
int len = m_bvh->calculateSerializeBufferSizeNew(); //make sure not to use calculateSerializeBufferSize because it is used for in-place
btChunk* chunk = serializer->allocate(len, 1);
const char* structType = m_bvh->serialize(chunk->m_oldPtr, serializer);
serializer->finalizeChunk(chunk, structType, BT_QUANTIZED_BVH_CODE, (void*)m_bvh);
}
}
void btBvhTriangleMeshShape::serializeSingleTriangleInfoMap(btSerializer* serializer) const
{
if (m_triangleInfoMap)
{
int len = m_triangleInfoMap->calculateSerializeBufferSize();
btChunk* chunk = serializer->allocate(len, 1);
const char* structType = m_triangleInfoMap->serialize(chunk->m_oldPtr, serializer);
serializer->finalizeChunk(chunk, structType, BT_TRIANLGE_INFO_MAP, (void*)m_triangleInfoMap);
}
}