2009-04-25 15:47:20 +00:00

371 lines
12 KiB
C++

#pragma region License
/******************************************************************************
This file is part of the PolyVox library
Copyright (C) 2006 David Williams
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
******************************************************************************/
#pragma endregion
#pragma region Headers
#include "Block.h"
#include "Log.h"
#include "VolumeIterator.h"
#include "Region.h"
#include "Vector.h"
#include <cassert>
#include <cstring> //For memcpy
#include <list>
#include <stdexcept> //For invalid_argument
#pragma endregion
namespace PolyVox
{
#pragma region Constructors/Destructors
template <typename VoxelType>
Volume<VoxelType>::Volume(uint16_t uWidth, uint16_t uHeight, uint16_t uDepth, uint16_t uBlockSideLength)
:m_pBlocks(0)
,m_uCurrentBlockForTidying(0)
{
//A values of zero for a block side length is a special value to indicate that the block
//side length should simply be made as large as possible. This can be useful if you are
//creating only a small volume which doesn't need to be broken down into many blocks. This
//'largest possible block size' will be equal to the shortest volume dimension, as a volume
//dimension can never be less than a block side length.
if(uBlockSideLength == 0)
{
uBlockSideLength = (std::min)((std::min)(uWidth,uHeight),uDepth);
}
//Debug mode validation
assert(isPowerOf2(uWidth));
assert(isPowerOf2(uHeight));
assert(isPowerOf2(uDepth));
assert(isPowerOf2(uBlockSideLength));
assert(uBlockSideLength <= uWidth);
assert(uBlockSideLength <= uHeight);
assert(uBlockSideLength <= uDepth);
//Release mode validation
if(!(isPowerOf2(uWidth) && isPowerOf2(uHeight) && isPowerOf2(uDepth)))
{
throw std::invalid_argument("Volume width, height, and depth must all be a power of two.");
}
if(!isPowerOf2(uBlockSideLength))
{
throw std::invalid_argument("Block side length must be a power of two.");
}
if(uBlockSideLength > uWidth)
{
throw std::invalid_argument("Block side length cannot be greater than volume width.");
}
if(uBlockSideLength > uHeight)
{
throw std::invalid_argument("Block side length cannot be greater than volume height.");
}
if(uBlockSideLength > uDepth)
{
throw std::invalid_argument("Block side length cannot be greater than volume depth.");
}
//Compute the volume side lengths
m_uWidth = uWidth;
m_uWidthPower = logBase2(m_uWidth);
m_uHeight = uHeight;
m_uHeightPower = logBase2(m_uHeight);
m_uDepth = uDepth;
m_uDepthPower = logBase2(m_uDepth);
//Compute the block side length
m_uBlockSideLength = uBlockSideLength;
m_uBlockSideLengthPower = logBase2(m_uBlockSideLength);
//Compute the side length in blocks
//m_uSideLengthInBlocks = m_uSideLength / m_uBlockSideLength;
m_uWidthInBlocks = m_uWidth / m_uBlockSideLength;
m_uHeightInBlocks = m_uHeight / m_uBlockSideLength;
m_uDepthInBlocks = m_uDepth / m_uBlockSideLength;
//Compute number of blocks in the volume
m_uNoOfBlocksInVolume = m_uWidthInBlocks * m_uHeightInBlocks * m_uDepthInBlocks;
//Create the blocks
m_pBlocks.resize(m_uNoOfBlocksInVolume);
m_vecBlockIsPotentiallyHomogenous.resize(m_uNoOfBlocksInVolume);
for(uint32_t i = 0; i < m_uNoOfBlocksInVolume; ++i)
{
m_pBlocks[i] = getHomogenousBlockData(0);
m_vecBlockIsPotentiallyHomogenous[i] = false;
}
//Other properties we might find useful later
m_uLongestSideLength = (std::max)((std::max)(m_uWidth,m_uHeight),m_uDepth);
m_uShortestSideLength = (std::min)((std::min)(m_uWidth,m_uHeight),m_uDepth);
m_fDiagonalLength = sqrtf(static_cast<float>(m_uWidth * m_uWidth + m_uHeight * m_uHeight + m_uDepth * m_uDepth));
}
/*template <typename VoxelType>
Volume<VoxelType>::Volume(const Volume<VoxelType>& rhs)
{
*this = rhs;
}*/
template <typename VoxelType>
Volume<VoxelType>::~Volume()
{
}
#pragma endregion
#pragma region Operators
/*template <typename VoxelType>
Volume<VoxelType>& Volume<VoxelType>::operator=(const Volume& rhs)
{
}*/
#pragma endregion
#pragma region Getters
template <typename VoxelType>
Region Volume<VoxelType>::getEnclosingRegion(void) const
{
return Region(Vector3DInt32(0,0,0), Vector3DInt32(m_uWidth-1,m_uHeight-1,m_uDepth-1));
}
template <typename VoxelType>
uint16_t Volume<VoxelType>::getDepth(void) const
{
return m_uDepth;
}
template <typename VoxelType>
float Volume<VoxelType>::getDiagonalLength(void) const
{
return m_fDiagonalLength;
}
template <typename VoxelType>
uint16_t Volume<VoxelType>::getHeight(void) const
{
return m_uHeight;
}
template <typename VoxelType>
uint16_t Volume<VoxelType>::getLongestSideLength(void) const
{
return m_uLongestSideLength;
}
template <typename VoxelType>
uint16_t Volume<VoxelType>::getShortestSideLength(void) const
{
return m_uShortestSideLength;
}
template <typename VoxelType>
uint16_t Volume<VoxelType>::getWidth(void) const
{
return m_uWidth;
}
template <typename VoxelType>
VoxelType Volume<VoxelType>::getVoxelAt(uint16_t uXPos, uint16_t uYPos, uint16_t uZPos) const
{
assert(uXPos < getWidth());
assert(uYPos < getHeight());
assert(uZPos < getDepth());
const uint16_t blockX = uXPos >> m_uBlockSideLengthPower;
const uint16_t blockY = uYPos >> m_uBlockSideLengthPower;
const uint16_t blockZ = uZPos >> m_uBlockSideLengthPower;
const uint16_t xOffset = uXPos - (blockX << m_uBlockSideLengthPower);
const uint16_t yOffset = uYPos - (blockY << m_uBlockSideLengthPower);
const uint16_t zOffset = uZPos - (blockZ << m_uBlockSideLengthPower);
const POLYVOX_SHARED_PTR< BlockData< VoxelType > >& block = m_pBlocks
[
blockX +
blockY * m_uWidthInBlocks +
blockZ * m_uWidthInBlocks * m_uHeightInBlocks
];
return block->getVoxelAt(xOffset,yOffset,zOffset);
}
template <typename VoxelType>
VoxelType Volume<VoxelType>::getVoxelAt(const Vector3DUint16& v3dPos) const
{
assert(v3dPos.getX() < m_uWidth);
assert(v3dPos.getY() < m_uHeight);
assert(v3dPos.getZ() < m_uDepth);
return getVoxelAt(v3dPos.getX(), v3dPos.getY(), v3dPos.getZ());
}
#pragma endregion
#pragma region Setters
template <typename VoxelType>
void Volume<VoxelType>::setVoxelAt(uint16_t uXPos, uint16_t uYPos, uint16_t uZPos, VoxelType tValue)
{
const uint16_t blockX = uXPos >> m_uBlockSideLengthPower;
const uint16_t blockY = uYPos >> m_uBlockSideLengthPower;
const uint16_t blockZ = uZPos >> m_uBlockSideLengthPower;
const uint16_t xOffset = uXPos - (blockX << m_uBlockSideLengthPower);
const uint16_t yOffset = uYPos - (blockY << m_uBlockSideLengthPower);
const uint16_t zOffset = uZPos - (blockZ << m_uBlockSideLengthPower);
uint32_t uBlockIndex =
blockX +
blockY * m_uWidthInBlocks +
blockZ * m_uWidthInBlocks * m_uHeightInBlocks;
POLYVOX_SHARED_PTR< BlockData<VoxelType> >& block = m_pBlocks[uBlockIndex];
//It's quite possible that the user might attempt to set a voxel to it's current value.
//We test for this case firstly because it could help performance, but more importantly
//because it lets us avoid unsharing blocks unnecessarily.
if(block->getVoxelAt(xOffset, yOffset, zOffset) != tValue)
{
if(block.unique())
{
block->setVoxelAt(xOffset,yOffset,zOffset, tValue);
//There is a chance that setting this voxel makes the block homogenous and therefore shareable.
//But checking this will take some time, so for now just set a flag.
m_vecBlockIsPotentiallyHomogenous[uBlockIndex] = true;
}
else
{
POLYVOX_SHARED_PTR< BlockData<VoxelType> > pNewBlockData(new BlockData<VoxelType>(*(block)));
block = pNewBlockData;
m_vecBlockIsPotentiallyHomogenous[uBlockIndex] = false;
block->setVoxelAt(xOffset,yOffset,zOffset, tValue);
}
}
}
template <typename VoxelType>
void Volume<VoxelType>::setVoxelAt(const Vector3DUint16& v3dPos, VoxelType tValue)
{
setVoxelAt(v3dPos.getX(), v3dPos.getY(), v3dPos.getZ(), tValue);
}
#pragma endregion
#pragma region Other
template <typename VoxelType>
void Volume<VoxelType>::tidyUpMemory(uint32_t uNoOfBlocksToProcess)
{
//Track the number of blocks we have processed.
uint32_t m_uNoOfProcessedBlocks = 0;
//We will loop around, and finish if we get back to our start position
uint32_t uFinishBlock = m_uCurrentBlockForTidying;
//Increment the current block, looping around if necessary
++m_uCurrentBlockForTidying;
m_uCurrentBlockForTidying %= m_uNoOfBlocksInVolume;
//While we have not reached the user specified limit and there are more blocks to process...
while((m_uNoOfProcessedBlocks < uNoOfBlocksToProcess) && (m_uCurrentBlockForTidying != uFinishBlock))
{
//We only do any work if the block is flagged as potentially homogeneous.
if(m_vecBlockIsPotentiallyHomogenous[m_uCurrentBlockForTidying])
{
//Check if it's really homogeneous (this can be slow).
if(m_pBlocks[m_uCurrentBlockForTidying]->isHomogeneous())
{
//If so, replace is with a block from out homogeneous collection.
VoxelType homogeneousValue = m_pBlocks[m_uCurrentBlockForTidying]->getVoxelAt(0,0,0);
m_pBlocks[m_uCurrentBlockForTidying] = getHomogenousBlockData(homogeneousValue);
}
//Either way, we have now determined whether the block was sharable. So it's not *potentially* sharable.
m_vecBlockIsPotentiallyHomogenous[m_uCurrentBlockForTidying] = false;
//We've processed a block. This is inside the 'if' because the path outside the 'if' is trivially fast.
++m_uNoOfProcessedBlocks;
}
//Increment the current block, looping around if necessary
++m_uCurrentBlockForTidying;
m_uCurrentBlockForTidying %= m_uNoOfBlocksInVolume;
}
//Identify and remove any homogeneous blocks which are not actually in use.
typename std::map<VoxelType, POLYVOX_SHARED_PTR< BlockData<VoxelType> > >::iterator iter = m_pHomogenousBlockData.begin();
while(iter != m_pHomogenousBlockData.end())
{
if(iter->second.unique())
{
m_pHomogenousBlockData.erase(iter++); //Increments the iterator and returns the previous position to be erased.
}
else
{
++iter; //Just increments the iterator.
}
}
}
template <typename VoxelType>
bool Volume<VoxelType>::isRegionHomogenous(const Region& region)
{
VolumeIterator<VoxelType> iter(*this);
iter.setValidRegion(region);
iter.setPosition(static_cast<Vector3DInt16>(region.getLowerCorner()));
VoxelType tFirst = iter.getVoxel();
iter.moveForwardInRegionXYZ();
do
{
VoxelType tCurrent = iter.getVoxel();
if(tCurrent != tFirst)
{
return false;
}
}while(iter.moveForwardInRegionXYZ());
return true;
}
#pragma endregion
#pragma region Private Implementation
template <typename VoxelType>
POLYVOX_SHARED_PTR< BlockData<VoxelType> > Volume<VoxelType>::getHomogenousBlockData(VoxelType tHomogenousValue) const
{
typename std::map<VoxelType, POLYVOX_SHARED_PTR< BlockData<VoxelType> > >::iterator iterResult = m_pHomogenousBlockData.find(tHomogenousValue);
if(iterResult == m_pHomogenousBlockData.end())
{
//Block<VoxelType> block;
POLYVOX_SHARED_PTR< BlockData<VoxelType> > pHomogeneousBlock(new BlockData<VoxelType>(m_uBlockSideLength));
//block.m_pBlockData = temp;
//block.m_uReferenceCount++;
pHomogeneousBlock->fill(tHomogenousValue);
m_pHomogenousBlockData.insert(std::make_pair(tHomogenousValue, pHomogeneousBlock));
return pHomogeneousBlock;
}
else
{
//iterResult->second.m_uReferenceCount++;
//POLYVOX_SHARED_PTR< BlockData<VoxelType> > result(iterResult->second);
return iterResult->second;
}
}
#pragma endregion
}