Hi Brad,
This is a bug, there shouldn't be a change in behavior w.r.t Texture
updating and PBO's. The mechanism behind is much changed but the
behavior should be the same.
I have looked at the code and found the that setImage and many other
others are doing a ++_modifiedCount rather than calling dirty(). This
will be bypassing the new mechanism that keeps the PBO in sync. I
have just gone through Image.cpp changing all the ++_modifiedCount
case to dirty(). The modified file is attached. The change are also
checked in to SVN.
I have tested the change on a movie file on my box and its working
fine, where a couple of days ago it wasn't working it was just loading
the first frame. I assumed it was a bug in the xine-lib plugin,
but... now know it wasn't, and now is fixed thanks to you uncovering
the right rock :-)
Could you test it out and let me know how you get on.
Robert.
On 8/17/07, Christiansen, Brad <[EMAIL PROTECTED]> wrote:
> Hi Robert,
>
> As part of the task in moving our applications to OSG 2.0 I discovered
> there is a change in the behaviour of the API when binding an image with
> a PBO to a TextureRectangle.
>
> Using 1.2 we could set the image on the texture and then simply call
> setImage on the image object and the TextureRectangle would automaticaly
> update. Since moving to 2.0 we have to call dirty on the Image after the
> setImage call. This is not a big deal for me as it is an easy change to
> make but I thought this might not have been an intentional change. I had
> a quick look to see what this was the case but couldn't track it down.
>
> This is what the code looks like:
>
> //set things up
> image = new Image();
> image->setPixelBufferObject(new PixelBufferObject(image));
> m_YTexture->setImage(image);
>
> //the on update call...
> Vimage->setImage(0.5 * m_imageWidth,0.5 * m_imageHeight,1,GL_LUMINANCE,
> GL_LUMINANCE,
> GL_UNSIGNED_BYTE,VBuffer,Image::AllocationMode::NO_DELETE);
> //Since 2.0 we now also have to call the following for the texture to
> update
> Vimage->dirty();
>
>
> As I said this is not an issue as such just thought you might have
> wanted the behaviour to be consistent between versions.
>
> Cheers,
>
> Brad
>
>
>
> DISCLAIMER:---------------------------------------------------------------------------
> This Email may contain confidential and/or privileged information and is
> intended
> solely for the addressee(s) named. If you have received this information in
> error, or
> are advised that you have been posted this Email by accident, please notify
> the
> sender by return Email, do not redistribute it, delete the Email and keep no
> copies.
> --------------------------------------------------------------------------------------
>
> _______________________________________________
> osg-users mailing list
> [email protected]
> http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
>
/* -*-c++-*- OpenSceneGraph - Copyright (C) 1998-2006 Robert Osfield
*
* This library is open source and may be redistributed and/or modified under
* the terms of the OpenSceneGraph Public License (OSGPL) version 0.0 or
* (at your option) any later version. The full license is in LICENSE file
* included with this distribution, and on the openscenegraph.org website.
*
* This library 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
* OpenSceneGraph Public License for more details.
*/
#include <osg/GL>
#include <osg/GLU>
#include <osg/Image>
#include <osg/Notify>
#include <osg/Geode>
#include <osg/Geometry>
#include <osg/StateSet>
#include <osg/Texture2D>
#include <osg/Texture3D>
#include "dxtctool.h"
using namespace osg;
using namespace std;
Image::Image()
{
setDataVariance(STATIC);
_fileName = "";
_origin = BOTTOM_LEFT;
_s = _t = _r = 0;
_internalTextureFormat = 0;
_pixelFormat = (unsigned int)0;
_dataType = (unsigned int)0;
_packing = 4;
_allocationMode = USE_NEW_DELETE;
_data = (unsigned char *)0L;
_modifiedCount = 0;
}
Image::Image(const Image& image,const CopyOp& copyop):
Object(image,copyop),
_fileName(image._fileName),
_origin(image._origin),
_s(image._s), _t(image._t), _r(image._r),
_internalTextureFormat(image._internalTextureFormat),
_pixelFormat(image._pixelFormat),
_dataType(image._dataType),
_packing(image._packing),
_data(0L),
_modifiedCount(image._modifiedCount),
_mipmapData(image._mipmapData)
{
if (image._data)
{
int size = image.getTotalSizeInBytesIncludingMipmaps();
setData(new unsigned char [size],USE_NEW_DELETE);
memcpy(_data,image._data,size);
}
}
Image::~Image()
{
deallocateData();
}
void Image::deallocateData()
{
if (_data) {
if (_allocationMode==USE_NEW_DELETE) delete [] _data;
else if (_allocationMode==USE_MALLOC_FREE) ::free(_data);
_data = 0;
}
}
int Image::compare(const Image& rhs) const
{
// if at least one filename is empty, then need to test buffer
// pointers because images could have been created on the fly
// and therefore we can't rely on file names to get an accurate
// comparison
if (getFileName().empty() || rhs.getFileName().empty())
{
if (_data<rhs._data) return -1;
if (_data>rhs._data) return 1;
}
// need to test against image contents here...
COMPARE_StateAttribute_Parameter(_s)
COMPARE_StateAttribute_Parameter(_t)
COMPARE_StateAttribute_Parameter(_internalTextureFormat)
COMPARE_StateAttribute_Parameter(_pixelFormat)
COMPARE_StateAttribute_Parameter(_dataType)
COMPARE_StateAttribute_Parameter(_packing)
COMPARE_StateAttribute_Parameter(_mipmapData)
COMPARE_StateAttribute_Parameter(_modifiedCount)
// same buffer + same parameters = same image
if ((_data || rhs._data) && (_data == rhs._data)) return 0;
// slowest comparison at the bottom!
COMPARE_StateAttribute_Parameter(getFileName())
return 0;
}
void Image::setFileName(const std::string& fileName)
{
_fileName = fileName;
}
void Image::setData(unsigned char* data, AllocationMode mode)
{
deallocateData();
_data = data;
_allocationMode = mode;
}
bool Image::isPackedType(GLenum type)
{
switch(type)
{
case(GL_UNSIGNED_BYTE_3_3_2):
case(GL_UNSIGNED_BYTE_2_3_3_REV):
case(GL_UNSIGNED_SHORT_5_6_5):
case(GL_UNSIGNED_SHORT_5_6_5_REV):
case(GL_UNSIGNED_SHORT_4_4_4_4):
case(GL_UNSIGNED_SHORT_4_4_4_4_REV):
case(GL_UNSIGNED_SHORT_5_5_5_1):
case(GL_UNSIGNED_SHORT_1_5_5_5_REV):
case(GL_UNSIGNED_INT_8_8_8_8):
case(GL_UNSIGNED_INT_8_8_8_8_REV):
case(GL_UNSIGNED_INT_10_10_10_2):
case(GL_UNSIGNED_INT_2_10_10_10_REV): return true;
default: return false;
}
}
GLenum Image::computePixelFormat(GLenum format)
{
switch(format)
{
case(GL_ALPHA16F_ARB):
case(GL_ALPHA32F_ARB):
return GL_ALPHA;
case(GL_LUMINANCE16F_ARB):
case(GL_LUMINANCE32F_ARB):
return GL_LUMINANCE;
case(GL_INTENSITY16F_ARB):
case(GL_INTENSITY32F_ARB):
return GL_INTENSITY;
case(GL_LUMINANCE_ALPHA16F_ARB):
case(GL_LUMINANCE_ALPHA32F_ARB):
return GL_LUMINANCE_ALPHA;
case(GL_RGB32F_ARB):
case(GL_RGB16F_ARB):
return GL_RGB;
case(GL_RGBA32F_ARB):
case(GL_RGBA16F_ARB):
return GL_RGBA;
default:
return format;
}
}
unsigned int Image::computeNumComponents(GLenum pixelFormat)
{
switch(pixelFormat)
{
case(GL_COMPRESSED_RGB_S3TC_DXT1_EXT): return 3;
case(GL_COMPRESSED_RGBA_S3TC_DXT1_EXT): return 4;
case(GL_COMPRESSED_RGBA_S3TC_DXT3_EXT): return 4;
case(GL_COMPRESSED_RGBA_S3TC_DXT5_EXT): return 4;
case(GL_COLOR_INDEX): return 1;
case(GL_STENCIL_INDEX): return 1;
case(GL_DEPTH_COMPONENT): return 1;
case(GL_RED): return 1;
case(GL_GREEN): return 1;
case(GL_BLUE): return 1;
case(GL_ALPHA): return 1;
case(GL_ALPHA16F_ARB): return 1;
case(GL_ALPHA32F_ARB): return 1;
case(GL_RGB): return 3;
case(GL_BGR): return 3;
case(GL_RGB16F_ARB): return 3;
case(GL_RGB32F_ARB): return 3;
case(GL_RGBA): return 4;
case(GL_BGRA): return 4;
case(GL_LUMINANCE): return 1;
case(GL_LUMINANCE4): return 1;
case(GL_LUMINANCE8): return 1;
case(GL_LUMINANCE12): return 1;
case(GL_LUMINANCE16): return 1;
case(GL_LUMINANCE16F_ARB): return 1;
case(GL_LUMINANCE32F_ARB): return 1;
case(GL_LUMINANCE4_ALPHA4): return 2;
case(GL_LUMINANCE6_ALPHA2): return 2;
case(GL_LUMINANCE8_ALPHA8): return 2;
case(GL_LUMINANCE12_ALPHA4): return 2;
case(GL_LUMINANCE12_ALPHA12): return 2;
case(GL_LUMINANCE16_ALPHA16): return 2;
case(GL_INTENSITY): return 1;
case(GL_INTENSITY4): return 1;
case(GL_INTENSITY8): return 1;
case(GL_INTENSITY12): return 1;
case(GL_INTENSITY16): return 1;
case(GL_INTENSITY16F_ARB): return 1;
case(GL_INTENSITY32F_ARB): return 1;
case(GL_LUMINANCE_ALPHA): return 2;
case(GL_LUMINANCE_ALPHA16F_ARB): return 2;
case(GL_LUMINANCE_ALPHA32F_ARB): return 2;
case(GL_HILO_NV): return 2;
case(GL_DSDT_NV): return 2;
case(GL_DSDT_MAG_NV): return 3;
case(GL_DSDT_MAG_VIB_NV): return 4;
default:
{
notify(WARN)<<"error pixelFormat = "<<std::hex<<pixelFormat<<std::endl;
return 0;
}
}
}
unsigned int Image::computePixelSizeInBits(GLenum format,GLenum type)
{
switch(format)
{
case(GL_COMPRESSED_RGB_S3TC_DXT1_EXT): return 4;
case(GL_COMPRESSED_RGBA_S3TC_DXT1_EXT): return 4;
case(GL_COMPRESSED_RGBA_S3TC_DXT3_EXT): return 8;
case(GL_COMPRESSED_RGBA_S3TC_DXT5_EXT): return 8;
default: break;
}
// note, haven't yet added proper handling of the ARB GL_COMPRESSRED_* pathways
// yet, no clear size for these since its probably implementation dependent
// which raises the question of how to actually querry for these sizes...
// will need to revisit this issue, for now just report an error.
// this is possible a bit of mute point though as since the ARB compressed formats
// arn't yet used for storing images to disk, so its likely that users wont have
// osg::Image's for pixel formats set the ARB compressed formats, just using these
// compressed formats as internal texture modes. This is very much speculation though
// if get the below error then its time to revist this issue :-)
// Robert Osfield, Jan 2005.
switch(format)
{
case(GL_COMPRESSED_ALPHA):
case(GL_COMPRESSED_LUMINANCE):
case(GL_COMPRESSED_LUMINANCE_ALPHA):
case(GL_COMPRESSED_INTENSITY):
case(GL_COMPRESSED_RGB):
case(GL_COMPRESSED_RGBA):
notify(WARN)<<"Image::computePixelSizeInBits(format,type) : cannot compute correct size of compressed format ("<<format<<") returning 0."<<std::endl;
return 0;
default: break;
}
switch(format)
{
case(GL_LUMINANCE4): return 4;
case(GL_LUMINANCE8): return 8;
case(GL_LUMINANCE12): return 12;
case(GL_LUMINANCE16): return 16;
case(GL_LUMINANCE4_ALPHA4): return 8;
case(GL_LUMINANCE6_ALPHA2): return 8;
case(GL_LUMINANCE8_ALPHA8): return 16;
case(GL_LUMINANCE12_ALPHA4): return 16;
case(GL_LUMINANCE12_ALPHA12): return 24;
case(GL_LUMINANCE16_ALPHA16): return 32;
case(GL_INTENSITY4): return 4;
case(GL_INTENSITY8): return 8;
case(GL_INTENSITY12): return 12;
case(GL_INTENSITY16): return 16;
default: break;
}
switch(type)
{
case(GL_BITMAP): return computeNumComponents(format);
case(GL_BYTE):
case(GL_UNSIGNED_BYTE): return 8*computeNumComponents(format);
case(GL_HALF_FLOAT_NV):
case(GL_SHORT):
case(GL_UNSIGNED_SHORT): return 16*computeNumComponents(format);
case(GL_INT):
case(GL_UNSIGNED_INT):
case(GL_FLOAT): return 32*computeNumComponents(format);
case(GL_UNSIGNED_BYTE_3_3_2):
case(GL_UNSIGNED_BYTE_2_3_3_REV): return 8;
case(GL_UNSIGNED_SHORT_5_6_5):
case(GL_UNSIGNED_SHORT_5_6_5_REV):
case(GL_UNSIGNED_SHORT_4_4_4_4):
case(GL_UNSIGNED_SHORT_4_4_4_4_REV):
case(GL_UNSIGNED_SHORT_5_5_5_1):
case(GL_UNSIGNED_SHORT_1_5_5_5_REV): return 16;
case(GL_UNSIGNED_INT_8_8_8_8):
case(GL_UNSIGNED_INT_8_8_8_8_REV):
case(GL_UNSIGNED_INT_10_10_10_2):
case(GL_UNSIGNED_INT_2_10_10_10_REV): return 32;
default:
{
notify(WARN)<<"error type = "<<type<<std::endl;
return 0;
}
}
}
unsigned int Image::computeRowWidthInBytes(int width,GLenum pixelFormat,GLenum type,int packing)
{
unsigned int pixelSize = computePixelSizeInBits(pixelFormat,type);
int widthInBits = width*pixelSize;
int packingInBits = packing*8;
//notify(INFO) << "width="<<width<<" pixelSize="<<pixelSize<<" width in bit="<<widthInBits<<" packingInBits="<<packingInBits<<" widthInBits%packingInBits="<<widthInBits%packingInBits<<std::endl;
return (widthInBits/packingInBits + ((widthInBits%packingInBits)?1:0))*packing;
}
int Image::computeNearestPowerOfTwo(int s,float bias)
{
if ((s & (s-1))!=0)
{
// it isn't so lets find the closest power of two.
// yes, logf and powf are slow, but this code should
// only be called during scene graph initilization,
// if at all, so not critical in the greater scheme.
float p2 = logf((float)s)/logf(2.0f);
float rounded_p2 = floorf(p2+bias);
s = (int)(powf(2.0f,rounded_p2));
}
return s;
}
unsigned int Image::getTotalSizeInBytesIncludingMipmaps() const
{
if (_mipmapData.empty())
{
// no mips so just return size of main image
return getTotalSizeInBytes();
}
int s = _s;
int t = _t;
int r = _r;
unsigned int maxValue = 0;
for(unsigned int i=0;i<_mipmapData.size() && _mipmapData[i];++i)
{
s >>= 1;
t >>= 1;
r >>= 1;
maxValue = maximum(maxValue,_mipmapData[i]);
}
if (s==0) s=1;
if (t==0) t=1;
if (r==0) r=1;
unsigned int sizeOfLastMipMap = computeRowWidthInBytes(s,_pixelFormat,_dataType,_packing)* r*t;
switch(_pixelFormat)
{
case(GL_COMPRESSED_RGB_S3TC_DXT1_EXT):
case(GL_COMPRESSED_RGBA_S3TC_DXT1_EXT):
sizeOfLastMipMap = maximum(sizeOfLastMipMap, 8u); // block size of 8
break;
case(GL_COMPRESSED_RGBA_S3TC_DXT3_EXT):
case(GL_COMPRESSED_RGBA_S3TC_DXT5_EXT):
sizeOfLastMipMap = maximum(sizeOfLastMipMap, 16u); // block size of 16
break;
default: break;
}
// notify(INFO)<<"sizeOfLastMipMap="<<sizeOfLastMipMap<<"\ts="<<s<<"\tt="<<t<<"\tr"<<r<<std::endl;
return maxValue+sizeOfLastMipMap;
}
void Image::setInternalTextureFormat(GLint internalFormat)
{
// won't do any sanity checking right now, leave it to
// OpenGL to make the call.
_internalTextureFormat = internalFormat;
}
void Image::setPixelFormat(GLenum pixelFormat)
{
if (_pixelFormat==pixelFormat) return; // do nothing if the same.
if (_pixelFormat==0 || computeNumComponents(_pixelFormat)==computeNumComponents(pixelFormat))
{
// if the two formats have the same number of componets then
// we can do a straight swap.
_pixelFormat = pixelFormat;
}
else
{
notify(WARN)<<"Image::setPixelFormat(..) - warning, attempt to reset the pixel format with a different number of components."<<std::endl;
}
}
void Image::setDataType(GLenum dataType)
{
if (_dataType==dataType) return; // do nothing if the same.
if (_dataType==0)
{
// setting the datatype for the first time
_dataType = dataType;
}
else
{
notify(WARN)<<"Image::setDataType(..) - warning, attempt to reset the data type not permitted."<<std::endl;
}
}
void Image::allocateImage(int s,int t,int r,
GLenum format,GLenum type,
int packing)
{
_mipmapData.clear();
unsigned int previousTotalSize = 0;
if (_data) previousTotalSize = computeRowWidthInBytes(_s,_pixelFormat,_dataType,_packing)*_t*_r;
unsigned int newTotalSize = computeRowWidthInBytes(s,format,type,packing)*t*r;
if (newTotalSize!=previousTotalSize)
{
if (newTotalSize)
setData(new unsigned char [newTotalSize],USE_NEW_DELETE);
else
deallocateData(); // and sets it to NULL.
}
if (_data)
{
_s = s;
_t = t;
_r = r;
_pixelFormat = format;
_dataType = type;
_packing = packing;
_internalTextureFormat = format;
}
else
{
// throw exception?? not for now, will simply set values to 0.
_s = 0;
_t = 0;
_r = 0;
_pixelFormat = 0;
_dataType = 0;
_packing = 0;
_internalTextureFormat = 0;
}
dirty();
}
void Image::setImage(int s,int t,int r,
GLint internalTextureFormat,
GLenum format,GLenum type,
unsigned char *data,
AllocationMode mode,
int packing)
{
_mipmapData.clear();
_s = s;
_t = t;
_r = r;
_internalTextureFormat = internalTextureFormat;
_pixelFormat = format;
_dataType = type;
setData(data,mode);
_packing = packing;
dirty();
}
void Image::readPixels(int x,int y,int width,int height,
GLenum format,GLenum type)
{
allocateImage(width,height,1,format,type);
glPixelStorei(GL_PACK_ALIGNMENT,_packing);
glReadPixels(x,y,width,height,format,type,_data);
}
void Image::readImageFromCurrentTexture(unsigned int contextID, bool copyMipMapsIfAvailable, GLenum type)
{
const osg::Texture::Extensions* extensions = osg::Texture::getExtensions(contextID,true);
const osg::Texture3D::Extensions* extensions3D = osg::Texture3D::getExtensions(contextID,true);
GLboolean binding1D, binding2D, binding3D;
glGetBooleanv(GL_TEXTURE_BINDING_1D, &binding1D);
glGetBooleanv(GL_TEXTURE_BINDING_2D, &binding2D);
glGetBooleanv(GL_TEXTURE_BINDING_3D, &binding3D);
GLenum textureMode = binding1D ? GL_TEXTURE_1D : binding2D ? GL_TEXTURE_2D : binding3D ? GL_TEXTURE_3D : 0;
if (textureMode==0) return;
GLint internalformat;
GLint width;
GLint height;
GLint depth;
GLint packing;
GLint numMipMaps = 0;
if (copyMipMapsIfAvailable)
{
for(;numMipMaps<20;++numMipMaps)
{
glGetTexLevelParameteriv(textureMode, numMipMaps, GL_TEXTURE_WIDTH, &width);
glGetTexLevelParameteriv(textureMode, numMipMaps, GL_TEXTURE_HEIGHT, &height);
glGetTexLevelParameteriv(textureMode, numMipMaps, GL_TEXTURE_DEPTH, &depth);
if (width==0 || height==0 || depth==0) break;
}
}
else
{
numMipMaps = 1;
}
GLint compressed = 0;
if (textureMode==GL_TEXTURE_2D)
{
if (extensions->isCompressedTexImage2DSupported())
{
glGetTexLevelParameteriv(GL_TEXTURE_2D, 0, GL_TEXTURE_COMPRESSED_ARB,&compressed);
}
}
else if (textureMode==GL_TEXTURE_3D)
{
if (extensions3D->isCompressedTexImage3DSupported())
{
glGetTexLevelParameteriv(GL_TEXTURE_3D, 0, GL_TEXTURE_COMPRESSED_ARB,&compressed);
}
}
/* if the compression has been successful */
if (compressed == GL_TRUE)
{
MipmapDataType mipMapData;
unsigned int total_size = 0;
GLint i;
for(i=0;i<numMipMaps;++i)
{
if (i>0) mipMapData.push_back(total_size);
GLint compressed_size;
glGetTexLevelParameteriv(textureMode, i, GL_TEXTURE_COMPRESSED_IMAGE_SIZE_ARB, &compressed_size);
total_size += compressed_size;
}
unsigned char* data = new unsigned char[total_size];
if (!data)
{
osg::notify(osg::WARN)<<"Warning: Image::readImageFromCurrentTexture(..) out of memory, now image read."<<std::endl;
return;
}
deallocateData(); // and sets it to NULL.
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_INTERNAL_FORMAT, &internalformat);
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_WIDTH, &width);
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_HEIGHT, &height);
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_DEPTH, &depth);
glGetIntegerv(GL_UNPACK_ALIGNMENT, &packing);
glPixelStorei(GL_PACK_ALIGNMENT, packing);
_data = data;
_s = width;
_t = height;
_r = depth;
_pixelFormat = internalformat;
_dataType = type;
_internalTextureFormat = internalformat;
_mipmapData = mipMapData;
_allocationMode=USE_NEW_DELETE;
_packing = packing;
for(i=0;i<numMipMaps;++i)
{
extensions->glGetCompressedTexImage(textureMode, i, getMipmapData(i));
}
dirty();
}
else
{
MipmapDataType mipMapData;
// Get the internal texture format and packing value from OpenGL,
// instead of using possibly outdated values from the class.
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_INTERNAL_FORMAT, &internalformat);
glGetIntegerv(GL_UNPACK_ALIGNMENT, &packing);
glPixelStorei(GL_PACK_ALIGNMENT, packing);
unsigned int total_size = 0;
GLint i;
for(i=0;i<numMipMaps;++i)
{
if (i>0) mipMapData.push_back(total_size);
glGetTexLevelParameteriv(textureMode, i, GL_TEXTURE_WIDTH, &width);
glGetTexLevelParameteriv(textureMode, i, GL_TEXTURE_HEIGHT, &height);
glGetTexLevelParameteriv(textureMode, i, GL_TEXTURE_DEPTH, &depth);
unsigned int level_size = computeRowWidthInBytes(width,internalformat,type,packing)*height*depth;
total_size += level_size;
}
unsigned char* data = new unsigned char[total_size];
if (!data)
{
osg::notify(osg::WARN)<<"Warning: Image::readImageFromCurrentTexture(..) out of memory, now image read."<<std::endl;
return;
}
deallocateData(); // and sets it to NULL.
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_WIDTH, &width);
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_HEIGHT, &height);
glGetTexLevelParameteriv(textureMode, 0, GL_TEXTURE_DEPTH, &depth);
_data = data;
_s = width;
_t = height;
_r = depth;
_pixelFormat = computePixelFormat(internalformat);
_dataType = type;
_internalTextureFormat = internalformat;
_mipmapData = mipMapData;
_allocationMode=USE_NEW_DELETE;
_packing = packing;
for(i=0;i<numMipMaps;++i)
{
glGetTexImage(textureMode,i,_pixelFormat,_dataType,getMipmapData(i));
}
dirty();
}
}
void Image::scaleImage(int s,int t,int r, GLenum newDataType)
{
if (_s==s && _t==t && _r==r) return;
if (_data==NULL)
{
notify(WARN) << "Error Image::scaleImage() do not succeed : cannot scale NULL image."<<std::endl;
return;
}
if (_r!=1 || r!=1)
{
notify(WARN) << "Error Image::scaleImage() do not succeed : scaling of volumes not implemented."<<std::endl;
return;
}
unsigned int newTotalSize = computeRowWidthInBytes(s,_pixelFormat,newDataType,_packing)*t;
// need to sort out what size to really use...
unsigned char* newData = new unsigned char [newTotalSize];
if (!newData)
{
// should we throw an exception??? Just return for time being.
notify(FATAL) << "Error Image::scaleImage() do not succeed : out of memory."<<newTotalSize<<std::endl;
return;
}
glPixelStorei(GL_PACK_ALIGNMENT,_packing);
glPixelStorei(GL_UNPACK_ALIGNMENT,_packing);
GLint status = gluScaleImage(_pixelFormat,
_s,
_t,
_dataType,
_data,
s,
t,
newDataType,
newData);
if (status==0)
{
// free old image.
_s = s;
_t = t;
_dataType = newDataType;
setData(newData,USE_NEW_DELETE);
}
else
{
delete [] newData;
notify(WARN) << "Error Image::scaleImage() did not succeed : errorString = "<<gluErrorString((GLenum)status)<<std::endl;
}
dirty();
}
void Image::copySubImage(int s_offset,int t_offset,int r_offset,osg::Image* source)
{
if (!source) return;
if (s_offset<0 || t_offset<0 || r_offset<0)
{
notify(WARN)<<"Warning: negative offsets passed to Image::copySubImage(..) not supported, operation ignored."<<std::endl;
return;
}
if (!_data)
{
notify(INFO)<<"allocating image"<<endl;
allocateImage(s_offset+source->s(),t_offset+source->t(),r_offset+source->r(),
source->getPixelFormat(),source->getDataType(),
source->getPacking());
}
if (s_offset>=_s || t_offset>=_t || r_offset>=_r)
{
notify(WARN)<<"Warning: offsets passed to Image::copySubImage(..) outside destination image, operation ignored."<<std::endl;
return;
}
if (_pixelFormat != source->getPixelFormat())
{
notify(WARN)<<"Warning: image with an incompatible pixel formats passed to Image::copySubImage(..), operation ignored."<<std::endl;
return;
}
void* data_destination = data(s_offset,t_offset,r_offset);
glPixelStorei(GL_PACK_ALIGNMENT,source->getPacking());
glPixelStorei(GL_PACK_ROW_LENGTH,_s);
glPixelStorei(GL_UNPACK_ALIGNMENT,_packing);
GLint status = gluScaleImage(_pixelFormat,
source->s(),
source->t(),
source->getDataType(),
source->data(),
source->s(),
source->t(),
_dataType,
data_destination);
glPixelStorei(GL_PACK_ROW_LENGTH,0);
if (status!=0)
{
notify(WARN) << "Error Image::scaleImage() do not succeed : errorString = "<<gluErrorString((GLenum)status)<<std::endl;
}
}
void Image::flipHorizontal()
{
if (_data==NULL)
{
notify(WARN) << "Error Image::flipHorizontal() did not succeed : cannot flip NULL image."<<std::endl;
return;
}
unsigned int elemSize = getPixelSizeInBits()/8;
if (_mipmapData.empty())
{
for(int r=0;r<_r;++r)
{
for (int t=0; t<_t; ++t)
{
unsigned char* rowData = _data+t*getRowSizeInBytes()+r*getImageSizeInBytes();
unsigned char* left = rowData ;
unsigned char* right = rowData + ((_s-1)*getPixelSizeInBits())/8;
while (left < right)
{
char tmp[32]; // max elem size is four floats
memcpy(tmp, left, elemSize);
memcpy(left, right, elemSize);
memcpy(right, tmp, elemSize);
left += elemSize;
right -= elemSize;
}
}
}
}
else
{
notify(WARN) << "Error Image::flipHorizontal() did not succeed : cannot flip mipmapped image."<<std::endl;
return;
}
dirty();
}
void flipImageVertical(unsigned char* top, unsigned char* bottom, unsigned int rowSize)
{
while(top<bottom)
{
for(unsigned int i=0;i<rowSize;++i, ++top,++bottom)
{
unsigned char temp=*top;
*top = *bottom;
*bottom = temp;
}
bottom -= 2*rowSize;
}
}
void Image::flipVertical()
{
if (_data==NULL)
{
notify(WARN) << "Error Image::flipVertical() do not succeed : cannot flip NULL image."<<std::endl;
return;
}
if (!_mipmapData.empty() && _r>1)
{
notify(WARN) << "Error Image::flipVertical() do not succeed : flipping of mipmap 3d textures not yet supported."<<std::endl;
return;
}
if (_mipmapData.empty())
{
// no mipmaps,
// so we can safely handle 3d textures
for(int r=0;r<_r;++r)
{
if (!dxtc_tool::VerticalFlip(_s,_t,_pixelFormat,data(0,0,r)))
{
// its not a compressed image, so implement flip oursleves.
unsigned int rowSize = computeRowWidthInBytes(_s,_pixelFormat,_dataType,_packing);
unsigned char* top = data(0,0,r);
unsigned char* bottom = top + (_t-1)*rowSize;
flipImageVertical(top, bottom, rowSize);
}
}
}
else if (_r==1)
{
if (!dxtc_tool::VerticalFlip(_s,_t,_pixelFormat,_data))
{
// its not a compressed image, so implement flip oursleves.
unsigned int rowSize = computeRowWidthInBytes(_s,_pixelFormat,_dataType,_packing);
unsigned char* top = data(0,0,0);
unsigned char* bottom = top + (_t-1)*rowSize;
flipImageVertical(top, bottom, rowSize);
}
int s = _s;
int t = _t;
//int r = _r;
for(unsigned int i=0;i<_mipmapData.size() && _mipmapData[i];++i)
{
s >>= 1;
t >>= 1;
if (s==0) s=1;
if (t==0) t=1;
if (!dxtc_tool::VerticalFlip(s,t,_pixelFormat,_data+_mipmapData[i]))
{
// its not a compressed image, so implement flip oursleves.
unsigned int rowSize = computeRowWidthInBytes(s,_pixelFormat,_dataType,_packing);
unsigned char* top = _data+_mipmapData[i];
unsigned char* bottom = top + (t-1)*rowSize;
flipImageVertical(top, bottom, rowSize);
}
}
}
dirty();
}
void Image::ensureValidSizeForTexturing(GLint maxTextureSize)
{
int new_s = computeNearestPowerOfTwo(_s);
int new_t = computeNearestPowerOfTwo(_t);
if (new_s>maxTextureSize) new_s = maxTextureSize;
if (new_t>maxTextureSize) new_t = maxTextureSize;
if (new_s!=_s || new_t!=_t)
{
if (!_fileName.empty()) notify(NOTICE) << "Scaling image '"<<_fileName<<"' from ("<<_s<<","<<_t<<") to ("<<new_s<<","<<new_t<<")"<<std::endl;
else notify(NOTICE) << "Scaling image from ("<<_s<<","<<_t<<") to ("<<new_s<<","<<new_t<<")"<<std::endl;
scaleImage(new_s,new_t,_r);
}
}
template <typename T>
bool _findLowerAlphaValueInRow(unsigned int num, T* data,T value, unsigned int delta)
{
for(unsigned int i=0;i<num;++i)
{
if (*data<value) return true;
data += delta;
}
return false;
}
template <typename T>
bool _maskedFindLowerAlphaValueInRow(unsigned int num, T* data,T value, T mask, unsigned int delta)
{
for(unsigned int i=0;i<num;++i)
{
if ((*data & mask)<value) return true;
data += delta;
}
return false;
}
bool Image::isImageTranslucent() const
{
unsigned int offset = 0;
unsigned int delta = 1;
switch(_pixelFormat)
{
case(GL_ALPHA):
offset = 0;
delta = 1;
break;
case(GL_LUMINANCE_ALPHA):
offset = 1;
delta = 2;
break;
case(GL_RGBA):
offset = 3;
delta = 4;
break;
case(GL_BGRA):
offset = 3;
delta = 4;
break;
default:
return false;
}
for(int ir=0;ir<r();++ir)
{
for(int it=0;it<t();++it)
{
const unsigned char* d = data(0,it,ir);
switch(_dataType)
{
case(GL_BYTE):
if (_findLowerAlphaValueInRow(s(), (char*)d +offset, (char)127, delta))
return true;
break;
case(GL_UNSIGNED_BYTE):
if (_findLowerAlphaValueInRow(s(), (unsigned char*)d + offset, (unsigned char)255, delta))
return true;
break;
case(GL_SHORT):
if (_findLowerAlphaValueInRow(s(), (short*)d + offset, (short)32767, delta))
return true;
break;
case(GL_UNSIGNED_SHORT):
if (_findLowerAlphaValueInRow(s(), (unsigned short*)d + offset, (unsigned short)65535, delta))
return true;
break;
case(GL_INT):
if (_findLowerAlphaValueInRow(s(), (int*)d + offset, (int)2147483647, delta))
return true;
break;
case(GL_UNSIGNED_INT):
if (_findLowerAlphaValueInRow(s(), (unsigned int*)d + offset, 4294967295u, delta))
return true;
break;
case(GL_FLOAT):
if (_findLowerAlphaValueInRow(s(), (float*)d + offset, 1.0f, delta))
return true;
break;
case(GL_UNSIGNED_SHORT_5_5_5_1):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned short*)d,
(unsigned short)0x0001,
(unsigned short)0x0001, 1))
return true;
break;
case(GL_UNSIGNED_SHORT_1_5_5_5_REV):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned short*)d,
(unsigned short)0x8000,
(unsigned short)0x8000, 1))
return true;
break;
case(GL_UNSIGNED_SHORT_4_4_4_4):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned short*)d,
(unsigned short)0x000f,
(unsigned short)0x000f, 1))
return true;
break;
case(GL_UNSIGNED_SHORT_4_4_4_4_REV):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned short*)d,
(unsigned short)0xf000,
(unsigned short)0xf000, 1))
return true;
break;
case(GL_UNSIGNED_INT_10_10_10_2):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned int*)d,
0x00000003u,
0x00000003u, 1))
return true;
break;
case(GL_UNSIGNED_INT_2_10_10_10_REV):
if (_maskedFindLowerAlphaValueInRow(s(), (unsigned int*)d,
0xc0000000u,
0xc0000000u, 1))
return true;
break;
case(GL_HALF_FLOAT_NV):
if (_findLowerAlphaValueInRow(s(), (unsigned short*)d + offset,
(unsigned short)0x3c00, delta))
return true;
break;
}
}
}
return false;
}
///////////////////////////////////////////////////////////////////////////////
Geode* osg::createGeodeForImage(osg::Image* image)
{
return createGeodeForImage(image,image->s(),image->t());
}
Geode* osg::createGeodeForImage(osg::Image* image,float s,float t)
{
if (image)
{
if (s>0 && t>0)
{
float y = 1.0;
float x = y*(s/t);
// set up the texture.
osg::Texture2D* texture = new osg::Texture2D;
texture->setImage(image);
// set up the drawstate.
osg::StateSet* dstate = new osg::StateSet;
dstate->setMode(GL_CULL_FACE,osg::StateAttribute::OFF);
dstate->setMode(GL_LIGHTING,osg::StateAttribute::OFF);
dstate->setTextureAttributeAndModes(0, texture,osg::StateAttribute::ON);
// set up the geoset.
Geometry* geom = new Geometry;
geom->setStateSet(dstate);
Vec3Array* coords = new Vec3Array(4);
(*coords)[0].set(-x,0.0f,y);
(*coords)[1].set(-x,0.0f,-y);
(*coords)[2].set(x,0.0f,-y);
(*coords)[3].set(x,0.0f,y);
geom->setVertexArray(coords);
Vec2Array* tcoords = new Vec2Array(4);
(*tcoords)[0].set(0.0f,1.0f);
(*tcoords)[1].set(0.0f,0.0f);
(*tcoords)[2].set(1.0f,0.0f);
(*tcoords)[3].set(1.0f,1.0f);
geom->setTexCoordArray(0,tcoords);
osg::Vec4Array* colours = new osg::Vec4Array(1);
(*colours)[0].set(1.0f,1.0f,1.0,1.0f);
geom->setColorArray(colours);
geom->setColorBinding(Geometry::BIND_OVERALL);
geom->addPrimitiveSet(new DrawArrays(PrimitiveSet::QUADS,0,4));
// set up the geode.
osg::Geode* geode = new osg::Geode;
geode->addDrawable(geom);
return geode;
}
else
{
return NULL;
}
}
else
{
return NULL;
}
}
_______________________________________________
osg-users mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
