You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

2131 lines
97 KiB

/*
This file is part of Magnum.
Copyright © 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019
Vladimír Vondruš <mosra@centrum.cz>
Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and associated documentation files (the "Software"),
to deal in the Software without restriction, including without limitation
the rights to use, copy, modify, merge, publish, distribute, sublicense,
and/or sell copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included
in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
*/
#include "AbstractTexture.h"
#include <Corrade/Containers/Array.h>
#include "Magnum/Array.h"
#include "Magnum/Image.h"
#ifndef MAGNUM_TARGET_GLES2
#include "Magnum/GL/BufferImage.h"
#endif
#include "Magnum/GL/Context.h"
#include "Magnum/GL/Extensions.h"
#include "Magnum/GL/PixelFormat.h"
#include "Magnum/GL/TextureFormat.h"
#ifndef MAGNUM_TARGET_WEBGL
#include "Magnum/GL/Implementation/DebugState.h"
#endif
#include "Magnum/GL/Implementation/RendererState.h"
#include "Magnum/GL/Implementation/State.h"
#include "Magnum/GL/Implementation/TextureState.h"
#include "Magnum/Math/Color.h"
#include "Magnum/Math/Range.h"
namespace Magnum { namespace GL {
#ifndef MAGNUM_TARGET_GLES2
Float AbstractTexture::maxLodBias() {
GLfloat& value = Context::current().state().texture->maxLodBias;
/* Get the value, if not already cached */
if(value == 0.0f)
glGetFloatv(GL_MAX_TEXTURE_LOD_BIAS, &value);
return value;
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
Int AbstractTexture::maxColorSamples() {
#ifndef MAGNUM_TARGET_GLES
if(!Context::current().isExtensionSupported<Extensions::ARB::texture_multisample>())
#else
if(!Context::current().isVersionSupported(Version::GLES310))
#endif
return 0;
GLint& value = Context::current().state().texture->maxColorSamples;
/* Get the value, if not already cached */
if(value == 0)
glGetIntegerv(GL_MAX_COLOR_TEXTURE_SAMPLES, &value);
return value;
}
Int AbstractTexture::maxDepthSamples() {
#ifndef MAGNUM_TARGET_GLES
if(!Context::current().isExtensionSupported<Extensions::ARB::texture_multisample>())
#else
if(!Context::current().isVersionSupported(Version::GLES310))
#endif
return 0;
GLint& value = Context::current().state().texture->maxDepthSamples;
/* Get the value, if not already cached */
if(value == 0)
glGetIntegerv(GL_MAX_DEPTH_TEXTURE_SAMPLES, &value);
return value;
}
Int AbstractTexture::maxIntegerSamples() {
#ifndef MAGNUM_TARGET_GLES
if(!Context::current().isExtensionSupported<Extensions::ARB::texture_multisample>())
#else
if(!Context::current().isVersionSupported(Version::GLES310))
#endif
return 0;
GLint& value = Context::current().state().texture->maxIntegerSamples;
/* Get the value, if not already cached */
if(value == 0)
glGetIntegerv(GL_MAX_INTEGER_SAMPLES, &value);
return value;
}
#endif
void AbstractTexture::unbind(const Int textureUnit) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* If given texture unit is already unbound, nothing to do */
if(textureState.bindings[textureUnit].second == 0) return;
/* Unbind the texture, reset state tracker */
Context::current().state().texture->unbindImplementation(textureUnit);
/* libstdc++ since GCC 6.3 can't handle just = {} (ambiguous overload of
operator=) */
textureState.bindings[textureUnit] = std::pair<GLenum, GLuint>{};
}
void AbstractTexture::unbindImplementationDefault(const GLint textureUnit) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* Activate given texture unit if not already active, update state tracker */
if(textureState.currentTextureUnit != textureUnit)
glActiveTexture(GL_TEXTURE0 + (textureState.currentTextureUnit = textureUnit));
CORRADE_INTERNAL_ASSERT(textureState.bindings[textureUnit].first != 0);
glBindTexture(textureState.bindings[textureUnit].first, 0);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::unbindImplementationMulti(const GLint textureUnit) {
constexpr static GLuint zero = 0;
glBindTextures(textureUnit, 1, &zero);
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::unbindImplementationDSA(const GLint textureUnit) {
CORRADE_INTERNAL_ASSERT(Context::current().state().texture->bindings[textureUnit].first != 0);
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glBindTextureUnit(textureUnit, 0);
}
#endif
void AbstractTexture::unbind(const Int firstTextureUnit, const std::size_t count) {
/* State tracker is updated in the implementations */
Context::current().state().texture->bindMultiImplementation(firstTextureUnit, {nullptr, count});
}
/** @todoc const std::initializer_list makes Doxygen grumpy */
void AbstractTexture::bind(const Int firstTextureUnit, std::initializer_list<AbstractTexture*> textures) {
/* State tracker is updated in the implementations */
Context::current().state().texture->bindMultiImplementation(firstTextureUnit, {textures.begin(), textures.size()});
}
void AbstractTexture::bindImplementationFallback(const GLint firstTextureUnit, const Containers::ArrayView<AbstractTexture* const> textures) {
for(std::size_t i = 0; i != textures.size(); ++i)
textures && textures[i] ? textures[i]->bind(firstTextureUnit + i) : unbind(firstTextureUnit + i);
}
#ifndef MAGNUM_TARGET_GLES
/** @todoc const Containers::ArrayView makes Doxygen grumpy */
void AbstractTexture::bindImplementationMulti(const GLint firstTextureUnit, Containers::ArrayView<AbstractTexture* const> textures) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* Create array of IDs and also update bindings in state tracker */
/** @todo VLAs */
Containers::Array<GLuint> ids{textures ? textures.size() : 0};
bool different = false;
for(std::size_t i = 0; i != textures.size(); ++i) {
const GLuint id = textures && textures[i] ? textures[i]->_id : 0;
if(textures) {
if(textures[i]) textures[i]->createIfNotAlready();
ids[i] = id;
}
if(textureState.bindings[firstTextureUnit + i].second != id) {
different = true;
textureState.bindings[firstTextureUnit + i].second = id;
}
}
/* Avoid doing the binding if there is nothing different */
if(different) glBindTextures(firstTextureUnit, textures.size(), ids);
}
#endif
#ifndef MAGNUM_TARGET_GLES
Int AbstractTexture::compressedBlockDataSize(const GLenum target, const TextureFormat format) {
return (Context::current().state().texture->compressedBlockDataSizeImplementation)(target, format);
}
Int AbstractTexture::compressedBlockDataSizeImplementationDefault(const GLenum target, const TextureFormat format) {
GLint value;
glGetInternalformativ(target, GLenum(format), GL_TEXTURE_COMPRESSED_BLOCK_SIZE, 1, &value);
return value;
}
Int AbstractTexture::compressedBlockDataSizeImplementationBitsWorkaround(const GLenum target, const TextureFormat format) {
/* NVidia (358.16) reports the value in bits instead of bytes */
return compressedBlockDataSizeImplementationDefault(target, format)/8;
}
#endif
AbstractTexture::AbstractTexture(GLenum target): _target{target}, _flags{ObjectFlag::DeleteOnDestruction} {
(this->*Context::current().state().texture->createImplementation)();
CORRADE_INTERNAL_ASSERT(_id != Implementation::State::DisengagedBinding);
}
void AbstractTexture::createImplementationDefault() {
glGenTextures(1, &_id);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::createImplementationDSA() {
glCreateTextures(_target, 1, &_id);
_flags |= ObjectFlag::Created;
}
#endif
AbstractTexture::~AbstractTexture() {
/* Moved out or not deleting on destruction, nothing to do */
if(!_id || !(_flags & ObjectFlag::DeleteOnDestruction)) return;
/* Remove all bindings */
for(auto& binding: Context::current().state().texture->bindings) {
/* MSVC 2015 needs the parentheses around */
/* libstdc++ since GCC 6.3 can't handle just = {} (ambiguous overload
of operator=) */
if(binding.second == _id) binding = std::pair<GLenum, GLuint>{};
}
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
/* Remove all image bindings */
for(auto& binding: Context::current().state().texture->imageBindings) {
/* MSVC 2015 needs the parentheses around */
if(std::get<0>(binding) == _id) binding = {};
}
#endif
glDeleteTextures(1, &_id);
}
void AbstractTexture::createIfNotAlready() {
if(_flags & ObjectFlag::Created) return;
/* glGen*() does not create the object, just reserves the name. Some
commands (such as glBindTextures() or glObjectLabel()) operate with IDs
directly and they require the object to be created. Binding the texture
to desired target finally creates it. Also all EXT DSA functions
implicitly create it. */
bindInternal();
CORRADE_INTERNAL_ASSERT(_flags & ObjectFlag::Created);
}
#ifndef MAGNUM_TARGET_WEBGL
std::string AbstractTexture::label() {
createIfNotAlready();
return Context::current().state().debug->getLabelImplementation(GL_TEXTURE, _id);
}
AbstractTexture& AbstractTexture::setLabelInternal(const Containers::ArrayView<const char> label) {
createIfNotAlready();
Context::current().state().debug->labelImplementation(GL_TEXTURE, _id, label);
return *this;
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
void AbstractTexture::unbindImage(const Int imageUnit) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* If already unbound in given image unit, nothing to do */
if(std::get<0>(textureState.imageBindings[imageUnit]) == 0) return;
/* Update state tracker, bind the texture to the unit */
std::get<0>(textureState.imageBindings[imageUnit]) = 0;
glBindImageTexture(imageUnit, 0, 0, false, 0, GL_READ_ONLY, GL_RGBA8);
}
#ifndef MAGNUM_TARGET_GLES
/** @todoc const Containers::ArrayView makes Doxygen grumpy */
void AbstractTexture::bindImagesInternal(const Int firstImageUnit, Containers::ArrayView<AbstractTexture* const> textures) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* Create array of IDs and also update bindings in state tracker */
Containers::Array<GLuint> ids{textures ? textures.size() : 0};
bool different = false;
for(std::size_t i = 0; i != textures.size(); ++i) {
const std::tuple<GLuint, GLint, GLboolean, GLint, GLenum> state = textures && textures[i] ?
std::tuple<GLuint, GLint, GLboolean, GLint, GLenum>(textures[i]->_id, 0, true, 0, GL_READ_WRITE) :
std::tuple<GLuint, GLint, GLboolean, GLint, GLenum>(0, 0, false, 0, GL_READ_ONLY);
if(textures) {
if(textures[i]) {
textures[i]->createIfNotAlready();
}
ids[i] = std::get<0>(state);
}
if(textureState.imageBindings[firstImageUnit + i] != state) {
different = true;
textureState.imageBindings[firstImageUnit + i] = state;
}
}
/* Avoid doing the binding if there is nothing different */
if(different) glBindImageTextures(firstImageUnit, textures.size(), ids);
}
#endif
void AbstractTexture::bindImageInternal(const Int imageUnit, const Int level, const bool layered, const Int layer, const ImageAccess access, const ImageFormat format) {
Implementation::TextureState& textureState = *Context::current().state().texture;
const std::tuple<GLuint, GLint, GLboolean, GLint, GLenum> state{_id, level, layered, layer, GLenum(access)};
/* If already bound in given texture unit, nothing to do */
if(textureState.imageBindings[imageUnit] == state) return;
/* Update state tracker, bind the texture to the unit */
textureState.imageBindings[imageUnit] = state;
glBindImageTexture(imageUnit, _id, level, layered, layer, GLenum(access), GLenum(format));
}
#endif
void AbstractTexture::bind(Int textureUnit) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* If already bound in given texture unit, nothing to do */
if(textureState.bindings[textureUnit].second == _id) return;
/* Update state tracker, bind the texture to the unit */
textureState.bindings[textureUnit] = {_target, _id};
(this->*textureState.bindImplementation)(textureUnit);
}
void AbstractTexture::bindImplementationDefault(GLint textureUnit) {
Implementation::TextureState& textureState = *Context::current().state().texture;
/* Activate given texture unit if not already active, update state tracker */
if(textureState.currentTextureUnit != textureUnit)
glActiveTexture(GL_TEXTURE0 + (textureState.currentTextureUnit = textureUnit));
/* Binding the texture finally creates it */
_flags |= ObjectFlag::Created;
glBindTexture(_target, _id);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::bindImplementationMulti(GLint textureUnit) {
createIfNotAlready();
glBindTextures(textureUnit, 1, &_id);
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::bindImplementationDSA(const GLint textureUnit) {
glBindTextureUnit(textureUnit, _id);
}
#ifdef CORRADE_TARGET_WINDOWS
void AbstractTexture::bindImplementationDSAIntelWindows(const GLint textureUnit) {
/* See the "intel-windows-half-baked-dsa-texture-bind" workaround */
if(_target == GL_TEXTURE_CUBE_MAP) bindImplementationDefault(textureUnit);
else bindImplementationDSA(textureUnit);
}
#endif
#endif
#ifndef MAGNUM_TARGET_GLES2
void AbstractTexture::setBaseLevel(Int level) {
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_BASE_LEVEL, level);
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::setMaxLevel(Int level) {
(this->*Context::current().state().texture->parameteriImplementation)(
#ifndef MAGNUM_TARGET_GLES2
GL_TEXTURE_MAX_LEVEL
#else
GL_TEXTURE_MAX_LEVEL_APPLE
#endif
, level);
}
#endif
void AbstractTexture::setMinificationFilter(SamplerFilter filter, SamplerMipmap mipmap) {
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_MIN_FILTER, GLint(filter)|GLint(mipmap));
}
void AbstractTexture::setMagnificationFilter(const SamplerFilter filter) {
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_MAG_FILTER, GLint(filter));
}
#ifndef MAGNUM_TARGET_GLES2
void AbstractTexture::setMinLod(const Float lod) {
(this->*Context::current().state().texture->parameterfImplementation)(GL_TEXTURE_MIN_LOD, lod);
}
void AbstractTexture::setMaxLod(const Float lod) {
(this->*Context::current().state().texture->parameterfImplementation)(GL_TEXTURE_MAX_LOD, lod);
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::setLodBias(const Float bias) {
(this->*Context::current().state().texture->parameterfImplementation)(GL_TEXTURE_LOD_BIAS, bias);
}
#endif
#ifndef MAGNUM_TARGET_WEBGL
void AbstractTexture::setBorderColor(const Color4& color) {
(this->*Context::current().state().texture->parameterfvImplementation)(
#ifndef MAGNUM_TARGET_GLES2
GL_TEXTURE_BORDER_COLOR,
#else
GL_TEXTURE_BORDER_COLOR_EXT,
#endif
color.data());
}
#ifndef MAGNUM_TARGET_GLES2
void AbstractTexture::setBorderColor(const Vector4ui& color) {
(this->*Context::current().state().texture->parameterIuivImplementation)(GL_TEXTURE_BORDER_COLOR, color.data());
}
void AbstractTexture::setBorderColor(const Vector4i& color) {
(this->*Context::current().state().texture->parameterIivImplementation)(GL_TEXTURE_BORDER_COLOR, color.data());
}
#endif
#endif
void AbstractTexture::setMaxAnisotropy(const Float anisotropy) {
(this->*Context::current().state().texture->setMaxAnisotropyImplementation)(anisotropy);
}
#ifndef MAGNUM_TARGET_WEBGL
void AbstractTexture::setSrgbDecode(bool decode) {
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_SRGB_DECODE_EXT,
decode ? GL_DECODE_EXT : GL_SKIP_DECODE_EXT);
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
void AbstractTexture::setSwizzleInternal(const GLint r, const GLint g, const GLint b, const GLint a) {
#ifndef MAGNUM_TARGET_GLES
const GLint rgba[] = {r, g, b, a};
(this->*Context::current().state().texture->parameterivImplementation)(GL_TEXTURE_SWIZZLE_RGBA, rgba);
#else
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_SWIZZLE_R, r);
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_SWIZZLE_G, g);
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_SWIZZLE_B, b);
(this->*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_SWIZZLE_A, a);
#endif
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::setCompareMode(const SamplerCompareMode mode) {
(this->*Context::current().state().texture->parameteriImplementation)(
#ifndef MAGNUM_TARGET_GLES2
GL_TEXTURE_COMPARE_MODE
#else
GL_TEXTURE_COMPARE_MODE_EXT
#endif
, GLenum(mode));
}
void AbstractTexture::setCompareFunction(const SamplerCompareFunction function) {
(this->*Context::current().state().texture->parameteriImplementation)(
#ifndef MAGNUM_TARGET_GLES2
GL_TEXTURE_COMPARE_FUNC
#else
GL_TEXTURE_COMPARE_FUNC_EXT
#endif
, GLenum(function));
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
void AbstractTexture::setDepthStencilMode(const SamplerDepthStencilMode mode) {
(this->*Context::current().state().texture->parameteriImplementation)(GL_DEPTH_STENCIL_TEXTURE_MODE, GLenum(mode));
}
#endif
void AbstractTexture::invalidateImage(const Int level) {
(this->*Context::current().state().texture->invalidateImageImplementation)(level);
}
void AbstractTexture::generateMipmap() {
(this->*Context::current().state().texture->mipmapImplementation)();
}
void AbstractTexture::mipmapImplementationDefault() {
bindInternal();
glGenerateMipmap(_target);
}
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::mipmapImplementationDSA() {
glGenerateTextureMipmap(_id);
}
#endif
void AbstractTexture::bindInternal() {
/* Using glBindTextures() here is meaningless, because the non-DSA
functions need to have the texture bound in *currently active* unit,
so we would need to call glActiveTexture() afterwards anyway. */
Implementation::TextureState& textureState = *Context::current().state().texture;
/* If the texture is already bound in current unit, nothing to do */
if(textureState.bindings[textureState.currentTextureUnit].second == _id)
return;
/* Set internal unit as active if not already, update state tracker */
CORRADE_INTERNAL_ASSERT(textureState.maxTextureUnits > 1);
const GLint internalTextureUnit = textureState.maxTextureUnits-1;
if(textureState.currentTextureUnit != internalTextureUnit)
glActiveTexture(GL_TEXTURE0 + (textureState.currentTextureUnit = internalTextureUnit));
/* Bind the texture to internal unit if not already, update state tracker */
if(textureState.bindings[internalTextureUnit].second == _id) return;
textureState.bindings[internalTextureUnit] = {_target, _id};
/* Binding the texture finally creates it */
_flags |= ObjectFlag::Created;
glBindTexture(_target, _id);
}
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
namespace {
PixelFormat pixelFormatForInternalFormat(const TextureFormat internalFormat) {
switch(internalFormat) {
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::Red:
case TextureFormat::R8:
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R8Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::R16:
case TextureFormat::R16Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R16F:
case TextureFormat::R32F:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedRed:
case TextureFormat::CompressedRedRgtc1:
case TextureFormat::CompressedSignedRedRgtc1:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::CompressedR11Eac:
case TextureFormat::CompressedSignedR11Eac:
#endif
return PixelFormat::Red;
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R8UI:
case TextureFormat::R8I:
case TextureFormat::R16UI:
case TextureFormat::R16I:
case TextureFormat::R32UI:
case TextureFormat::R32I:
return PixelFormat::RedInteger;
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::RG:
case TextureFormat::RG8:
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RG8Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RG16:
case TextureFormat::RG16Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RG16F:
case TextureFormat::RG32F:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedRG:
case TextureFormat::CompressedRGRgtc2:
case TextureFormat::CompressedSignedRGRgtc2:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::CompressedRG11Eac:
case TextureFormat::CompressedSignedRG11Eac:
#endif
return PixelFormat::RG;
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RG8UI:
case TextureFormat::RG8I:
case TextureFormat::RG16UI:
case TextureFormat::RG16I:
case TextureFormat::RG32UI:
case TextureFormat::RG32I:
return PixelFormat::RGInteger;
#endif
case TextureFormat::RGB:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::RGB8:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGB8Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGB16:
case TextureFormat::RGB16Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGB16F:
case TextureFormat::RGB32F:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::R3B3G2:
case TextureFormat::RGB4:
case TextureFormat::RGB5:
#endif
case TextureFormat::RGB565:
#if !defined(MAGNUM_TARGET_GLES) || (defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL))
case TextureFormat::RGB10:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGB12:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R11FG11FB10F:
case TextureFormat::RGB9E5:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedRGB:
case TextureFormat::CompressedRGBBptcUnsignedFloat:
case TextureFormat::CompressedRGBBptcSignedFloat:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::CompressedRGB8Etc2:
case TextureFormat::CompressedSRGB8Etc2:
#endif
case TextureFormat::CompressedRGBS3tcDxt1:
return PixelFormat::RGB;
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
case TextureFormat::SRGB:
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::SRGB8:
#endif
#ifndef MAGNUM_TARGET_GLES2
return PixelFormat::RGB;
#else
return PixelFormat::SRGB;
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGB8UI:
case TextureFormat::RGB8I:
case TextureFormat::RGB16UI:
case TextureFormat::RGB16I:
case TextureFormat::RGB32UI:
case TextureFormat::RGB32I:
return PixelFormat::RGBInteger;
#endif
case TextureFormat::RGBA:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::RGBA8:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGBA8Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGBA16:
case TextureFormat::RGBA16Snorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGBA16F:
case TextureFormat::RGBA32F:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGBA2:
#endif
case TextureFormat::RGBA4:
case TextureFormat::RGB5A1:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::RGB10A2:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGBA12:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedRGBA:
case TextureFormat::CompressedRGBABptcUnorm:
case TextureFormat::CompressedSRGBAlphaBptcUnorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::CompressedRGB8PunchthroughAlpha1Etc2:
case TextureFormat::CompressedSRGB8PunchthroughAlpha1Etc2:
case TextureFormat::CompressedRGBA8Etc2Eac:
case TextureFormat::CompressedSRGB8Alpha8Etc2Eac:
#endif
case TextureFormat::CompressedRGBAS3tcDxt1:
case TextureFormat::CompressedRGBAS3tcDxt3:
case TextureFormat::CompressedRGBAS3tcDxt5:
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::CompressedRGBAAstc4x4:
case TextureFormat::CompressedSRGB8Alpha8Astc4x4:
case TextureFormat::CompressedRGBAAstc5x4:
case TextureFormat::CompressedSRGB8Alpha8Astc5x4:
case TextureFormat::CompressedRGBAAstc5x5:
case TextureFormat::CompressedSRGB8Alpha8Astc5x5:
case TextureFormat::CompressedRGBAAstc6x5:
case TextureFormat::CompressedSRGB8Alpha8Astc6x5:
case TextureFormat::CompressedRGBAAstc6x6:
case TextureFormat::CompressedSRGB8Alpha8Astc6x6:
case TextureFormat::CompressedRGBAAstc8x5:
case TextureFormat::CompressedSRGB8Alpha8Astc8x5:
case TextureFormat::CompressedRGBAAstc8x6:
case TextureFormat::CompressedSRGB8Alpha8Astc8x6:
case TextureFormat::CompressedRGBAAstc8x8:
case TextureFormat::CompressedSRGB8Alpha8Astc8x8:
case TextureFormat::CompressedRGBAAstc10x5:
case TextureFormat::CompressedSRGB8Alpha8Astc10x5:
case TextureFormat::CompressedRGBAAstc10x6:
case TextureFormat::CompressedSRGB8Alpha8Astc10x6:
case TextureFormat::CompressedRGBAAstc10x8:
case TextureFormat::CompressedSRGB8Alpha8Astc10x8:
case TextureFormat::CompressedRGBAAstc10x10:
case TextureFormat::CompressedSRGB8Alpha8Astc10x10:
case TextureFormat::CompressedRGBAAstc12x10:
case TextureFormat::CompressedSRGB8Alpha8Astc12x10:
case TextureFormat::CompressedRGBAAstc12x12:
case TextureFormat::CompressedSRGB8Alpha8Astc12x12:
#endif
return PixelFormat::RGBA;
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
case TextureFormat::SRGBAlpha:
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::SRGB8Alpha8:
#endif
#ifndef MAGNUM_TARGET_GLES2
return PixelFormat::RGBA;
#else
return PixelFormat::SRGBAlpha;
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGBA8UI:
case TextureFormat::RGBA8I:
case TextureFormat::RGBA16UI:
case TextureFormat::RGBA16I:
case TextureFormat::RGBA32UI:
case TextureFormat::RGBA32I:
case TextureFormat::RGB10A2UI:
return PixelFormat::RGBAInteger;
#endif
#ifdef MAGNUM_TARGET_GLES2
case TextureFormat::Luminance:
return PixelFormat::Luminance;
case TextureFormat::LuminanceAlpha:
return PixelFormat::LuminanceAlpha;
#endif
case TextureFormat::DepthComponent:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::DepthComponent16:
case TextureFormat::DepthComponent24:
#endif
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::DepthComponent32:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::DepthComponent32F:
#endif
return PixelFormat::DepthComponent;
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::StencilIndex8:
return PixelFormat::StencilIndex;
#endif
case TextureFormat::DepthStencil:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::Depth24Stencil8:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::Depth32FStencil8:
#endif
return PixelFormat::DepthStencil;
}
CORRADE_ASSERT_UNREACHABLE(); /* LCOV_EXCL_LINE */
}
PixelType pixelTypeForInternalFormat(const TextureFormat internalFormat) {
switch(internalFormat) {
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::Red:
case TextureFormat::RG:
#endif
case TextureFormat::RGB:
case TextureFormat::RGBA:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::R8:
case TextureFormat::RG8:
case TextureFormat::RGB8:
case TextureFormat::RGBA8:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R8UI:
case TextureFormat::RG8UI:
case TextureFormat::RGB8UI:
case TextureFormat::RGBA8UI:
#endif
#ifdef MAGNUM_TARGET_GLES2
case TextureFormat::Luminance:
case TextureFormat::LuminanceAlpha:
#endif
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
case TextureFormat::SRGB:
case TextureFormat::SRGBAlpha:
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::SRGB8:
case TextureFormat::SRGB8Alpha8:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGBA2: /**< @todo really? */
case TextureFormat::CompressedRed:
case TextureFormat::CompressedRG:
case TextureFormat::CompressedRGB:
case TextureFormat::CompressedRGBA:
case TextureFormat::CompressedRedRgtc1:
case TextureFormat::CompressedRGRgtc2:
case TextureFormat::CompressedRGBABptcUnorm:
case TextureFormat::CompressedSRGBAlphaBptcUnorm:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::CompressedRGB8Etc2:
case TextureFormat::CompressedSRGB8Etc2:
case TextureFormat::CompressedRGB8PunchthroughAlpha1Etc2:
case TextureFormat::CompressedSRGB8PunchthroughAlpha1Etc2:
case TextureFormat::CompressedRGBA8Etc2Eac:
case TextureFormat::CompressedSRGB8Alpha8Etc2Eac:
case TextureFormat::CompressedR11Eac:
case TextureFormat::CompressedSignedR11Eac:
case TextureFormat::CompressedRG11Eac:
case TextureFormat::CompressedSignedRG11Eac:
#endif
case TextureFormat::CompressedRGBS3tcDxt1:
case TextureFormat::CompressedRGBAS3tcDxt1:
case TextureFormat::CompressedRGBAS3tcDxt3:
case TextureFormat::CompressedRGBAS3tcDxt5:
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::CompressedRGBAAstc4x4:
case TextureFormat::CompressedSRGB8Alpha8Astc4x4:
case TextureFormat::CompressedRGBAAstc5x4:
case TextureFormat::CompressedSRGB8Alpha8Astc5x4:
case TextureFormat::CompressedRGBAAstc5x5:
case TextureFormat::CompressedSRGB8Alpha8Astc5x5:
case TextureFormat::CompressedRGBAAstc6x5:
case TextureFormat::CompressedSRGB8Alpha8Astc6x5:
case TextureFormat::CompressedRGBAAstc6x6:
case TextureFormat::CompressedSRGB8Alpha8Astc6x6:
case TextureFormat::CompressedRGBAAstc8x5:
case TextureFormat::CompressedSRGB8Alpha8Astc8x5:
case TextureFormat::CompressedRGBAAstc8x6:
case TextureFormat::CompressedSRGB8Alpha8Astc8x6:
case TextureFormat::CompressedRGBAAstc8x8:
case TextureFormat::CompressedSRGB8Alpha8Astc8x8:
case TextureFormat::CompressedRGBAAstc10x5:
case TextureFormat::CompressedSRGB8Alpha8Astc10x5:
case TextureFormat::CompressedRGBAAstc10x6:
case TextureFormat::CompressedSRGB8Alpha8Astc10x6:
case TextureFormat::CompressedRGBAAstc10x8:
case TextureFormat::CompressedSRGB8Alpha8Astc10x8:
case TextureFormat::CompressedRGBAAstc10x10:
case TextureFormat::CompressedSRGB8Alpha8Astc10x10:
case TextureFormat::CompressedRGBAAstc12x10:
case TextureFormat::CompressedSRGB8Alpha8Astc12x10:
case TextureFormat::CompressedRGBAAstc12x12:
case TextureFormat::CompressedSRGB8Alpha8Astc12x12:
#endif
return PixelType::UnsignedByte;
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R8Snorm:
case TextureFormat::RG8Snorm:
case TextureFormat::RGB8Snorm:
case TextureFormat::RGBA8Snorm:
case TextureFormat::R8I:
case TextureFormat::RG8I:
case TextureFormat::RGB8I:
case TextureFormat::RGBA8I:
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedSignedRedRgtc1:
case TextureFormat::CompressedSignedRGRgtc2:
#endif
return PixelType::Byte;
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::R16:
case TextureFormat::RG16:
case TextureFormat::RGB16:
case TextureFormat::RGBA16:
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R16UI:
case TextureFormat::RG16UI:
case TextureFormat::RGB16UI:
case TextureFormat::RGBA16UI:
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGB12: /**< @todo really? */
#endif
case TextureFormat::RGBA4: /**< @todo really? */
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGBA12: /**< @todo really? */
#endif
return PixelType::UnsignedShort;
#ifndef MAGNUM_TARGET_GLES2
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::R16Snorm:
case TextureFormat::RG16Snorm:
case TextureFormat::RGB16Snorm:
case TextureFormat::RGBA16Snorm:
#endif
case TextureFormat::R16I:
case TextureFormat::RG16I:
case TextureFormat::RGB16I:
case TextureFormat::RGBA16I:
return PixelType::Short;
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R16F:
case TextureFormat::RG16F:
case TextureFormat::RGB16F:
case TextureFormat::RGBA16F:
return PixelType::HalfFloat;
case TextureFormat::R32UI:
case TextureFormat::RG32UI:
case TextureFormat::RGB32UI:
case TextureFormat::RGBA32UI:
return PixelType::UnsignedInt;
case TextureFormat::R32I:
case TextureFormat::RG32I:
case TextureFormat::RGB32I:
case TextureFormat::RGBA32I:
return PixelType::Int;
case TextureFormat::R32F:
case TextureFormat::RG32F:
case TextureFormat::RGB32F:
case TextureFormat::RGBA32F:
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::CompressedRGBBptcUnsignedFloat:
case TextureFormat::CompressedRGBBptcSignedFloat:
#endif
return PixelType::Float;
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::R3B3G2:
return PixelType::UnsignedByte332;
case TextureFormat::RGB4:
return PixelType::UnsignedShort4444;
#endif
#ifndef MAGNUM_TARGET_GLES
case TextureFormat::RGB5:
#endif
case TextureFormat::RGB5A1:
return PixelType::UnsignedShort5551;
case TextureFormat::RGB565:
return PixelType::UnsignedShort565;
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
case TextureFormat::RGB10:
#endif
case TextureFormat::RGB10A2:
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::RGB10A2UI:
#endif
return PixelType::UnsignedInt2101010Rev; /**< @todo Rev for all? */
#endif
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::R11FG11FB10F:
return PixelType::UnsignedInt10F11F11FRev;
case TextureFormat::RGB9E5:
return PixelType::UnsignedInt5999Rev;
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::DepthComponent16:
return PixelType::UnsignedShort;
#endif
case TextureFormat::DepthComponent:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::DepthComponent24:
#endif
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::DepthComponent32:
#endif
return PixelType::UnsignedInt;
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::DepthComponent32F:
return PixelType::Float;
#endif
#ifndef MAGNUM_TARGET_WEBGL
case TextureFormat::StencilIndex8:
return PixelType::UnsignedByte;
#endif
case TextureFormat::DepthStencil:
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
case TextureFormat::Depth24Stencil8:
#endif
return PixelType::UnsignedInt248;
#ifndef MAGNUM_TARGET_GLES2
case TextureFormat::Depth32FStencil8:
return PixelType::Float32UnsignedInt248Rev;
#endif
}
CORRADE_ASSERT_UNREACHABLE(); /* LCOV_EXCL_LINE */
}
}
#endif
void AbstractTexture::parameterImplementationDefault(GLenum parameter, GLint value) {
bindInternal();
glTexParameteri(_target, parameter, value);
}
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterImplementationDSA(const GLenum parameter, const GLint value) {
glTextureParameteri(_id, parameter, value);
}
#endif
void AbstractTexture::parameterImplementationDefault(GLenum parameter, GLfloat value) {
bindInternal();
glTexParameterf(_target, parameter, value);
}
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterImplementationDSA(const GLenum parameter, const GLfloat value) {
glTextureParameterf(_id, parameter, value);
}
#endif
#ifndef MAGNUM_TARGET_GLES2
void AbstractTexture::parameterImplementationDefault(GLenum parameter, const GLint* values) {
bindInternal();
glTexParameteriv(_target, parameter, values);
}
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterImplementationDSA(const GLenum parameter, const GLint* const values) {
glTextureParameteriv(_id, parameter, values);
}
#endif
#endif
void AbstractTexture::parameterImplementationDefault(GLenum parameter, const GLfloat* values) {
bindInternal();
glTexParameterfv(_target, parameter, values);
}
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterImplementationDSA(const GLenum parameter, const GLfloat* const values) {
glTextureParameterfv(_id, parameter, values);
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
void AbstractTexture::parameterIImplementationDefault(GLenum parameter, const GLuint* values) {
bindInternal();
glTexParameterIuiv(_target, parameter, values);
}
#ifdef MAGNUM_TARGET_GLES
void AbstractTexture::parameterIImplementationEXT(GLenum parameter, const GLuint* values) {
bindInternal();
glTexParameterIuivEXT(_target, parameter, values);
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterIImplementationDSA(const GLenum parameter, const GLuint* const values) {
glTextureParameterIuiv(_id, parameter, values);
}
#endif
void AbstractTexture::parameterIImplementationDefault(GLenum parameter, const GLint* values) {
bindInternal();
glTexParameterIiv(_target, parameter, values);
}
#ifdef MAGNUM_TARGET_GLES
void AbstractTexture::parameterIImplementationEXT(GLenum parameter, const GLint* values) {
bindInternal();
glTexParameterIivEXT(_target, parameter, values);
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::parameterIImplementationDSA(const GLenum parameter, const GLint* const values) {
glTextureParameterIiv(_id, parameter, values);
}
#endif
#endif
void AbstractTexture::setMaxAnisotropyImplementationNoOp(GLfloat) {}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::setMaxAnisotropyImplementationArb(GLfloat anisotropy) {
(this->*Context::current().state().texture->parameterfImplementation)(GL_TEXTURE_MAX_ANISOTROPY, anisotropy);
}
#endif
void AbstractTexture::setMaxAnisotropyImplementationExt(GLfloat anisotropy) {
(this->*Context::current().state().texture->parameterfImplementation)(GL_TEXTURE_MAX_ANISOTROPY_EXT, anisotropy);
}
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
void AbstractTexture::getLevelParameterImplementationDefault(const GLint level, const GLenum parameter, GLint* const values) {
bindInternal();
glGetTexLevelParameteriv(_target, level, parameter, values);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::getLevelParameterImplementationDSA(const GLint level, const GLenum parameter, GLint* const values) {
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glGetTextureLevelParameteriv(_id, level, parameter, values);
}
#endif
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationFallback(const GLsizei levels, const TextureFormat internalFormat, const Math::Vector<1, GLsizei>& size) {
const PixelFormat format = pixelFormatForInternalFormat(internalFormat);
const PixelType type = pixelTypeForInternalFormat(internalFormat);
for(GLsizei level = 0; level != levels; ++level)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
DataHelper<1>::setImage(*this, level, internalFormat,
ImageView1D{format, type, Math::max(Math::Vector<1, GLsizei>(1), size >> level)});
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDefault(GLsizei levels, TextureFormat internalFormat, const Math::Vector<1, GLsizei>& size) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexStorage1D(_target, levels, GLenum(internalFormat), size[0]);
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDSA(const GLsizei levels, const TextureFormat internalFormat, const Math::Vector<1, GLsizei>& size) {
glTextureStorage1D(_id, levels, GLenum(internalFormat), size[0]);
}
#endif
#if !defined(MAGNUM_TARGET_GLES) || defined(MAGNUM_TARGET_GLES2)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationFallback(const GLsizei levels, const TextureFormat internalFormat, const Vector2i& size) {
const PixelFormat format = pixelFormatForInternalFormat(internalFormat);
const PixelType type = pixelTypeForInternalFormat(internalFormat);
/* If EXT_texture_storage is not available on ES2, passing e.g.
TextureFormat::RGBA8 would cause an error. On ES2 it's required to have
internalFormat equal to format, so we do exactly that. */
#ifndef MAGNUM_TARGET_GLES2
const TextureFormat finalInternalFormat = internalFormat;
#else
const TextureFormat finalInternalFormat = TextureFormat(GLenum(format));
#endif
/* Common code for classic types */
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
if(_target == GL_TEXTURE_2D || _target == GL_TEXTURE_RECTANGLE)
#else
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
if(_target == GL_TEXTURE_2D)
#endif
{
for(GLsizei level = 0; level != levels; ++level)
DataHelper<2>::setImage(*this, level, finalInternalFormat,
ImageView2D{format, type, Math::max(Vector2i(1), size >> level)});
/* Cube map additionally needs to specify all faces */
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
} else if(_target == GL_TEXTURE_CUBE_MAP) {
for(GLsizei level = 0; level != levels; ++level) {
for(GLenum face: {GL_TEXTURE_CUBE_MAP_POSITIVE_X,
GL_TEXTURE_CUBE_MAP_NEGATIVE_X,
GL_TEXTURE_CUBE_MAP_POSITIVE_Y,
GL_TEXTURE_CUBE_MAP_NEGATIVE_Y,
GL_TEXTURE_CUBE_MAP_POSITIVE_Z,
GL_TEXTURE_CUBE_MAP_NEGATIVE_Z})
DataHelper<2>::setImage(*this, face, level, finalInternalFormat,
ImageView2D{format, type, Math::max(Vector2i(1), size >> level)});
}
#ifndef MAGNUM_TARGET_GLES
/* Array texture is not scaled in "layer" dimension */
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
} else if(_target == GL_TEXTURE_1D_ARRAY) {
for(GLsizei level = 0; level != levels; ++level)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
DataHelper<2>::setImage(*this, level, internalFormat,
ImageView2D{format, type, Vector2i{Math::max(1, size.x() >> level), size.y()}});
#endif
/* No other targets are available */
} else CORRADE_ASSERT_UNREACHABLE(); /* LCOV_EXCL_LINE */
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDefault(GLsizei levels, TextureFormat internalFormat, const Vector2i& size) {
bindInternal();
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexStorage2D(_target, levels, GLenum(internalFormat), size.x(), size.y());
#else
glTexStorage2DEXT(_target, levels, GLenum(internalFormat), size.x(), size.y());
#endif
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDSA(const GLsizei levels, const TextureFormat internalFormat, const Vector2i& size) {
glTextureStorage2D(_id, levels, GLenum(internalFormat), size.x(), size.y());
}
#endif
#if !defined(MAGNUM_TARGET_GLES) || (defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL))
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationFallback(GLsizei levels, TextureFormat internalFormat, const Vector3i& size) {
const PixelFormat format = pixelFormatForInternalFormat(internalFormat);
const PixelType type = pixelTypeForInternalFormat(internalFormat);
/* If EXT_texture_storage is not available on ES2, passing e.g.
TextureFormat::RGBA8 would cause an error. On ES2 it's required to have
internalFormat equal to format, so we do exactly that. */
#ifndef MAGNUM_TARGET_GLES2
const TextureFormat finalInternalFormat = internalFormat;
#else
const TextureFormat finalInternalFormat = TextureFormat(GLenum(format));
#endif
/* Common code for classic type */
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
if(_target == GL_TEXTURE_3D)
#else
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
if(_target == GL_TEXTURE_3D_OES)
#endif
for(GLsizei level = 0; level != levels; ++level)
DataHelper<3>::setImage(*this, level, finalInternalFormat,
ImageView3D{format, type, Math::max(Vector3i(1), size >> level)});
#ifndef MAGNUM_TARGET_GLES2
/* Array texture is not scaled in "layer" dimension */
else if(_target == GL_TEXTURE_2D_ARRAY || _target ==
#ifndef MAGNUM_TARGET_GLES
GL_TEXTURE_CUBE_MAP_ARRAY
#else
GL_TEXTURE_CUBE_MAP_ARRAY_EXT
#endif
) {
for(GLsizei level = 0; level != levels; ++level)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
DataHelper<3>::setImage(*this, level, internalFormat,
ImageView3D{format, type, Vector3i{Math::max(Vector2i{1}, size.xy() >> level), size.z()}});
}
#endif
/* No other targets are available */
else CORRADE_ASSERT_UNREACHABLE(); /* LCOV_EXCL_LINE */
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDefault(GLsizei levels, TextureFormat internalFormat, const Vector3i& size) {
bindInternal();
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexStorage3D(_target, levels, GLenum(internalFormat), size.x(), size.y(), size.z());
#else
glTexStorage3DEXT(_target, levels, GLenum(internalFormat), size.x(), size.y(), size.z());
#endif
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageImplementationDSA(const GLsizei levels, const TextureFormat internalFormat, const Vector3i& size) {
glTextureStorage3D(_id, levels, GLenum(internalFormat), size.x(), size.y(), size.z());
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageMultisampleImplementationFallback(const GLsizei samples, const TextureFormat internalFormat, const Vector2i& size, const GLboolean fixedSampleLocations) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexImage2DMultisample(_target, samples, GLenum(internalFormat), size.x(), size.y(), fixedSampleLocations);
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageMultisampleImplementationDefault(const GLsizei samples, const TextureFormat internalFormat, const Vector2i& size, const GLboolean fixedSampleLocations) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexStorage2DMultisample(_target, samples, GLenum(internalFormat), size.x(), size.y(), fixedSampleLocations);
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageMultisampleImplementationDSA(const GLsizei samples, const TextureFormat internalFormat, const Vector2i& size, const GLboolean fixedSampleLocations) {
glTextureStorage2DMultisample(_id, samples, GLenum(internalFormat), size.x(), size.y(), fixedSampleLocations);
}
void AbstractTexture::storageMultisampleImplementationFallback(const GLsizei samples, const TextureFormat internalFormat, const Vector3i& size, const GLboolean fixedSampleLocations) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexImage3DMultisample(_target, samples, GLenum(internalFormat), size.x(), size.y(), size.z(), fixedSampleLocations);
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageMultisampleImplementationDefault(const GLsizei samples, const TextureFormat internalFormat, const Vector3i& size, const GLboolean fixedSampleLocations) {
bindInternal();
glTexStorage3DMultisample(_target, samples, GLenum(internalFormat), size.x(), size.y(), size.z(), fixedSampleLocations);
}
#ifdef MAGNUM_TARGET_GLES
void AbstractTexture::storageMultisampleImplementationOES(const GLsizei samples, const TextureFormat internalFormat, const Vector3i& size, const GLboolean fixedSampleLocations) {
bindInternal();
glTexStorage3DMultisampleOES(_target, samples, GLenum(internalFormat), size.x(), size.y(), size.z(), fixedSampleLocations);
}
#endif
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::storageMultisampleImplementationDSA(const GLsizei samples, const TextureFormat internalFormat, const Vector3i& size, const GLboolean fixedSampleLocations) {
glTextureStorage3DMultisample(_id, samples, GLenum(internalFormat), size.x(), size.y(), size.z(), fixedSampleLocations);
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::getImageImplementationDefault(const GLint level, const PixelFormat format, const PixelType type, const std::size_t, GLvoid* const data) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glGetTexImage(_target, level, GLenum(format), GLenum(type), data);
}
void AbstractTexture::getCompressedImageImplementationDefault(const GLint level, std::size_t, GLvoid* const data) {
bindInternal();
glGetCompressedTexImage(_target, level, data);
}
void AbstractTexture::getImageImplementationDSA(const GLint level, const PixelFormat format, const PixelType type, const std::size_t dataSize, GLvoid* const data) {
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glGetTextureImage(_id, level, GLenum(format), GLenum(type), dataSize, data);
}
void AbstractTexture::getCompressedImageImplementationDSA(const GLint level, const std::size_t dataSize, GLvoid* const data) {
glGetCompressedTextureImage(_id, level, dataSize, data);
}
void AbstractTexture::getImageImplementationRobustness(const GLint level, const PixelFormat format, const PixelType type, const std::size_t dataSize, GLvoid* const data) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glGetnTexImageARB(_target, level, GLenum(format), GLenum(type), dataSize, data);
}
void AbstractTexture::getCompressedImageImplementationRobustness(const GLint level, const std::size_t dataSize, GLvoid* const data) {
bindInternal();
glGetnCompressedTexImageARB(_target, level, dataSize, data);
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::subImageImplementationDefault(GLint level, const Math::Vector<1, GLint>& offset, const Math::Vector<1, GLsizei>& size, PixelFormat format, PixelType type, const GLvoid* data) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexSubImage1D(_target, level, offset[0], size[0], GLenum(format), GLenum(type), data);
}
void AbstractTexture::compressedSubImageImplementationDefault(const GLint level, const Math::Vector<1, GLint>& offset, const Math::Vector<1, GLsizei>& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
bindInternal();
glCompressedTexSubImage1D(_target, level, offset[0], size[0], GLenum(format), dataSize, data);
}
void AbstractTexture::subImageImplementationDSA(const GLint level, const Math::Vector<1, GLint>& offset, const Math::Vector<1, GLsizei>& size, const PixelFormat format, const PixelType type, const GLvoid* const data) {
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTextureSubImage1D(_id, level, offset[0], size[0], GLenum(format), GLenum(type), data);
}
void AbstractTexture::compressedSubImageImplementationDSA(const GLint level, const Math::Vector<1, GLint>& offset, const Math::Vector<1, GLsizei>& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
glCompressedTextureSubImage1D(_id, level, offset[0], size[0], GLenum(format), dataSize, data);
}
#endif
void AbstractTexture::imageImplementationDefault(const GLenum target, const GLint level, const TextureFormat internalFormat, const Vector2i& size, const PixelFormat format, const PixelType type, const GLvoid* const data, const PixelStorage&) {
bindInternal();
glTexImage2D(target, level, GLint(internalFormat), size.x(), size.y(), 0, GLenum(format), GLenum(type), data);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::imageImplementationSvga3DSliceBySlice(const GLenum target, const GLint level, const TextureFormat internalFormat, const Vector2i& size, const PixelFormat format, const PixelType type, const GLvoid* const data, const PixelStorage& storage) {
/* Allocate and upload the first slice */
imageImplementationDefault(target, level, internalFormat, size, format, type, data, storage);
/* Upload the next slices slice by slice only if this is an array texture
with more than one slice and we are copying from user memory (not from a
buffer). The hard work is done by the subImage() implementation.
Moreover, I am simply calling the default implementation and not the DSA
one as just using glTexImage() pollutes the state already anyway so the
DSA cleanness is not worth it. */
/** @todo this will break when we support uploading from buffer offset (i.e. data != nullptr) */
if(target == GL_TEXTURE_1D_ARRAY && data && size.y() > 1)
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage2DImplementationDefault>(level, {0, 1}, {size.x(), size.y() - 1}, format, type, static_cast<const char*>(data) + std::get<1>(storage.dataProperties(pixelSize(format, type), {size, 1})).x(), storage);
}
#endif
void AbstractTexture::subImage2DImplementationDefault(GLint level, const Vector2i& offset, const Vector2i& size, PixelFormat format, PixelType type, const GLvoid* data, const PixelStorage&) {
bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexSubImage2D(_target, level, offset.x(), offset.y(), size.x(), size.y(), GLenum(format), GLenum(type), data);
}
/* Doxygen gets confused by the template parameter */
#if !defined(DOXYGEN_GENERATING_OUTPUT) && !defined(MAGNUM_TARGET_GLES)
template<void(AbstractTexture::*original)(GLint, const Vector2i&, const Vector2i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&)> void AbstractTexture::subImageImplementationSvga3DSliceBySlice(GLint level, const Vector2i& offset, const Vector2i& size, PixelFormat format, PixelType type, const GLvoid* const data, const PixelStorage& storage) {
/* Upload the data slice by slice only if this is an array texture and we
are copying from user memory (not from a buffer) */
if(_target == GL_TEXTURE_1D_ARRAY && data) {
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
const std::size_t stride = std::get<1>(storage.dataProperties(pixelSize(format, type), {size, 1})).x();
for(Int i = 0; i != size.y(); ++i)
(this->*original)(level, {offset.x(), offset.y() + i}, {size.x(), 1}, format, type, static_cast<const char*>(data) + stride*i, storage);
/* Otherwise just pass-though to the default implementation */
} else (this->*original)(level, offset, size, format, type, data, storage);
}
template void AbstractTexture::subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage2DImplementationDefault>(GLint, const Vector2i&, const Vector2i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&);
template void AbstractTexture::subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage2DImplementationDSA>(GLint, const Vector2i&, const Vector2i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&);
#endif
void AbstractTexture::compressedSubImageImplementationDefault(const GLint level, const Vector2i& offset, const Vector2i& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
bindInternal();
glCompressedTexSubImage2D(_target, level, offset.x(), offset.y(), size.x(), size.y(), GLenum(format), dataSize, data);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::subImage2DImplementationDSA(const GLint level, const Vector2i& offset, const Vector2i& size, const PixelFormat format, const PixelType type, const GLvoid* const data, const PixelStorage&) {
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTextureSubImage2D(_id, level, offset.x(), offset.y(), size.x(), size.y(), GLenum(format), GLenum(type), data);
}
void AbstractTexture::compressedSubImageImplementationDSA(const GLint level, const Vector2i& offset, const Vector2i& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
glCompressedTextureSubImage2D(_id, level, offset.x(), offset.y(), size.x(), size.y(), GLenum(format), dataSize, data);
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::imageImplementationDefault(GLint level, TextureFormat internalFormat, const Vector3i& size, PixelFormat format, PixelType type, const GLvoid* data, const PixelStorage&) {
bindInternal();
#ifndef MAGNUM_TARGET_GLES2
glTexImage3D
#else
glTexImage3DOES
#endif
(_target, level, GLint(internalFormat), size.x(), size.y(), size.z(), 0, GLenum(format), GLenum(type), data);
}
#ifndef MAGNUM_TARGET_WEBGL
void AbstractTexture::imageImplementationSvga3DSliceBySlice(GLint level, TextureFormat internalFormat, const Vector3i& size, PixelFormat format, PixelType type, const GLvoid* data, const PixelStorage& storage) {
/* Allocate and upload the first slice */
imageImplementationDefault(level, internalFormat, size, format, type, data, storage);
/* Upload the next slices slice by slice only if this is an array texture
with more than one slice or a 3D texture and we are copying from user
memory (not from a buffer). The hard work is done by the subImage()
implementation. Moreover, I am simply calling the default implementation
and not the DSA one as just using glTexImage() pollutes the state
already anyway so the DSA cleanness is not worth it. */
/** @todo this will break when we support uploading from buffer offset (i.e. data != nullptr) */
if((
#ifndef MAGNUM_TARGET_GLES2
_target == GL_TEXTURE_2D_ARRAY || _target == GL_TEXTURE_3D
#else
_target == GL_TEXTURE_3D_OES
#endif
) && data && size.z() > 1)
{
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage3DImplementationDefault>(level, {0, 0, 1}, {size.xy(), size.z() - 1}, format, type, static_cast<const char*>(data) + std::get<1>(storage.dataProperties(pixelSize(format, type), size)).xy().product(), storage);
}
}
#endif
void AbstractTexture::subImage3DImplementationDefault(GLint level, const Vector3i& offset, const Vector3i& size, PixelFormat format, PixelType type, const GLvoid* data, const PixelStorage&) {
bindInternal();
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexSubImage3D(_target, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), GLenum(type), data);
#else
glTexSubImage3DOES(_target, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), GLenum(type), data);
#endif
}
/* Doxygen gets confused by the template parameter */
#if !defined(DOXYGEN_GENERATING_OUTPUT) && !defined(MAGNUM_TARGET_WEBGL)
template<void(AbstractTexture::*original)(GLint, const Vector3i&, const Vector3i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&)> void AbstractTexture::subImageImplementationSvga3DSliceBySlice(GLint level, const Vector3i& offset, const Vector3i& size, PixelFormat format, PixelType type, const GLvoid* const data, const PixelStorage& storage) {
/* Upload the data slice by slice only if this is an array texture and we
are copying from user memory (not from a buffer) */
if(
#ifndef MAGNUM_TARGET_GLES2
_target == GL_TEXTURE_2D_ARRAY || _target == GL_TEXTURE_3D
#else
_target == GL_TEXTURE_3D_OES
#endif
)
{
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
const std::size_t stride = std::get<1>(storage.dataProperties(pixelSize(format, type), size)).xy().product();
for(Int i = 0; i != size.z(); ++i)
(this->*original)(level, {offset.xy(), offset.z() + i}, {size.xy(), 1}, format, type, static_cast<const char*>(data) + stride*i, storage);
/* Otherwise just pass-though to the default implementation */
} else (this->*original)(level, offset, size, format, type, data, storage);
}
template void AbstractTexture::subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage3DImplementationDefault>(GLint, const Vector3i&, const Vector3i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&);
#ifndef MAGNUM_TARGET_GLES
template void AbstractTexture::subImageImplementationSvga3DSliceBySlice<&AbstractTexture::subImage3DImplementationDSA>(GLint, const Vector3i&, const Vector3i&, PixelFormat, PixelType, const GLvoid*, const PixelStorage&);
#endif
#endif
void AbstractTexture::compressedSubImageImplementationDefault(const GLint level, const Vector3i& offset, const Vector3i& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
bindInternal();
#ifndef MAGNUM_TARGET_GLES2
glCompressedTexSubImage3D(_target, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), dataSize, data);
#else
glCompressedTexSubImage3DOES(_target, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), dataSize, data);
#endif
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::subImage3DImplementationDSA(const GLint level, const Vector3i& offset, const Vector3i& size, const PixelFormat format, const PixelType type, const GLvoid* const data, const PixelStorage&) {
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTextureSubImage3D(_id, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), GLenum(type), data);
}
void AbstractTexture::compressedSubImageImplementationDSA(const GLint level, const Vector3i& offset, const Vector3i& size, const CompressedPixelFormat format, const GLvoid* const data, const GLsizei dataSize) {
glCompressedTextureSubImage3D(_id, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z(), GLenum(format), dataSize, data);
}
#endif
void AbstractTexture::invalidateImageImplementationNoOp(GLint) {}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::invalidateImageImplementationARB(GLint level) {
createIfNotAlready();
glInvalidateTexImage(_id, level);
}
#endif
void AbstractTexture::invalidateSubImageImplementationNoOp(GLint, const Vector3i&, const Vector3i&) {}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::invalidateSubImageImplementationARB(GLint level, const Vector3i& offset, const Vector3i& size) {
createIfNotAlready();
glInvalidateTexSubImage(_id, level, offset.x(), offset.y(), offset.z(), size.x(), size.y(), size.z());
}
#endif
#ifndef DOXYGEN_GENERATING_OUTPUT
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
template<UnsignedInt dimensions> void AbstractTexture::image(GLint level, Image<dimensions>& image) {
const Math::Vector<dimensions, Int> size = DataHelper<dimensions>::imageSize(*this, level);
const std::size_t dataSize = Magnum::Implementation::imageDataSizeFor(image, size);
/* Reallocate only if needed */
Containers::Array<char> data{image.release()};
if(data.size() < dataSize)
data = Containers::Array<char>{dataSize};
Buffer::unbindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(this->*Context::current().state().texture->getImageImplementation)(level, pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), data.size(), data);
image = Image<dimensions>{image.storage(), image.format(), image.formatExtra(), image.pixelSize(), size, std::move(data)};
}
template void MAGNUM_GL_EXPORT AbstractTexture::image<1>(GLint, Image<1>&);
template void MAGNUM_GL_EXPORT AbstractTexture::image<2>(GLint, Image<2>&);
template void MAGNUM_GL_EXPORT AbstractTexture::image<3>(GLint, Image<3>&);
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
template<UnsignedInt dimensions> void AbstractTexture::image(GLint level, BufferImage<dimensions>& image, BufferUsage usage) {
const Math::Vector<dimensions, Int> size = DataHelper<dimensions>::imageSize(*this, level);
const std::size_t dataSize = Magnum::Implementation::imageDataSizeFor(image, size);
/* Reallocate only if needed */
if(image.dataSize() < dataSize)
image.setData(image.storage(), image.format(), image.type(), size, {nullptr, dataSize}, usage);
else
image.setData(image.storage(), image.format(), image.type(), size, nullptr, usage);
image.buffer().bindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
(this->*Context::current().state().texture->getImageImplementation)(level, image.format(), image.type(), dataSize, nullptr);
}
template void MAGNUM_GL_EXPORT AbstractTexture::image<1>(GLint, BufferImage<1>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::image<2>(GLint, BufferImage<2>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::image<3>(GLint, BufferImage<3>&, BufferUsage);
template<UnsignedInt dimensions> void AbstractTexture::compressedImage(const GLint level, CompressedImage<dimensions>& image) {
const Math::Vector<dimensions, Int> size = DataHelper<dimensions>::imageSize(*this, level);
/* If the user-provided pixel storage doesn't tell us all properties about
the compression, we need to ask GL for it */
std::size_t dataSize;
if(!image.storage().compressedBlockSize().product() || !image.storage().compressedBlockDataSize()) {
GLint textureDataSize;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_COMPRESSED_IMAGE_SIZE, &textureDataSize);
dataSize = textureDataSize;
} else dataSize = Magnum::Implementation::compressedImageDataSizeFor(image, size);
/* Internal texture format */
GLint format;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_INTERNAL_FORMAT, &format);
/* Reallocate only if needed */
Containers::Array<char> data{image.release()};
if(data.size() < dataSize)
data = Containers::Array<char>{dataSize};
Buffer::unbindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
(this->*Context::current().state().texture->getCompressedImageImplementation)(level, data.size(), data);
image = CompressedImage<dimensions>{image.storage(), CompressedPixelFormat(format), size, std::move(data)};
}
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<1>(GLint, CompressedImage<1>&);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<2>(GLint, CompressedImage<2>&);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<3>(GLint, CompressedImage<3>&);
template<UnsignedInt dimensions> void AbstractTexture::compressedImage(const GLint level, CompressedBufferImage<dimensions>& image, BufferUsage usage) {
const Math::Vector<dimensions, Int> size = DataHelper<dimensions>::imageSize(*this, level);
/* If the user-provided pixel storage doesn't tell us all properties about
the compression, we need to ask GL for it */
std::size_t dataSize;
if(!image.storage().compressedBlockSize().product() || !image.storage().compressedBlockDataSize()) {
GLint textureDataSize;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_COMPRESSED_IMAGE_SIZE, &textureDataSize);
dataSize = textureDataSize;
} else dataSize = Magnum::Implementation::compressedImageDataSizeFor(image, size);
/* Internal texture format */
GLint format;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_INTERNAL_FORMAT, &format);
/* Reallocate only if needed */
if(image.dataSize() < dataSize)
image.setData(image.storage(), CompressedPixelFormat(format), size, {nullptr, dataSize}, usage);
else
image.setData(image.storage(), CompressedPixelFormat(format), size, nullptr, usage);
image.buffer().bindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
(this->*Context::current().state().texture->getCompressedImageImplementation)(level, dataSize, nullptr);
}
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<1>(GLint, CompressedBufferImage<1>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<2>(GLint, CompressedBufferImage<2>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedImage<3>(GLint, CompressedBufferImage<3>&, BufferUsage);
template<UnsignedInt dimensions> void AbstractTexture::subImage(const GLint level, const RangeTypeFor<dimensions, Int>& range, Image<dimensions>& image) {
createIfNotAlready();
const Math::Vector<dimensions, Int> size = range.size();
const std::size_t dataSize = Magnum::Implementation::imageDataSizeFor(image, size);
const Vector3i paddedOffset = Vector3i::pad<dimensions>(range.min());
const Vector3i paddedSize = Vector3i::pad(size, 1);
/* Reallocate only if needed */
Containers::Array<char> data{image.release()};
if(data.size() < dataSize)
data = Containers::Array<char>{dataSize};
Buffer::unbindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
glGetTextureSubImage(_id, level, paddedOffset.x(), paddedOffset.y(), paddedOffset.z(), paddedSize.x(), paddedSize.y(), paddedSize.z(), GLenum(pixelFormat(image.format())), GLenum(pixelType(image.format(), image.formatExtra())), data.size(), data);
image = Image<dimensions>{image.storage(), image.format(), image.formatExtra(), image.pixelSize(), size, std::move(data)};
}
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<1>(GLint, const Range1Di&, Image<1>&);
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<2>(GLint, const Range2Di&, Image<2>&);
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<3>(GLint, const Range3Di&, Image<3>&);
template<UnsignedInt dimensions> void AbstractTexture::subImage(const GLint level, const RangeTypeFor<dimensions, Int>& range, BufferImage<dimensions>& image, const BufferUsage usage) {
createIfNotAlready();
const Math::Vector<dimensions, Int> size = range.size();
const std::size_t dataSize = Magnum::Implementation::imageDataSizeFor(image, size);
const Vector3i paddedOffset = Vector3i::pad<dimensions>(range.min());
const Vector3i paddedSize = Vector3i::pad(size, 1);
/* Reallocate only if needed */
if(image.dataSize() < dataSize)
image.setData(image.storage(), image.format(), image.type(), size, {nullptr, dataSize}, usage);
else
image.setData(image.storage(), image.format(), image.type(), size, nullptr, usage);
image.buffer().bindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
glGetTextureSubImage(_id, level, paddedOffset.x(), paddedOffset.y(), paddedOffset.z(), paddedSize.x(), paddedSize.y(), paddedSize.z(), GLenum(image.format()), GLenum(image.type()), dataSize, nullptr);
}
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<1>(GLint, const Range1Di&, BufferImage<1>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<2>(GLint, const Range2Di&, BufferImage<2>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::subImage<3>(GLint, const Range3Di&, BufferImage<3>&, BufferUsage);
template<UnsignedInt dimensions> std::size_t AbstractTexture::compressedSubImageSize(TextureFormat format, const Math::Vector<dimensions, Int>& size) {
/* Amount of blocks in given range (rounded up) multiplied by block
data size. The user is responsible for proper block-aligned size. */
const Math::Vector<dimensions, Int> blockSize = DataHelper<dimensions>::compressedBlockSize(_target, format);
return ((size + blockSize - Math::Vector<dimensions, Int>{1})/blockSize).product()*
compressedBlockDataSize(_target, format);
}
template<UnsignedInt dimensions> void AbstractTexture::compressedSubImage(const GLint level, const RangeTypeFor<dimensions, Int>& range, CompressedImage<dimensions>& image) {
createIfNotAlready();
const Math::Vector<dimensions, Int> size = range.size();
const Vector3i paddedOffset = Vector3i::pad<dimensions>(range.min());
const Vector3i paddedSize = Vector3i::pad(size, 1);
/* Internal texture format */
GLint format;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_INTERNAL_FORMAT, &format);
/* Calculate compressed subimage size. If the user-provided pixel storage
doesn't tell us all properties about the compression, we need to ask GL
for it. That requires GL_ARB_internalformat_query2. */
std::size_t dataSize;
if(!image.storage().compressedBlockSize().product() || !image.storage().compressedBlockDataSize())
dataSize = compressedSubImageSize<dimensions>(TextureFormat(format), size);
else dataSize = Magnum::Implementation::compressedImageDataSizeFor(image, size);
/* Reallocate only if needed */
Containers::Array<char> data{image.release()};
if(data.size() < dataSize)
data = Containers::Array<char>{dataSize};
Buffer::unbindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
glGetCompressedTextureSubImage(_id, level, paddedOffset.x(), paddedOffset.y(), paddedOffset.z(), paddedSize.x(), paddedSize.y(), paddedSize.z(), data.size(), data);
image = CompressedImage<dimensions>{CompressedPixelFormat(format), size, std::move(data)};
}
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<1>(GLint, const Range1Di&, CompressedImage<1>&);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<2>(GLint, const Range2Di&, CompressedImage<2>&);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<3>(GLint, const Range3Di&, CompressedImage<3>&);
template<UnsignedInt dimensions> void AbstractTexture::compressedSubImage(const GLint level, const RangeTypeFor<dimensions, Int>& range, CompressedBufferImage<dimensions>& image, const BufferUsage usage) {
createIfNotAlready();
const Math::Vector<dimensions, Int> size = range.size();
const Vector3i paddedOffset = Vector3i::pad<dimensions>(range.min());
const Vector3i paddedSize = Vector3i::pad(size, 1);
/* Internal texture format */
GLint format;
(this->*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_INTERNAL_FORMAT, &format);
/* Calculate compressed subimage size. If the user-provided pixel storage
doesn't tell us all properties about the compression, we need to ask GL
for it. That requires GL_ARB_internalformat_query2. */
std::size_t dataSize;
if(!image.storage().compressedBlockSize().product() || !image.storage().compressedBlockDataSize())
dataSize = compressedSubImageSize<dimensions>(TextureFormat(format), size);
else dataSize = Magnum::Implementation::compressedImageDataSizeFor(image, size);
/* Reallocate only if needed */
if(image.dataSize() < dataSize)
image.setData(image.storage(), CompressedPixelFormat(format), size, {nullptr, dataSize}, usage);
else
image.setData(image.storage(), CompressedPixelFormat(format), size, nullptr, usage);
image.buffer().bindInternal(Buffer::TargetHint::PixelPack);
Context::current().state().renderer->applyPixelStoragePack(image.storage());
glGetCompressedTextureSubImage(_id, level, paddedOffset.x(), paddedOffset.y(), paddedOffset.z(), paddedSize.x(), paddedSize.y(), paddedSize.z(), dataSize, nullptr);
}
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<1>(GLint, const Range1Di&, CompressedBufferImage<1>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<2>(GLint, const Range2Di&, CompressedBufferImage<2>&, BufferUsage);
template void MAGNUM_GL_EXPORT AbstractTexture::compressedSubImage<3>(GLint, const Range3Di&, CompressedBufferImage<3>&, BufferUsage);
#endif
#endif
#ifndef DOXYGEN_GENERATING_OUTPUT
#ifndef MAGNUM_TARGET_GLES
Math::Vector<1, GLint> AbstractTexture::DataHelper<1>::compressedBlockSize(const GLenum target, const TextureFormat format) {
GLint value;
glGetInternalformativ(target, GLenum(format), GL_TEXTURE_COMPRESSED_BLOCK_WIDTH, 1, &value);
return value;
}
Vector2i AbstractTexture::DataHelper<2>::compressedBlockSize(const GLenum target, const TextureFormat format) {
Vector2i value{Math::NoInit};
glGetInternalformativ(target, GLenum(format), GL_TEXTURE_COMPRESSED_BLOCK_WIDTH, 1, &value.x());
glGetInternalformativ(target, GLenum(format), GL_TEXTURE_COMPRESSED_BLOCK_HEIGHT, 1, &value.y());
return value;
}
Vector3i AbstractTexture::DataHelper<3>::compressedBlockSize(const GLenum target, const TextureFormat format) {
/** @todo use real value when OpenGL has proper queries for 3D compression formats */
return Vector3i{DataHelper<2>::compressedBlockSize(target, format), 1};
}
#endif
#ifndef MAGNUM_TARGET_GLES
Math::Vector<1, GLint> AbstractTexture::DataHelper<1>::imageSize(AbstractTexture& texture, const GLint level) {
Math::Vector<1, GLint> value;
(texture.*Context::current().state().texture->getLevelParameterivImplementation)(level, GL_TEXTURE_WIDTH, &value[0]);
return value;
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
Vector2i AbstractTexture::DataHelper<2>::imageSize(AbstractTexture& texture, const GLint level) {
const Implementation::TextureState& state = *Context::current().state().texture;
Vector2i value;
(texture.*state.getLevelParameterivImplementation)(level, GL_TEXTURE_WIDTH, &value[0]);
(texture.*state.getLevelParameterivImplementation)(level, GL_TEXTURE_HEIGHT, &value[1]);
return value;
}
Vector3i AbstractTexture::DataHelper<3>::imageSize(AbstractTexture& texture, const GLint level) {
const Implementation::TextureState& state = *Context::current().state().texture;
Vector3i value;
(texture.*state.getLevelParameterivImplementation)(level, GL_TEXTURE_WIDTH, &value[0]);
(texture.*state.getLevelParameterivImplementation)(level, GL_TEXTURE_HEIGHT, &value[1]);
(texture.*state.getLevelParameterivImplementation)(level, GL_TEXTURE_DEPTH, &value[2]);
return value;
}
#endif
#ifndef MAGNUM_TARGET_GLES
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<1>::setStorage(AbstractTexture& texture, const GLsizei levels, const TextureFormat internalFormat, const Math::Vector< 1, GLsizei >& size) {
(texture.*Context::current().state().texture->storage1DImplementation)(levels, internalFormat, size);
}
#endif
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<2>::setStorage(AbstractTexture& texture, const GLsizei levels, const TextureFormat internalFormat, const Vector2i& size) {
(texture.*Context::current().state().texture->storage2DImplementation)(levels, internalFormat, size);
}
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<3>::setStorage(AbstractTexture& texture, const GLsizei levels, const TextureFormat internalFormat, const Vector3i& size) {
(texture.*Context::current().state().texture->storage3DImplementation)(levels, internalFormat, size);
}
#endif
#if !defined(MAGNUM_TARGET_GLES2) && !defined(MAGNUM_TARGET_WEBGL)
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<2>::setStorageMultisample(AbstractTexture& texture, const GLsizei samples, const TextureFormat internalFormat, const Vector2i& size, const GLboolean fixedSampleLocations) {
(texture.*Context::current().state().texture->storage2DMultisampleImplementation)(samples, internalFormat, size, fixedSampleLocations);
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<3>::setStorageMultisample(AbstractTexture& texture, const GLsizei samples, const TextureFormat internalFormat, const Vector3i& size, const GLboolean fixedSampleLocations) {
(texture.*Context::current().state().texture->storage3DMultisampleImplementation)(samples, internalFormat, size, fixedSampleLocations);
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::DataHelper<1>::setImage(AbstractTexture& texture, const GLint level, const TextureFormat internalFormat, const ImageView1D& image) {
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
glTexImage1D(texture._target, level, GLint(internalFormat), image.size()[0], 0, GLenum(pixelFormat(image.format())), GLenum(pixelType(image.format(), image.formatExtra())), image.data());
}
void AbstractTexture::DataHelper<1>::setCompressedImage(AbstractTexture& texture, const GLint level, const CompressedImageView1D& image) {
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
glCompressedTexImage1D(texture._target, level, GLenum(image.format()), image.size()[0], 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()), image.data());
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<1>::setImage(AbstractTexture& texture, const GLint level, const TextureFormat internalFormat, BufferImage1D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexImage1D(texture._target, level, GLint(internalFormat), image.size()[0], 0, GLenum(image.format()), GLenum(image.type()), nullptr);
}
void AbstractTexture::DataHelper<1>::setCompressedImage(AbstractTexture& texture, const GLint level, CompressedBufferImage1D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
glCompressedTexImage1D(texture._target, level, GLenum(image.format()), image.size()[0], 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()), nullptr);
}
void AbstractTexture::DataHelper<1>::setSubImage(AbstractTexture& texture, const GLint level, const Math::Vector<1, GLint>& offset, const ImageView1D& image) {
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->subImage1DImplementation)(level, offset, image.size(), pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), image.data());
}
void AbstractTexture::DataHelper<1>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Math::Vector<1, GLint>& offset, const CompressedImageView1D& image) {
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->compressedSubImage1DImplementation)(level, offset, image.size(), compressedPixelFormat(image.format()), image.data(), Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()));
}
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<1>::setSubImage(AbstractTexture& texture, const GLint level, const Math::Vector<1, GLint>& offset, BufferImage1D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->subImage1DImplementation)(level, offset, image.size(), image.format(), image.type(), nullptr);
}
void AbstractTexture::DataHelper<1>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Math::Vector<1, GLint>& offset, CompressedBufferImage1D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->compressedSubImage1DImplementation)(level, offset, image.size(), image.format(), nullptr, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()));
}
#endif
void AbstractTexture::DataHelper<2>::setImage(AbstractTexture& texture, const GLenum target, const GLint level, const TextureFormat internalFormat, const ImageView2D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->image2DImplementation)(target, level, internalFormat, image.size(), pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), image.data()
#ifdef MAGNUM_TARGET_GLES2
+ Magnum::Implementation::pixelStorageSkipOffset(image)
#endif
, image.storage());
}
void AbstractTexture::DataHelper<2>::setCompressedImage(AbstractTexture& texture, const GLenum target, const GLint level, const CompressedImageView2D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
glCompressedTexImage2D(target, level, GLenum(compressedPixelFormat(image.format())), image.size().x(), image.size().y(), 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()), image.data());
}
#ifndef MAGNUM_TARGET_GLES2
void AbstractTexture::DataHelper<2>::setImage(AbstractTexture& texture, const GLenum target, const GLint level, const TextureFormat internalFormat, BufferImage2D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
glTexImage2D(target, level, GLint(internalFormat), image.size().x(), image.size().y(), 0, GLenum(image.format()), GLenum(image.type()), nullptr);
}
void AbstractTexture::DataHelper<2>::setCompressedImage(AbstractTexture& texture, const GLenum target, const GLint level, CompressedBufferImage2D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
glCompressedTexImage2D(target, level, GLenum(image.format()), image.size().x(), image.size().y(), 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()), nullptr);
}
#endif
void AbstractTexture::DataHelper<2>::setSubImage(AbstractTexture& texture, const GLint level, const Vector2i& offset, const ImageView2D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->subImage2DImplementation)(level, offset, image.size(), pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), image.data()
#ifdef MAGNUM_TARGET_GLES2
+ Magnum::Implementation::pixelStorageSkipOffset(image)
#endif
, image.storage());
}
void AbstractTexture::DataHelper<2>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Vector2i& offset, const CompressedImageView2D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->compressedSubImage2DImplementation)(level, offset, image.size(), compressedPixelFormat(image.format()), image.data(), Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()));
}
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<2>::setSubImage(AbstractTexture& texture, const GLint level, const Vector2i& offset, BufferImage2D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->subImage2DImplementation)(level, offset, image.size(), image.format(), image.type(), nullptr, image.storage());
}
void AbstractTexture::DataHelper<2>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Vector2i& offset, CompressedBufferImage2D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->compressedSubImage2DImplementation)(level, offset, image.size(), image.format(), nullptr, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()));
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::DataHelper<3>::setImage(AbstractTexture& texture, const GLint level, const TextureFormat internalFormat, const ImageView3D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->image3DImplementation)(level, internalFormat, image.size(), pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), image.data()
#ifdef MAGNUM_TARGET_GLES2
+ Magnum::Implementation::pixelStorageSkipOffset(image)
#endif
, image.storage());
}
void AbstractTexture::DataHelper<3>::setCompressedImage(AbstractTexture& texture, const GLint level, const CompressedImageView3D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
#ifndef MAGNUM_TARGET_GLES2
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
glCompressedTexImage3D(texture._target, level, GLenum(compressedPixelFormat(image.format())), image.size().x(), image.size().y(), image.size().z(), 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()), image.data());
#else
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
glCompressedTexImage3DOES(texture._target, level, GLenum(compressedPixelFormat(image.format())), image.size().x(), image.size().y(), image.size().z(), 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()), image.data());
#endif
}
#endif
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<3>::setImage(AbstractTexture& texture, const GLint level, const TextureFormat internalFormat, BufferImage3D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
glTexImage3D(texture._target, level, GLint(internalFormat), image.size().x(), image.size().y(), image.size().z(), 0, GLenum(image.format()), GLenum(image.type()), nullptr);
}
void AbstractTexture::DataHelper<3>::setCompressedImage(AbstractTexture& texture, const GLint level, CompressedBufferImage3D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
texture.bindInternal();
glCompressedTexImage3D(texture._target, level, GLenum(image.format()), image.size().x(), image.size().y(), image.size().z(), 0, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()), nullptr);
}
#endif
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::DataHelper<3>::setSubImage(AbstractTexture& texture, const GLint level, const Vector3i& offset, const ImageView3D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->subImage3DImplementation)(level, offset, image.size(), pixelFormat(image.format()), pixelType(image.format(), image.formatExtra()), image.data()
#ifdef MAGNUM_TARGET_GLES2
+ Magnum::Implementation::pixelStorageSkipOffset(image)
#endif
, image.storage());
}
void AbstractTexture::DataHelper<3>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Vector3i& offset, const CompressedImageView3D& image) {
#ifndef MAGNUM_TARGET_GLES2
Buffer::unbindInternal(Buffer::TargetHint::PixelUnpack);
#endif
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
Split the OpenGL layer out, pt 9: generic pixel formats. This is quite big, so: * There are new Magnum::PixelFormat and Magnum::CompressedPixelFormat enums, which contain generic API-independent formats. In particular, PixelFormat replaces GL::PixelFormat and GL::PixelType with a single value. * There's GL::pixelFormat(), GL::pixelType(), GL::compressedPixelFormat() to convert the generic enums to GL-specific. The mapping is only in one direction, done with a lookup table (generic enums are indices to that table). * GL classes taking the formats directly (such as GL::BufferImage) have overloads that take both the GL-specific and generic format. * The generic Image, CompressedImage, ImageView, CompressedImageView, and Trade::ImageData classes now accept the generic formats first-class. However, it's also possible to store an implementation-specific value to cover cases where a generic format enum doesn't have support for a particular format. This is done by wrapping the value using pixelFormatWrap() or compressedPixelFormatWrap(). Particular GPU APIs then assume it's their implementation-specific value and extract the value back using pixelFormatUnwrap() or compressedPixelFormatUnwrap(). There's also an isPixelFormatImplementationSpecific() and isCompressedPixelFormatImplementationSpecific() that distinguishes these values. * Many operations need pixel size and in order to have it even for implementation-specific formats, a corresponding pixelSize() overload is found via ADL on construction and the calculated size stored along the format. Previously the pixel size was only calculated on demand, but that's not possible now. In case such overload is not available, it's possible to pass pixel size manually as well. * In order to support the GL format+type pair, Image, ImageView and Trade::ImageData, there's now an additional untyped formatExtra() field that holds the second value. * The CompressedPixelStorage class is now unconditionally available on all targets, including OpenGL ES and WebGL. However, on OpenGL ES the GL APIs expect that it's all at default values. I attempted to preserve backwards compatibility as much as possible: * The PixelFormat and CompressedPixelFormat enum now contains generic API-independent values. The GL-specific formats are present there, but marked as deprecated. Use either the generic values or GL::PixelFormat (togehter with GL::PixelType) and GL::CompressedPixelFormat instead. There's a lot of ugliness caused by this, but seems to work well. * *Image::type() functions are deprecated as they were too GL-specific. Use formatExtra() and cast it to GL::PixelType instead. * Image constructors take templated format or format+extra arguments, so passing GL-specific values to them should still work.
8 years ago
(texture.*Context::current().state().texture->compressedSubImage3DImplementation)(level, offset, image.size(), compressedPixelFormat(image.format()), image.data(), Magnum::Implementation::occupiedCompressedImageDataSize(image, image.data().size()));
}
#endif
#ifndef MAGNUM_TARGET_GLES2
Full ARB_direct_state_access support in textures. ARB_DSA is now preferred in single-bind cases, as it is easier to use than passing pointers to ARB_multi_bind. ARB_multi_bind was preferred for single-bind previously simply because EXT_DSA was not in core. Because there is a lot to say about feature selection for each function, I took this as a opportunity to remove redundant documentation blocks, just refer to Texture documentation from everywhere and add extension requirements and deprecation where needed, so it's clear for each class what needs what. ARB_DSA also took the opportunity to finally remove all target enum values from function calls and because of that the CubeMapTexture has to handle a bunch of special cases. In order: - CubeMapTexture::imageSize() now doesn't take face parameter and returns one value for all faces. I'm thus now also assuming that the user is sane and called either setStorage() or setImage() with the same size for all faces. In non-ARB_DSA path I'm thus querying only size of +X face and returning it as size for all faces. The old imageSize(Coordinate, Int) overload is still present, but ignores the first parameter and calls imageSize(Int). It is marked as deprecated and will be removed in some future release. - CubeMapTexture::image() now needs to call glTextureSubImage() in ARB_DSA path to make it possible to extract single face. Other code paths (EXT_DSA, Robustness and "default") remain the same. - CubeMapTexture::setSubImage() calls glTextureSubImage3D() in ARB_DSA, because it is not possible to specify face index in glTextureSubImage2D(). Other code paths (EXT_DSA and "default") remain the same. Implementation of these special cases is extracted into CubeMapTexture class to avoid pollution of AbstractTexture with incompatible nonsense.
12 years ago
void AbstractTexture::DataHelper<3>::setSubImage(AbstractTexture& texture, const GLint level, const Vector3i& offset, BufferImage3D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->subImage3DImplementation)(level, offset, image.size(), image.format(), image.type(), nullptr, image.storage());
}
void AbstractTexture::DataHelper<3>::setCompressedSubImage(AbstractTexture& texture, const GLint level, const Vector3i& offset, CompressedBufferImage3D& image) {
image.buffer().bindInternal(Buffer::TargetHint::PixelUnpack);
Context::current().state().renderer->applyPixelStorageUnpack(image.storage());
(texture.*Context::current().state().texture->compressedSubImage3DImplementation)(level, offset, image.size(), image.format(), nullptr, Magnum::Implementation::occupiedCompressedImageDataSize(image, image.dataSize()));
}
#endif
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::DataHelper<1>::invalidateSubImage(AbstractTexture& texture, const GLint level, const Math::Vector<1, GLint>& offset, const Math::Vector<1, GLint>& size) {
(texture.*Context::current().state().texture->invalidateSubImageImplementation)(level, {offset[0], 0, 0}, {size[0], 1, 1});
}
#endif
void AbstractTexture::DataHelper<2>::invalidateSubImage(AbstractTexture& texture, const GLint level, const Vector2i& offset, const Vector2i& size) {
(texture.*Context::current().state().texture->invalidateSubImageImplementation)(level, {offset, 0}, {size, 1});
}
void AbstractTexture::DataHelper<3>::invalidateSubImage(AbstractTexture& texture, const GLint level, const Vector3i& offset, const Vector3i& size) {
(texture.*Context::current().state().texture->invalidateSubImageImplementation)(level, offset, size);
}
#ifndef MAGNUM_TARGET_GLES
void AbstractTexture::DataHelper<1>::setWrapping(AbstractTexture& texture, const Array1D<SamplerWrapping>& wrapping) {
(texture.*Context::current().state().texture->parameteriImplementation)(GL_TEXTURE_WRAP_S, GLint(wrapping.x()));
}
#endif
void AbstractTexture::DataHelper<2>::setWrapping(AbstractTexture& texture, const Array2D<SamplerWrapping>& wrapping) {
const Implementation::TextureState& state = *Context::current().state().texture;
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_S, GLint(wrapping.x()));
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_T, GLint(wrapping.y()));
}
#if !(defined(MAGNUM_TARGET_WEBGL) && defined(MAGNUM_TARGET_GLES2))
void AbstractTexture::DataHelper<3>::setWrapping(AbstractTexture& texture, const Array3D<SamplerWrapping>& wrapping) {
const Implementation::TextureState& state = *Context::current().state().texture;
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_S, GLint(wrapping.x()));
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_T, GLint(wrapping.y()));
#ifndef MAGNUM_TARGET_GLES2
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_R, GLint(wrapping.z()));
#else
(texture.*state.parameteriImplementation)(GL_TEXTURE_WRAP_R_OES, GLint(wrapping.z()));
#endif
}
#endif
#endif
}}