qtdeclarative/src/declarative/items/context2d/qsgcontext2dtexture.cpp

778 lines
20 KiB
C++
Raw Normal View History

/****************************************************************************
**
** Copyright (C) 2011 Nokia Corporation and/or its subsidiary(-ies).
** All rights reserved.
** Contact: Nokia Corporation (qt-info@nokia.com)
**
** This file is part of the QtDeclarative module of the Qt Toolkit.
**
** $QT_BEGIN_LICENSE:LGPL$
** GNU Lesser General Public License Usage
** This file may be used under the terms of the GNU Lesser General Public
** License version 2.1 as published by the Free Software Foundation and
** appearing in the file LICENSE.LGPL included in the packaging of this
** file. Please review the following information to ensure the GNU Lesser
** General Public License version 2.1 requirements will be met:
** http://www.gnu.org/licenses/old-licenses/lgpl-2.1.html.
**
** In addition, as a special exception, Nokia gives you certain additional
** rights. These rights are described in the Nokia Qt LGPL Exception
** version 1.1, included in the file LGPL_EXCEPTION.txt in this package.
**
** GNU General Public License Usage
** Alternatively, this file may be used under the terms of the GNU General
** Public License version 3.0 as published by the Free Software Foundation
** and appearing in the file LICENSE.GPL included in the packaging of this
** file. Please review the following information to ensure the GNU General
** Public License version 3.0 requirements will be met:
** http://www.gnu.org/copyleft/gpl.html.
**
** Other Usage
** Alternatively, this file may be used in accordance with the terms and
** conditions contained in a signed written agreement between you and Nokia.
**
**
**
**
**
** $QT_END_LICENSE$
**
****************************************************************************/
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
#include "qsgcontext2dtexture_p.h"
#include "qsgcontext2dtile_p.h"
#include "qsgcanvasitem_p.h"
#include <private/qsgitem_p.h>
#include <private/qsgtexture_p.h>
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
#include "qsgcontext2dcommandbuffer_p.h"
#include <QOpenGLPaintDevice>
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
#include <QOpenGLFramebufferObject>
#include <QOpenGLFramebufferObjectFormat>
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
#include <QtCore/QThread>
#define QT_MINIMUM_FBO_SIZE 64
static inline int qt_next_power_of_two(int v)
{
v--;
v |= v >> 1;
v |= v >> 2;
v |= v >> 4;
v |= v >> 8;
v |= v >> 16;
++v;
return v;
}
Q_GLOBAL_STATIC(QThread, globalCanvasThreadRenderInstance)
QSGContext2DTexture::QSGContext2DTexture()
: QSGDynamicTexture()
, m_context(0)
, m_canvasSize(QSize(1, 1))
, m_tileSize(QSize(1, 1))
, m_canvasWindow(QRect(0, 0, 1, 1))
, m_dirtyCanvas(false)
, m_dirtyTexture(false)
, m_threadRendering(false)
, m_smooth(false)
, m_tiledCanvas(false)
, m_doGrabImage(false)
, m_painting(false)
{
}
QSGContext2DTexture::~QSGContext2DTexture()
{
clearTiles();
}
QSize QSGContext2DTexture::textureSize() const
{
return m_canvasWindow.size();
}
void QSGContext2DTexture::markDirtyTexture()
{
lock();
m_dirtyTexture = true;
unlock();
emit textureChanged();
}
bool QSGContext2DTexture::setCanvasSize(const QSize &size)
{
if (m_canvasSize != size) {
m_canvasSize = size;
m_dirtyCanvas = true;
return true;
}
return false;
}
bool QSGContext2DTexture::setTileSize(const QSize &size)
{
if (m_tileSize != size) {
m_tileSize = size;
m_dirtyCanvas = true;
return true;
}
return false;
}
void QSGContext2DTexture::setSmooth(bool smooth)
{
m_smooth = smooth;
}
void QSGContext2DTexture::setItem(QSGCanvasItem* item)
{
if (!item) {
lock();
m_item = 0;
m_context = 0;
unlock();
wake();
} else if (m_item != item) {
lock();
m_item = item;
m_context = item->context();
m_state = m_context->state;
unlock();
connect(this, SIGNAL(textureChanged()), m_item, SIGNAL(painted()));
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
}
bool QSGContext2DTexture::setCanvasWindow(const QRect& r)
{
if (m_canvasWindow != r) {
m_canvasWindow = r;
return true;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
return false;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
bool QSGContext2DTexture::setDirtyRect(const QRect &r)
{
bool doDirty = false;
if (m_tiledCanvas) {
foreach (QSGContext2DTile* t, m_tiles) {
bool dirty = t->rect().intersected(r).isValid();
t->markDirty(dirty);
if (dirty)
doDirty = true;
}
} else {
doDirty = m_canvasWindow.intersected(r).isValid();
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
return doDirty;
}
void QSGContext2DTexture::canvasChanged(const QSize& canvasSize, const QSize& tileSize, const QRect& canvasWindow, const QRect& dirtyRect, bool smooth)
{
lock();
QSize ts = tileSize;
if (ts.width() > canvasSize.width())
ts.setWidth(canvasSize.width());
if (ts.height() > canvasSize.height())
ts.setHeight(canvasSize.height());
setCanvasSize(canvasSize);
setTileSize(ts);
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
if (canvasSize == canvasWindow.size()) {
m_tiledCanvas = false;
m_dirtyCanvas = false;
} else {
m_tiledCanvas = true;
}
bool doDirty = false;
if (dirtyRect.isValid())
doDirty = setDirtyRect(dirtyRect);
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
bool windowChanged = setCanvasWindow(canvasWindow);
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
if (windowChanged || doDirty) {
if (m_threadRendering)
QMetaObject::invokeMethod(this, "paint", Qt::QueuedConnection);
else if (supportDirectRendering()) {
QMetaObject::invokeMethod(this, "paint", Qt::DirectConnection);
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
setSmooth(smooth);
unlock();
}
void QSGContext2DTexture::paintWithoutTiles()
{
QSGContext2DCommandBuffer* ccb = m_context->buffer();
if (ccb->isEmpty() && m_threadRendering && !m_doGrabImage) {
lock();
if (m_item)
QMetaObject::invokeMethod(m_item, "_doPainting", Qt::QueuedConnection, Q_ARG(QRectF, QRectF(0, 0, m_canvasSize.width(), m_canvasSize.height())));
wait();
unlock();
}
if (ccb->isEmpty()) {
return;
}
QPaintDevice* device = beginPainting();
if (!device) {
endPainting();
return;
}
QPainter p;
p.begin(device);
if (m_smooth)
p.setRenderHints(QPainter::Antialiasing | QPainter::HighQualityAntialiasing
| QPainter::TextAntialiasing | QPainter::SmoothPixmapTransform);
else
p.setRenderHints(QPainter::Antialiasing | QPainter::HighQualityAntialiasing
| QPainter::TextAntialiasing | QPainter::SmoothPixmapTransform, false);
p.setCompositionMode(QPainter::CompositionMode_SourceOver);
m_state = ccb->replay(&p, m_state);
ccb->clear();
markDirtyTexture();
endPainting();
}
bool QSGContext2DTexture::canvasDestroyed()
{
bool noCanvas = false;
lock();
noCanvas = m_item == 0;
unlock();
return noCanvas;
}
void QSGContext2DTexture::paint()
{
if (canvasDestroyed())
return;
if (!m_tiledCanvas) {
paintWithoutTiles();
} else {
QSGContext2D::State oldState = m_state;
QSGContext2DCommandBuffer* ccb = m_context->buffer();
lock();
QRect tiledRegion = createTiles(m_canvasWindow.intersected(QRect(QPoint(0, 0), m_canvasSize)));
unlock();
if (!tiledRegion.isEmpty()) {
if (m_threadRendering && !m_doGrabImage) {
QRect dirtyRect;
lock();
foreach (QSGContext2DTile* tile, m_tiles) {
if (tile->dirty()) {
if (dirtyRect.isEmpty())
dirtyRect = tile->rect();
else
dirtyRect |= tile->rect();
}
}
unlock();
if (dirtyRect.isValid()) {
lock();
if (m_item)
QMetaObject::invokeMethod(m_item, "_doPainting", Qt::QueuedConnection, Q_ARG(QRectF, dirtyRect));
wait();
unlock();
}
}
if (beginPainting()) {
foreach (QSGContext2DTile* tile, m_tiles) {
bool dirtyTile = false, dirtyCanvas = false, smooth = false;
lock();
dirtyTile = tile->dirty();
smooth = m_smooth;
dirtyCanvas = m_dirtyCanvas;
unlock();
//canvas size or tile size may change during painting tiles
if (dirtyCanvas) {
if (m_threadRendering)
QMetaObject::invokeMethod(this, "paint", Qt::QueuedConnection);
endPainting();
return;
} else if (dirtyTile) {
m_state = ccb->replay(tile->createPainter(smooth), oldState);
tile->drawFinished();
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
lock();
tile->markDirty(false);
unlock();
}
compositeTile(tile);
}
ccb->clear();
endPainting();
markDirtyTexture();
}
}
}
}
QRect QSGContext2DTexture::tiledRect(const QRectF& window, const QSize& tileSize)
{
if (window.isEmpty())
return QRect();
const int tw = tileSize.width();
const int th = tileSize.height();
const int h1 = window.left() / tw;
const int v1 = window.top() / th;
const int htiles = ((window.right() - h1 * tw) + tw - 1)/tw;
const int vtiles = ((window.bottom() - v1 * th) + th - 1)/th;
return QRect(h1 * tw, v1 * th, htiles * tw, vtiles * th);
}
QRect QSGContext2DTexture::createTiles(const QRect& window)
{
QList<QSGContext2DTile*> oldTiles = m_tiles;
m_tiles.clear();
if (window.isEmpty()) {
m_dirtyCanvas = false;
return QRect();
}
QRect r = tiledRect(window, m_tileSize);
const int tw = m_tileSize.width();
const int th = m_tileSize.height();
const int h1 = window.left() / tw;
const int v1 = window.top() / th;
const int htiles = r.width() / tw;
const int vtiles = r.height() / th;
for (int yy = 0; yy < vtiles; ++yy) {
for (int xx = 0; xx < htiles; ++xx) {
int ht = xx + h1;
int vt = yy + v1;
QSGContext2DTile* tile = 0;
QPoint pos(ht * tw, vt * th);
QRect rect(pos, m_tileSize);
for (int i = 0; i < oldTiles.size(); i++) {
if (oldTiles[i]->rect() == rect) {
tile = oldTiles.takeAt(i);
break;
}
}
if (!tile)
tile = createTile();
tile->setRect(rect);
m_tiles.append(tile);
}
}
qDeleteAll(oldTiles);
m_dirtyCanvas = false;
return r;
}
void QSGContext2DTexture::clearTiles()
{
qDeleteAll(m_tiles);
m_tiles.clear();
}
QSGContext2DFBOTexture::QSGContext2DFBOTexture()
: QSGContext2DTexture()
, m_fbo(0)
, m_multisampledFbo(0)
, m_paint_device(0)
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
{
m_threadRendering = false;
}
QSGContext2DFBOTexture::~QSGContext2DFBOTexture()
{
delete m_fbo;
delete m_multisampledFbo;
delete m_paint_device;
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
bool QSGContext2DFBOTexture::setCanvasSize(const QSize &size)
{
QSize s = QSize(qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(size.width()))
, qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(size.height())));
if (m_canvasSize != s) {
m_canvasSize = s;
m_dirtyCanvas = true;
return true;
}
return false;
}
bool QSGContext2DFBOTexture::setTileSize(const QSize &size)
{
QSize s = QSize(qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(size.width()))
, qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(size.height())));
if (m_tileSize != s) {
m_tileSize = s;
m_dirtyCanvas = true;
return true;
}
return false;
}
bool QSGContext2DFBOTexture::setCanvasWindow(const QRect& canvasWindow)
{
QSize s = QSize(qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(canvasWindow.size().width()))
, qMax(QT_MINIMUM_FBO_SIZE, qt_next_power_of_two(canvasWindow.size().height())));
bool doChanged = false;
if (m_fboSize != s) {
m_fboSize = s;
doChanged = true;
}
if (m_canvasWindow != canvasWindow)
m_canvasWindow = canvasWindow;
return doChanged;
}
void QSGContext2DFBOTexture::bind()
{
glBindTexture(GL_TEXTURE_2D, textureId());
updateBindOptions();
}
QRectF QSGContext2DFBOTexture::textureSubRect() const
{
return QRectF(0
, 0
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
, qreal(m_canvasWindow.width()) / m_fboSize.width()
, qreal(m_canvasWindow.height()) / m_fboSize.height());
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
int QSGContext2DFBOTexture::textureId() const
{
return m_fbo? m_fbo->texture() : 0;
}
bool QSGContext2DFBOTexture::updateTexture()
{
if (!m_context->buffer()->isEmpty()) {
paint();
}
bool textureUpdated = m_dirtyTexture;
m_dirtyTexture = false;
if (m_doGrabImage) {
grabImage();
m_condition.wakeOne();
m_doGrabImage = false;
}
return textureUpdated;
}
QSGContext2DTile* QSGContext2DFBOTexture::createTile() const
{
return new QSGContext2DFBOTile();
}
void QSGContext2DFBOTexture::grabImage()
{
if (m_fbo) {
m_grabedImage = m_fbo->toImage();
}
}
bool QSGContext2DFBOTexture::doMultisampling() const
{
static bool extensionsChecked = false;
static bool multisamplingSupported = false;
if (!extensionsChecked) {
QList<QByteArray> extensions = QByteArray((const char *)glGetString(GL_EXTENSIONS)).split(' ');
multisamplingSupported = extensions.contains("GL_EXT_framebuffer_multisample")
&& extensions.contains("GL_EXT_framebuffer_blit");
extensionsChecked = true;
}
return multisamplingSupported && m_smooth;
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
QImage QSGContext2DFBOTexture::toImage(const QRectF& region)
{
#define QML_CONTEXT2D_WAIT_MAX 5000
m_doGrabImage = true;
if (m_item)
m_item->update();
QImage grabbed;
m_mutex.lock();
bool ok = m_condition.wait(&m_mutex, QML_CONTEXT2D_WAIT_MAX);
if (!ok)
grabbed = QImage();
if (region.isValid())
grabbed = m_grabedImage.copy(region.toRect());
else
grabbed = m_grabedImage;
m_grabedImage = QImage();
return grabbed;
}
void QSGContext2DFBOTexture::compositeTile(QSGContext2DTile* tile)
{
QSGContext2DFBOTile* t = static_cast<QSGContext2DFBOTile*>(tile);
QRect target = t->rect().intersect(m_canvasWindow);
if (target.isValid()) {
QRect source = target;
source.moveTo(source.topLeft() - t->rect().topLeft());
target.moveTo(target.topLeft() - m_canvasWindow.topLeft());
QOpenGLFramebufferObject::blitFramebuffer(m_fbo, target, t->fbo(), source);
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
}
QSGCanvasItem::RenderTarget QSGContext2DFBOTexture::renderTarget() const
{
return QSGCanvasItem::FramebufferObject;
}
QPaintDevice* QSGContext2DFBOTexture::beginPainting()
{
QSGContext2DTexture::beginPainting();
if (m_canvasWindow.size().isEmpty() && !m_threadRendering) {
delete m_fbo;
delete m_multisampledFbo;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
m_fbo = 0;
m_multisampledFbo = 0;
return 0;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
} else if (!m_fbo || m_fbo->size() != m_fboSize) {
delete m_fbo;
delete m_multisampledFbo;
if (doMultisampling()) {
{
QOpenGLFramebufferObjectFormat format;
format.setAttachment(QOpenGLFramebufferObject::CombinedDepthStencil);
format.setSamples(8);
m_multisampledFbo = new QOpenGLFramebufferObject(m_fboSize, format);
}
{
QOpenGLFramebufferObjectFormat format;
format.setAttachment(QOpenGLFramebufferObject::NoAttachment);
m_fbo = new QOpenGLFramebufferObject(m_fboSize, format);
}
} else {
QOpenGLFramebufferObjectFormat format;
format.setAttachment(QOpenGLFramebufferObject::CombinedDepthStencil);
m_fbo = new QOpenGLFramebufferObject(m_fboSize, format);
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
if (doMultisampling())
m_multisampledFbo->bind();
else
m_fbo->bind();
if (!m_paint_device) {
QOpenGLPaintDevice *gl_device = new QOpenGLPaintDevice(m_fbo->size());
gl_device->setPaintFlipped(true);
m_paint_device = gl_device;
}
return m_paint_device;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
void QSGContext2DFBOTexture::endPainting()
{
QSGContext2DTexture::endPainting();
if (m_multisampledFbo) {
QOpenGLFramebufferObject::blitFramebuffer(m_fbo, m_multisampledFbo);
m_multisampledFbo->release();
} else if (m_fbo)
m_fbo->release();
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
void qt_quit_context2d_render_thread()
{
QThread* thread = globalCanvasThreadRenderInstance();
if (thread->isRunning()) {
thread->exit(0);
thread->wait(1000);
}
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
QSGContext2DImageTexture::QSGContext2DImageTexture(bool threadRendering)
: QSGContext2DTexture()
, m_texture(new QSGPlainTexture())
{
m_texture->setOwnsTexture(true);
m_texture->setHasMipmaps(false);
m_threadRendering = threadRendering;
if (m_threadRendering) {
QThread* thread = globalCanvasThreadRenderInstance();
moveToThread(thread);
if (!thread->isRunning()) {
qAddPostRoutine(qt_quit_context2d_render_thread);
thread->start();
}
}
}
QSGContext2DImageTexture::~QSGContext2DImageTexture()
{
delete m_texture;
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
int QSGContext2DImageTexture::textureId() const
{
return m_texture->textureId();
}
void QSGContext2DImageTexture::lock()
{
if (m_threadRendering)
m_mutex.lock();
}
void QSGContext2DImageTexture::unlock()
{
if (m_threadRendering)
m_mutex.unlock();
}
void QSGContext2DImageTexture::wait()
{
if (m_threadRendering)
m_waitCondition.wait(&m_mutex);
}
void QSGContext2DImageTexture::wake()
{
if (m_threadRendering)
m_waitCondition.wakeOne();
}
bool QSGContext2DImageTexture::supportDirectRendering() const
{
return !m_threadRendering;
}
QSGCanvasItem::RenderTarget QSGContext2DImageTexture::renderTarget() const
{
return QSGCanvasItem::Image;
}
void QSGContext2DImageTexture::bind()
{
m_texture->bind();
}
bool QSGContext2DImageTexture::updateTexture()
{
lock();
bool textureUpdated = m_dirtyTexture;
if (m_dirtyTexture) {
m_texture->setImage(m_image);
m_dirtyTexture = false;
}
unlock();
return textureUpdated;
}
QSGContext2DTile* QSGContext2DImageTexture::createTile() const
{
return new QSGContext2DImageTile();
}
void QSGContext2DImageTexture::grabImage(const QRect& r)
{
m_doGrabImage = true;
paint();
m_doGrabImage = false;
m_grabedImage = m_image.copy(r);
}
QImage QSGContext2DImageTexture::toImage(const QRectF& region)
{
QRect r = region.isValid() ? region.toRect() : QRect(QPoint(0, 0), m_canvasWindow.size());
if (threadRendering()) {
wake();
QMetaObject::invokeMethod(this, "grabImage", Qt::BlockingQueuedConnection, Q_ARG(QRect, r));
} else {
QMetaObject::invokeMethod(this, "grabImage", Qt::DirectConnection, Q_ARG(QRect, r));
}
QImage image = m_grabedImage;
m_grabedImage = QImage();
return image;
}
QPaintDevice* QSGContext2DImageTexture::beginPainting()
{
QSGContext2DTexture::beginPainting();
if (m_canvasWindow.size().isEmpty())
return 0;
lock();
if (m_image.size() != m_canvasWindow.size()) {
m_image = QImage(m_canvasWindow.size(), QImage::Format_ARGB32_Premultiplied);
m_image.fill(0x00000000);
canvas item refactors 1.Supports tiled canvas with canvasSize, tileSize and canvasWindow 2.Supports different rendering targets: Canvas.Image and Canvas.FrameBufferObject by renderTarget property 3.Supports thread rendering when possible by threadRendering property. 4.Refactors QSGContext2D code, move some logic to QSGContext2DCommandBuffer,QSGContext2DTexture,QSGContext2DTile, etc 5.Updates/adds some canvas examples 6.Some improvements for context2d API 6.1 drawImage() now loads image asynchoronously and draw images automatically when they are ready 6.2 adds fillRule supports 6.3 add svg path supports 6.4 Pixel operations (getImageData/putImageData/createImageData) now have better performance by using V8 indexed array accessors 6.5 Uses QTransform instead of QMatrix 6.6 Gradients/patterns now are V8 values, not QObjects 6.7 Supports measureText and TextMetrics interface 6.8 Gives not support warnings for unimplemented functions (drawFocusRing,setCaretSelectionRect,caretBlinkRate) 6.9 Better error handling, throw standard DOM exceptions according to the HTML5 context2d spec. 6.10 Adds shear, resetTransform to matrix operations 6.11 Adds roundedRect, ellipse, text to path operations 6.12 Adds new features to CanvasImageData interface 1) adds mirror() function 2) adds filter() function, include the following filters: Threshold GrayScale Brightness Invert Blur Blend Opaque Convolute 7. Adds documentations Change-Id: Id19224260d6a3fdc589d1f9681c34a88a7e7b3e5 Reviewed-on: http://codereview.qt-project.org/3621 Reviewed-by: Charles Yin <charles.yin@nokia.com>
2011-08-09 06:44:38 +00:00
}
unlock();
return &m_image;
}
void QSGContext2DImageTexture::compositeTile(QSGContext2DTile* tile)
{
Q_ASSERT(!tile->dirty());
QSGContext2DImageTile* t = static_cast<QSGContext2DImageTile*>(tile);
QRect target = t->rect().intersect(m_canvasWindow);
if (target.isValid()) {
QRect source = target;
source.moveTo(source.topLeft() - t->rect().topLeft());
target.moveTo(target.topLeft() - m_canvasWindow.topLeft());
lock();
m_painter.begin(&m_image);
m_painter.setCompositionMode(QPainter::CompositionMode_Source);
m_painter.drawImage(target, t->image(), source);
m_painter.end();
unlock();
}
}