CDC Help for color an object - visual-studio

hi
There is a function in C++ Visual studio CDC::ExtFloodFill(int x, int y, COLORREF crColor,UINT nFillType);
my question is that what we suppose to write in place of
int x , int y, COLORREF crColor, UINT nFillType
Like if I have a Object Which I want to Color How to do it
enter code here
#include "afxwin.h"
class fr : public CFrameWnd
{
public:
CPoint st;
CPoint en;
fr()
{
Create(0,"First Frame");
}
//////////////////////
void OnLButtonDown(UINT fl,CPoint p )
{
st.x=p.x;
st.y=p.y;
}
//////////////////////
void OnLButtonUp(UINT fl,CPoint r)
{
en.x=r.x;
en.y=r.y;
CClientDC d(this);
d.Ellipse(st.x,st.y,en.x,en.y);
}
void OnRButtonDown(UINT fl,CPoint q)
{
CClientDC e(this);
e.ExtFloodFill(............);
}
DECLARE_MESSAGE_MAP()
};
BEGIN_MESSAGE_MAP(fr,CFrameWnd)
ON_WM_LBUTTONDOWN()
ON_WM_RBUTTONDOWN()
END_MESSAGE_MAP()
class app : public CWinApp
{
public:
int InitInstance()
{
fr*sal;
sal=new fr;
m_pMainWnd=sal;
sal->ShowWindow(1);
return true;
}
};
app a;

For your example, ExtFloodFill (or any other version of FloodFill) isn't really the right choice.
Instead, you normally want to set the current brush to the color/pattern you want, then draw your object (and it'll automatically be filled with the current brush). Let's say, for example, that you want to draw a red ellipse:
CMyView::OnDraw(CDC *pDC) {
CBrush red_brush;
red_brush.CreateSolidBrush(RGB(255, 0, 0));
pDC->SelectObject(red_brush);
pDC->Ellipse(0, 0, 100, 50);
}
Edit: Okay, if you really insist it has to be a flood-fill, and you're doing it in response to a button click, you'd probably do something like this:
void CYourView::OnRButtonDown(UINT nFlags, CPoint point)
{
CClientDC dc(this);
CBrush blue_brush;
blue_brush.CreateSolidBrush(RGB(0, 0, 255));
dc.SelectObject(blue_brush);
dc.ExtFloodFill(point.x, point.y, RGB(0, 0,0), FLOODFILLBORDER);
CView::OnRButtonDown(nFlags, point);
}

Related

AnimateWindow() incorrectly draws background on high dpi (Win10)

I am creating a simple Win32/MFC application with a main window, and a child-window that uses AnimateWindow() to show (slide up) and hide (slide down) the window. When running the application on 100% dpi scaling, everything behaves normally.
I have overridden WM_ERASEBKGND to render a random red color, to demonstrate the effect. As the window slides-down (hide) on every "step" of the animation the "update rectangle" of the background is repainted, exactly where the child-window disappeared, and the background is to become visible again.
However, when changing the dpi-scaling via Windows Settings (in this case to 125%), an incorrect area is redrawn. It appears as if the area passed into OnEraseBkgnd() is still using the 100% dpi-scaling size, but also the position is off: it appears as if the x/y position of the upper-left corner is passed in screen-space instead of client-space. So the redrawn area looks different, depending on where the on the screen the window is positioned.
The white area is where the child window was actually positioned, and where the redraw of the background should actually have happened.
I have confirmed this effect on Win10 (1803 and 1809) and on Win8.1
Is this a bug in the OS, or is there something I can do to avoid the problem - other than not using AnimateWindow()? ShowWindow() (with SW_SHOW or SW_HIDE) works just fine, btw.
Update: added the full source-code to reproduce the issue.
The problem occurs when using no dpi-aware manifest at all, but it also occurs when using <gdiScaling xmlns="http://schemas.microsoft.com/SMI/2017/WindowsSettings">true</gdiScaling>
class CDialogTestApp : public CWinApp
{
virtual BOOL InitInstance();
};
CDialogTestApp theApp;
class CAboutDlg : public CDialogEx
{
public:
CAboutDlg() : CDialogEx(IDD_ABOUTBOX) {}
};
class CDialogTestDlg : public CDialogEx
{
public:
CDialogTestDlg(CWnd* pParent = nullptr) : CDialogEx(IDD_DIALOGTEST_DIALOG, pParent) {}
protected:
virtual BOOL OnInitDialog();
afx_msg BOOL OnEraseBkgnd(CDC* pDC);
afx_msg void OnLButtonUp(UINT nFlags, CPoint point);
DECLARE_MESSAGE_MAP()
private:
CAboutDlg mDialog;
};
BEGIN_MESSAGE_MAP(CDialogTestDlg, CDialogEx)
ON_WM_ERASEBKGND()
ON_WM_LBUTTONUP()
END_MESSAGE_MAP()
BOOL CDialogTestApp::InitInstance()
{
CWinApp::InitInstance();
CDialogTestDlg dlg;
m_pMainWnd = &dlg;
INT_PTR nResponse = dlg.DoModal();
return FALSE;
}
BOOL CDialogTestDlg::OnInitDialog()
{
CDialogEx::OnInitDialog();
mDialog.Create(IDD_ABOUTBOX, this);
return TRUE;
}
BOOL CDialogTestDlg::OnEraseBkgnd(CDC* pDC)
{
COLORREF color = RGB(rand() & 255, 20, 40);
CRect rect;
GetClipBox(pDC->m_hDC, &rect); // retrieve the update-rectangle
CBrush brush(color);
FillRect(pDC->m_hDC, &rect, (HBRUSH)brush.m_hObject);
return TRUE;
}
void CDialogTestDlg::OnLButtonUp(UINT nFlags, CPoint point)
{
if (mDialog.IsWindowVisible())
{
mDialog.AnimateWindow(200, AW_HIDE | AW_SLIDE | AW_VER_POSITIVE);
}
else
{
mDialog.SetWindowPos(&CWnd::wndTop, 0, 50, 0, 0, SWP_NOSIZE);
mDialog.AnimateWindow(200, AW_ACTIVATE | AW_SLIDE | AW_VER_NEGATIVE);
}
CDialogEx::OnLButtonUp(nFlags, point);
}

QDockWidget does not remember floating size and location when IsFloat is toggled

QDockWidget has a feature where you can double click on the title bar and the dock will toggle to a floating window and back to its docked state. The problem is if you move and resize the floating window and then toggle back to the dock and then back to floating again, your position and size are lost.
I've looked for solutions to resize and move a QDockWidget and I've taken a look at the Qt source code for QDockWidget. I've created a small subclass of QDockWidget that appears to solve the problem. It overrides the mouseDoubleClick, resize and move events, filtering out the unwanted "random" resizing and positioning by Qt and stores the information about screen, position and size in a struct that I store in QSettings for persistence between sessions.
// header
#include <QDockWidget>
#include "global.h"
class DockWidget : public QDockWidget
{
Q_OBJECT
public:
DockWidget(const QString &title, QWidget *parent = nullptr);
QSize sizeHint() const;
void rpt(QString s);
struct DWLoc {
int screen;
QPoint pos;
QSize size;
};
DWLoc dw;
bool ignore;
protected:
bool event(QEvent *event);
void resizeEvent(QResizeEvent *event);
void moveEvent(QMoveEvent *event);
};
// cpp
#include "dockwidget.h"
DockWidget::DockWidget(const QString &title, QWidget *parent)
: QDockWidget(title, parent)
{
ignore = false;
}
bool DockWidget::event(QEvent *event)
{
if (event->type() == QEvent::MouseButtonDblClick) {
ignore = true;
setFloating(!isFloating());
if (isFloating()) {
// move and size to previous state
QRect screenres = QApplication::desktop()->screenGeometry(dw.screen);
move(QPoint(screenres.x() + dw.pos.x(), screenres.y() + dw.pos.y()));
ignore = false;
adjustSize();
}
ignore = false;
return true;
}
QDockWidget::event(event);
return true;
}
void DockWidget::resizeEvent(QResizeEvent *event)
{
if (ignore) {
return;
}
if (isFloating()) {
dw.screen = QApplication::desktop()->screenNumber(this);
QRect r = geometry();
QRect a = QApplication::desktop()->screen(dw.screen)->geometry();
dw.pos = QPoint(r.x() - a.x(), r.y() - a.y());
dw.size = event->size();
}
}
QSize DockWidget::sizeHint() const
{
return dw.size;
}
void DockWidget::moveEvent(QMoveEvent *event)
{
if (ignore || !isFloating()) return;
dw.screen = QApplication::desktop()->screenNumber(this);
QRect r = geometry();
QRect a = QApplication::desktop()->screen(dw.screen)->geometry();
dw.pos = QPoint(r.x() - a.x(), r.y() - a.y());
dw.size = QSize(r.width(), r.height());
}
While this appears to be working is there a simpler way to accomplish this? What do I do if a QDockWidget was on a screen that is now turned off?

Stop and Restart game in Processing on button press

I'm currently working on a small game to learn Processing.
I want my game to stop when I press "Stop" and want my game to reset/restart when "Stop" changed to Start.
(When I click on the button, it changes Stop to Start and back to Stop etc etc. (so basically, I have 1 'button')
I'm struggling and can't find a solution on internet / stackoverflow so maybe someone can help me?
(# mousePressed's if else I need the 'stop and restart function')
float x = width/2;
float speed = 2;
boolean textHasBeenClicked = false;
int aantalRaak = 0;
int aantalMis = 0;
int positieText = 20;
void setup() {
background(0);
size(600,500);
}
void draw() {
clear();
move();
display();
smooth();
//Scoreboard bovenaan
fill(255);
textSize(20);
textAlign(LEFT);
text("Aantal geraakt: " + aantalRaak,0, positieText); text("Aantal gemist: " + aantalMis, width/2, positieText);
//button onderaan
fill(0,255,0);
rect(width/2-40, height-40, 100, 50);// draw anyway...
}
void mousePressed() {
// toggle
textHasBeenClicked = ! textHasBeenClicked;
fill(0);
if (textHasBeenClicked) {
// display text 2
textSize(30);
textAlign(CENTER);
text("Stop" , width/2,height-10);
}
else {
// display text 1
textSize(30);
textAlign(CENTER);
text("Start" , width/2,height-10);
}
}
void move() {
x = x + speed;
if (x > width) {
x = 0;
}
}
void display(){
//schietschijf
float y = height/2;
noStroke();
fill(255, 0, 0);
ellipse(x, y, 40, 40);
fill(255);
ellipse(x, y, 30, 30);
fill(255, 0, 0);
ellipse(x, y, 20, 20);
fill(255);
ellipse(x, y, 10, 10);
}
You should try to break your problem down into smaller steps and take those steps on one at a time. You're really asking two questions:
How do I show a stop button that turns into a start button?
How do I reset a sketch?
For the first question, you can create a boolean variable. Use that variable in your draw() function, and then modify that variable in the mousePressed() function.
boolean running = false;
void draw() {
fill(0);
if (running) {
background(255, 0, 0);
text("Stop", 25, 25);
} else {
background(0, 255, 0);
text("Start", 25, 25);
}
}
void mousePressed() {
running = !running;
}
Then for resetting the sketch, you can create a function that changes all of your variables back to their default values. Here's a simple example:
float circleY;
void setup() {
size(100, 500);
}
void draw() {
background(0);
circleY++;
ellipse(width/2, circleY, 20, 20);
}
void reset() {
circleY = 0;
}
void mousePressed() {
reset();
}
Try to work from small examples like this instead of your full program, and post a MCVE if you get stuck. Good luck.
You may consider implementing a while loop. I don't know what library you're using for input, so I can't tell you exactly what to do. But something along the lines of:
while(!InputReceived) {
if(CheckForMouseInput()) // Assuming CheckForMouseInput returns true if input was detected
break // Input was detected, now do stuff based on that.
else {
// Must #include <thread> and #include <chrono>
// Wait a bit...
continue; // Jump back to the top of the loop, effectively restarting it.
}
Would probably suit your needs. That is what I would do, at least. After the loop breaks, the game is effectively restarted, and you can do whatever you need to do based on that.

Show grey level value of pixel in Qt when the mouse is over it

Currently I am working on the display of gray level image with zoom feature. I am able to get the position of the pixel and the zoom feature is working well. However I encountered two problems:
1.) How can I get the grey level value of the pixel that is pointed by the mouse? I only managed to obtain the rgb value through “QRgb rgbValue = pix.toImage().pixel(x,y)”. How can I convert it into grey level value? Or is there any direct way to get the grey level value of the pixel.
2.) I have implemented “mouseMoveEvent(QMouseEvent *event)” and “setMouseTracking(true)”. However the function of “mouseMoveEvent(QMouseEvent *event)” is not functioning when I move the mouse. What is wrong with my code?
mainwindow.h
#ifndef MAINWINDOW_H
#define MAINWINDOW_H
#include <QMainWindow>
#include <QGraphicsScene>
#include <QGraphicsItem>
namespace Ui {
class MainWindow;
}
class MainWindow : public QMainWindow
{
Q_OBJECT
public:
explicit MainWindow(QWidget *parent = 0);
~MainWindow();
private slots:
void on_pushButton_clicked();
protected:
void mouseMoveEvent(QMouseEvent * event);
private:
Ui::MainWindow *ui;
QGraphicsScene* scene;
QGraphicsItem* item;
QPixmap pix;
};
#endif // MAINWINDOW_H
mainwindow.cpp
#include "mainwindow.h"
#include "ui_mainwindow.h"
MainWindow::MainWindow(QWidget *parent) :
QMainWindow(parent),
ui(new Ui::MainWindow)
{
ui->setupUi(this);
QImage image("E:/image_00002.bmp");
pix = QPixmap::fromImage(image);
scene = new QGraphicsScene(this);
ui->graphicsView->setScene(scene);
scene->addPixmap(pix);
}
MainWindow::~MainWindow()
{
delete ui;
}
void MainWindow::on_pushButton_clicked()
{
ui->graphicsView->setMouseTracking(true);
}
void MainWindow::mouseMoveEvent(QMouseEvent *event)
{
QPoint local_pt = ui->graphicsView->mapFromGlobal(event->globalPos());
QPointF img_coord_pt = ui->graphicsView->mapToScene(local_pt);
double x = img_coord_pt.x();
double y = img_coord_pt.y();
/* How can I get a gray level image here */
QRgb rgbValue = pix.toImage().pixel(x,y);
ui->label_X->setText(QString::number(x));
ui->label_Y->setText(QString::number(y));
ui->label_Value->setText(QString::number(rgbValue));
}
To elaborate a bit on what hank said. In order for your QMainWindow to receive the events from the QGraphicsScene you need to install an event filter (see http://qt-project.org/doc/qt-4.8/qobject.html#installEventFilter ). Quoting from the DOCs:
An event filter is an object that receives all events that are sent to this object.
The filter can either stop the event or forward it to this object.
In order to process the events you need to define an eventFilter method in your main window class. Below are the on_pushButton_clicked() and eventFilter() methods that should do what you want:
void MainWindow::on_pushButton_clicked()
{
ui->graphicsView->setMouseTracking(true);
scene->installEventFilter(this);
}
bool MainWindow::eventFilter(QObject *obj, QEvent *event)
{
if (event->type() == QEvent::GraphicsSceneMouseMove ) {
QGraphicsSceneMouseEvent *mouseEvent = static_cast<QGraphicsSceneMouseEvent*>(event);
QPointF img_coord_pt = mouseEvent->scenePos();
double x = img_coord_pt.x();
double y = img_coord_pt.y();
QColor color = QColor(pix.toImage().pixel(x,y));
int average = (color.red()+color.green()+color.blue())/3;
ui->label_X->setText(QString::number(x));
ui->label_Y->setText(QString::number(y));
ui->label_Value->setText(QString::number(average));
return true;
} else {
return QObject::eventFilter(obj, event);
}
}
Does this help?

QGraphicsScene/QGraphicsView performance

I am using QGraphicsScene/QGraphicsView pair in my project
I have performance issue with this pair.
I added my custom graphics items to scene and displayed the contents with view. After that my custom graphics items paint method continuously called by scene(just like infinite loop). This makes %25 of CPU usage(approximately 400 items on scene). What may cause this behaviour?
Here is one my item implementation:
class LevelCrossingItem : public QGraphicsWidget
{
public:
LevelCrossingItem(QString _id,qreal _x,qreal _y);
~LevelCrossingItem();
QRectF boundingRect() const;
QSizeF sizeHint(Qt::SizeHint which, const QSizeF &constraint = QSizeF()) const;
void paint(QPainter *painter, const QStyleOptionGraphicsItem *option, QWidget *widget /* = 0 */);
void readStateBits();
bool isClosed();
bool isGateArmBroken();
bool isOpenedDuringRouteTanzimCompleted();
bool hasDataConsistencyWarning();
int type() const {return Type;}
private slots:
void setVisible(bool);
private:
enum {Type = FIELDLEVELCROSSING};
QString m_id;
QString m_source;
short m_closedState;
short m_brokenGateArmState;
short m_openedDuringRouteTanzimCompletedState;
short m_dataConsistencyWarningState;
QBitArray stateBitArray;
qreal x,y;
QSvgRenderer *renderer;
};
#include "levelcrossing.h"
LevelCrossingItem::LevelCrossingItem(QString _id,qreal _x,qreal _y):m_id(_id),x(_x),y(_y),stateBitArray(4)
{
m_source = LEVELCROSSING_RESOURCE_PATH.arg("open");
renderer = new QSvgRenderer;
setStateArray(stateBitArray);
setZValue(-0.5);
}
LevelCrossingItem::~LevelCrossingItem()
{
delete renderer;
}
void LevelCrossingItem::setVisible(bool visible)
{
QGraphicsItem::setVisible(visible);
}
QRectF LevelCrossingItem::boundingRect() const
{
return QRectF(QPointF(x,y),sizeHint(Qt::PreferredSize));
}
QSizeF LevelCrossingItem::sizeHint(Qt::SizeHint which, const QSizeF &constraint) const
{
return QSizeF(50,270);
}
void LevelCrossingItem::readStateBits()
{
m_closedState = property("Closed").toInt();
m_brokenGateArmState = property("Broken").toInt();
m_openedDuringRouteTanzimCompletedState = property("OpenedOnRouteWarning").toInt();
m_dataConsistencyWarningState = property("DataConsistencyWarning").toInt();
stateBitArray.setBit(0,qvariant_cast<bool>(m_closedState));
stateBitArray.setBit(1,qvariant_cast<bool>(m_brokenGateArmState));
stateBitArray.setBit(2,qvariant_cast<bool>(m_openedDuringRouteTanzimCompletedState));
stateBitArray.setBit(3,qvariant_cast<bool>(m_dataConsistencyWarningState));
setStateArray(stateBitArray);
}
void LevelCrossingItem::paint(QPainter *painter, const QStyleOptionGraphicsItem *option, QWidget *widget )
{
Q_UNUSED(option);
Q_UNUSED(widget);
readStateBits();
m_closedState == Positive ? m_source = LEVELCROSSING_RESOURCE_PATH.arg("closed")
: m_source = LEVELCROSSING_RESOURCE_PATH.arg("open");
m_brokenGateArmState == Positive ? m_source = LEVELCROSSING_RESOURCE_PATH.arg("broken")
: m_source = m_source;
if(m_openedDuringRouteTanzimCompletedState == Positive)
{
setWarningVisible(OOR_WRN.arg(name()).arg(interlockingRegionId()),true);
if(stateChanged())
emit itemAlarmOccured(m_id,LevelCrossingIsOpenDuringTanzimCompleted);
}
else
setWarningVisible(OOR_WRN.arg(name()).arg(interlockingRegionId()),false);
if(m_dataConsistencyWarningState == Positive)
{
setWarningVisible(DC_WRN.arg(name()).arg(interlockingRegionId()),true);
if(stateChanged())
emit itemAlarmOccured(m_id,LevelCrossingDataConsistency);
}
else
setWarningVisible(DC_WRN.arg(name()).arg(interlockingRegionId()),false);
renderer->load(m_source);
renderer->render(painter,boundingRect());
}
bool LevelCrossingItem::isClosed()
{
return m_closedState == Positive;
}
bool LevelCrossingItem::isGateArmBroken()
{
return m_brokenGateArmState == Positive;
}
bool LevelCrossingItem::isOpenedDuringRouteTanzimCompleted()
{
return m_openedDuringRouteTanzimCompletedState == Positive;
}
bool LevelCrossingItem::hasDataConsistencyWarning()
{
return m_dataConsistencyWarningState == Positive;
}
I read x and y coordinates from xml file. For this item x and y coordinates are 239,344 respectively
Most likely your graphics item has mistakes in the implementation. I had similar behavior, but I was unable to figure out what exactly causes it. I suspect that this happens when you draw outside of the bounding rect. This triggers some clean up routine, which in turn causes redraw of the item, and here goes the loop. Eventually I resolved the issue by carefully inspecting the implementation of my custom graphics item and making sure that:
These is no painting outside of the MyGraphicsItem::boundingRect. Use painter->setClipRect(boundingRect())
MyGraphicsItem::shape does not cross with the MyGraphicsItem::boundingRect.
If you override any other QGraphicsItem functions, make sure that your implementation is correct.
Hope this helps. Feel free to post the source code of your graphics item, it will be easier to find the problem.

Resources