I'm trying to find a way to draw something on the screen using C++ that would always be on top, even over programs that are in fullscreen.
I want to do this in Windows 10 and so far I've found this https://archive.codeplex.com/?p=DirectDrawOverlayLib
On Windows OS, you can get the device context of the screen then draw to this device. Here is an example :
#include <iostream>
#include <Windows.h>
#include <WinUser.h>
int main()
{
std::cout << "Drawing a line on screen during ten seconds ....\n";
for (int i = 0; i < 100; i++) {
HDC hdc = GetDC(NULL);
MoveToEx(hdc, 0, 0, NULL);
LineTo(hdc, 1000, 1000);
ReleaseDC(NULL, hdc);
Sleep(100);
}
}
Related
I want the code below to take a screenshot of a specified window only, becuse this way BitBlt is faster.
The code below takes a screenshot of a window, specified by the name of window, loads the pixel data in a buffer and then re-draws the picture on your screen just to prove the copying worked.
I'd like to take screenshots of browser windows like Google Chrome, but it doesn't seem to work. It draws a black rectangle on my screen with the correct dimensions of the window.
Seems to be working all the time with the window of Minecraft.
Also worked with the Tor Browser.
I noticed that after querying window info, minecraft's window had no child-windows, but when it didn't work with the others, all of them had multiple child-windows.
#include<iostream>
#include<Windows.h>
#include<vector>
using namespace std;
int main() {
HWND wnd = FindWindow(NULL, "Minecraft 1.15.1");//Name of window to be screenshoted
if (!wnd) {
std::cout << "e1\n";
std::cin.get();
return 0;
}
WINDOWINFO wi = { 0 };
wi.cbSize = sizeof(WINDOWINFO);
GetWindowInfo(wnd, &wi);
RECT rect;
GetWindowRect(wnd, &rect);
int width = wi.rcClient.right - wi.rcClient.left;
int height = wi.rcClient.bottom - wi.rcClient.top;
cout << width << height;
/////Fetch window info^^^^^^^^
BYTE* ScreenData = new BYTE[4 * width*height];
// copy screen to bitmap
HDC hScreen = GetWindowDC(wnd);
HDC hDC = CreateCompatibleDC(hScreen);
HBITMAP hBitmap = CreateCompatibleBitmap(hScreen, width, height);
HGDIOBJ old_obj = SelectObject(hDC, hBitmap);
BitBlt(hDC, 0, 0, width, height, hScreen, 0, 0, SRCCOPY);
SelectObject(hDC, old_obj);
BITMAPINFO bmi = { 0 };
bmi.bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
std::cout << GetDIBits(hDC, hBitmap, 0, 0, NULL, &bmi, DIB_RGB_COLORS)<<endl;
//std::cout << bmi.bmiHeader.biHeight<< " "<< bmi.bmiHeader.biWidth<<endl;
BYTE*buffer = new BYTE[4* bmi.bmiHeader.biHeight*bmi.bmiHeader.biWidth];
bmi.bmiHeader.biHeight *= -1;
std::cout<<GetDIBits(hDC, hBitmap, 0, bmi.bmiHeader.biHeight, buffer, &bmi, DIB_RGB_COLORS)<<endl;//DIB_PAL_COLORS
/////Load window pixels into a buffer^^^^^^^^
POINT p;
int r = 0;
int g = 0;
int b = 0;
int x = 0;
int y = 0;
HDC sc = GetDC(NULL);
COLORREF color;
while (true) {
if ((y*-1) == bmi.bmiHeader.biHeight+1 && x == bmi.bmiHeader.biWidth-1) { break; }
r = (int)buffer[4 * ((y*bmi.bmiHeader.biWidth) + x)+2];
g = (int)buffer[4 * ((y*bmi.bmiHeader.biWidth) + x)+1];
b = (int)buffer[4 * ((y*bmi.bmiHeader.biWidth) + x) ];
color = RGB(r, g, b);
SetPixel(sc, x, y, color);
x++;
if (x == bmi.bmiHeader.biWidth) {
y++;
x = 0;
}
}
/////Prove that the copying was successful and buffer is full^^^^^^^^
Sleep(5000);
std::cout << "fin\n";
std::cin.get();
return 0;
}
I think the problem is the order you're doing things.
Before you put the bitmap on the clipboard, you should select it out of your memory device context.
Once you put the bitmaps on the clipboard, you no longer own it, so you shouldn't try to delete it.
The best way to do that is probably to move your // clean up section before the // save bitmap to clipboard section and to eliminate your DelectObject(hBitmap) statement. So the tail end of your code should probably be:
BitBlt(hDC, 0, 0, width, height, hScreen, 0, 0, SRCCOPY);
// clean up
SelectObject(hDC, old_obj); // selects hBitmap out of hDC
DeleteDC(hDC);
ReleaseDC(NULL, hScreen);
// save bitmap to clipboard
OpenClipboard(NULL);
EmptyClipboard();
SetClipboardData(CF_BITMAP, hBitmap); // clipboard now owns the bitmap
CloseClipboard();
If you still have a problem after those changes, I would check the return value of the SetClipboardData call. If that's failing, GetLastError may give a clue.
New here. Okay so in my project I have a 20x70 2d array, but I can't see it all using the usual 2d array printing. By "see it all" I mean my console is too small. Is there any way of printing the full matrix and see it in the console? Or is there any library that could help me printing it like in a canvas?
Edit: I'm using Codeblocks as my IDE and working in a Windows Console. I've searched on google for some time and I didn't find an answer to this question. I found only answers to how to print a 10x10 2d array.
You can try to resize the console window as described here:
#include <iostream>
//the following line is necessary for the
// GetConsoleWindow() function to work!
//it basically says that you are running this
// program on Windows 2000 or higher
#define _WIN32_WINNT 0x0500
//it is important that the above line be typed
// BEFORE <windows.h> is included
#include <windows.h>
using namespace std;
int main (void)
{
HWND console = GetConsoleWindow();
RECT r;
GetWindowRect(console, &r); //stores the console's current dimensions
//MoveWindow(window_handle, x, y, width, height, redraw_window);
MoveWindow(console, r.left, r.top, 800, 600, TRUE);
for (int j = 0; j < 100; ++j)
{
for (int i = 0x41; i < 0x5B; ++i)
cout << (char)i;
}
cout << endl;
Sleep(1000);
MoveWindow(console, r.left, r.top, r.right - r.left, r.bottom - r.top, TRUE);
}
I wanna draw some pixels with mouse and i wanna do it in C++ just like in Paint but so far i couldnt get it working.
What i have done:
// NewPaint.cpp : main project file.
#include "stdafx.h"
#include <windows.h>
#include <iostream>
#include <conio.h>
#include <dos.h>
#include <Winuser.h>
int main()
{
SetConsoleTitle(L"PaintER"); // Set text of the console so you can find the window
HWND hWnd = GetConsoleWindow(); // Get the HWND
HDC hdc = GetDC(hWnd); // Get the DC from that HWND
textbackgroundcolor(WHITE); // Background color
textcolor(BLACK); // Text color
POINT p; // Pointer
while(1) // Smt like mouse button 1
{
if (ScreenToClient(hWnd, &p)) // Mouse position
{
SetPixel(hdc, p.x, p.y, RGB(0,0,0)); // Draw black pixels
}
}
ReleaseDC(hWnd, hdc); // Release the DC
DeleteDC(hdc); // Delete the DC
system("pause");
return 0;
}
I am using Visual Studio 2010 for this and using Windows 7. I heard these could be inportant infos..
I dont have i only got GDI+ references so if you know a way to use it for this please let me know.
I need coding examples and explanations with it as i've given in my code.
In C++ using Code::Blocks v10.05, how do I draw a single pixel on the console screen? Is this easy at all, or would it be easier to just draw a rectangle? How do I color it?
I'm sorry, but I just can't get any code from SOF, HF, or even cplusplus.com to work. This is for a Super Mario World figure on the screen. The game I think is 16-bit, and is for the SNES system. C::B says I need SDK for C::B. It says "afxwin.h" doesn't exist. Download maybe?
This is what I'm trying to make:
It depends on your OS. I suppose you are programming in a Windows platform, therefore you can use SetPixel but you have to use "windows.h" to get a console handle, so here an example for drawing the cos() function:
#include<windows.h>
#include<iostream>
#include <cmath>
using namespace std;
#define PI 3.14
int main()
{
//Get a console handle
HWND myconsole = GetConsoleWindow();
//Get a handle to device context
HDC mydc = GetDC(myconsole);
int pixel =0;
//Choose any color
COLORREF COLOR= RGB(255,255,255);
//Draw pixels
for(double i = 0; i < PI * 4; i += 0.05)
{
SetPixel(mydc,pixel,(int)(50+25*cos(i)),COLOR);
pixel+=1;
}
ReleaseDC(myconsole, mydc);
cin.ignore();
return 0;
}
You can also use some others libraries like: conio.h allegro.h sdl, etc.
If you're willing to have the image look blocky, you could take advantage of the block characters from the console code page.
█ = '\xDB' = U+2588 FULL BLOCK
▄ = '\xDC' = U+2584 LOWER HALF BLOCK
▀ = '\xDF' = U+2580 UPPER HALF BLOCK
and space
By using the half-blocks in combination with colored text, you can turn an 80×25 console window into an 80×50 16-color display. (This was the approach used by the QBasic version of Nibbles.)
Then, you just need to convert your image to the 16-color palette and a reasonably small size.
windows.h provides a function SetPixel() to print a pixel at specified location of a window. The general form of the function is
SetPixel(HDC hdc, int x, int y, COLORREF& color);
where, x and y are coordinates of pixel to be display and color is the color of pixel.
Important: to print the pixel in your machine with Code::blocks IDE, add a link library libgdi32.a (it is usually inside MinGW\lib ) in linker setting.
Console is a text device, so in general you don't write to individual pixels. You can create a special font and select it as a font for console, but it will be monochromatic. There are libraries which simplify writing console UI (e.g. Curses), but I believe that you also have more gamelike functionality in mind besides just showing a sprite.
if you want to write a game, I suggest taking a look at some of the graphics/game frameworks/libs, e.g. SDL
I have drawn the straight line using windows.h in code::blocks. I can't explain it in details, but I can provide you a code and procedure to compile it in code::blocks.
go to setting menu and select compiler and debugger.
Click on linker tab and add a link library libgdi32.a which is at C:\Program Files\CodeBlocks\MinGW\lib directory.
Now compile this program
#include <windows.h>
#include <cmath>
#define ROUND(a) ((int) (a + 0.5))
/* set window handle */
static HWND sHwnd;
static COLORREF redColor=RGB(255,0,0);
static COLORREF blueColor=RGB(0,0,255);
static COLORREF greenColor=RGB(0,255,0);
void SetWindowHandle(HWND hwnd){
sHwnd=hwnd;
}
/* SetPixel */
void setPixel(int x,int y,COLORREF& color=redColor){
if(sHwnd==NULL){
MessageBox(NULL,"sHwnd was not initialized !","Error",MB_OK|MB_ICONERROR);
exit(0);
}
HDC hdc=GetDC(sHwnd);
SetPixel(hdc,x,y,color);
ReleaseDC(sHwnd,hdc);
return;
// NEVERREACH //
}
void drawLineDDA(int xa, int ya, int xb, int yb){
int dx = xb - xa, dy = yb - ya, steps, k;
float xIncrement, yIncrement, x = xa, y = ya;
if(abs(dx) > abs(dy)) steps = abs(dx);
else steps = abs(dy);
xIncrement = dx / (float) steps;
yIncrement = dy / (float) steps;
setPixel(ROUND(x), ROUND(y));
for(int k = 0; k < steps; k++){
x += xIncrement;
y += yIncrement;
setPixel(x, y);
}
}
/* Window Procedure WndProc */
LRESULT CALLBACK WndProc(HWND hwnd,UINT message,WPARAM wParam,LPARAM lParam){
switch(message){
case WM_PAINT:
SetWindowHandle(hwnd);
drawLineDDA(10, 20, 250, 300);
break;
case WM_CLOSE: // FAIL THROUGH to call DefWindowProc
break;
case WM_DESTROY:
PostQuitMessage(0);
return 0;
default:
break; // FAIL to call DefWindowProc //
}
return DefWindowProc(hwnd,message,wParam,lParam);
}
int WINAPI WinMain(HINSTANCE hInstance,HINSTANCE hPrevInstance,LPSTR lpCmdLine,int iCmdShow){
static TCHAR szAppName[] = TEXT("Straight Line");
WNDCLASS wndclass;
wndclass.style = CS_HREDRAW|CS_VREDRAW ;
wndclass.lpfnWndProc = WndProc ;
wndclass.cbClsExtra = 0 ;
wndclass.cbWndExtra = 0 ;
wndclass.hInstance = hInstance ;
wndclass.hIcon = LoadIcon (NULL, IDI_APPLICATION) ;
wndclass.hCursor = LoadCursor (NULL, IDC_ARROW) ;
wndclass.hbrBackground = (HBRUSH) GetStockObject (WHITE_BRUSH) ;
wndclass.lpszMenuName = NULL ;
wndclass.lpszClassName = szAppName ;
// Register the window //
if(!RegisterClass(&wndclass)){
MessageBox(NULL,"Registering the class failled","Error",MB_OK|MB_ICONERROR);
exit(0);
}
// CreateWindow //
HWND hwnd=CreateWindow(szAppName,"DDA - Programming Techniques",
WS_OVERLAPPEDWINDOW,
CW_USEDEFAULT,
CW_USEDEFAULT,
CW_USEDEFAULT,
CW_USEDEFAULT,
NULL,
NULL,
hInstance,
NULL);
if(!hwnd){
MessageBox(NULL,"Window Creation Failed!","Error",MB_OK);
exit(0);
}
// ShowWindow and UpdateWindow //
ShowWindow(hwnd,iCmdShow);
UpdateWindow(hwnd);
// Message Loop //
MSG msg;
while(GetMessage(&msg,NULL,0,0)){
TranslateMessage(&msg);
DispatchMessage(&msg);
}
/* return no error to the operating system */
return 0;
}
In this program I have used DDA line drawing algorithm. Pixel drawing tasks is done by setPixel(ROUND(x), ROUND(y)) function.
This is windows programing which you can learn details here
To use in CodeBlocks I found this (you have to add a linker option -lgdi32):
//Code Blocks: Project Build Options Linker settings Othoer linker options: add -lgdi32
I forgot: You have to put this before including windows.h:
#define _WIN32_WINNT 0x0500
The whole cosine code again. Ready to compile:
//Code Blocks: Project Build Options Linker settings Othoer linker options: add -lgdi32
#define _WIN32_WINNT 0x0500
#include "windows.h"
#include <iostream>
#include <cmath>
using namespace std;
#define PI 3.14
int main(){
HWND myconsole = GetConsoleWindow();
HDC mydc = GetDC(myconsole);
int pixel =0;
COLORREF COLOR= RGB(255,255,255);
//Draw pixels
for(double i = 0; i < PI * 4; i += 0.05)
{
SetPixel(mydc,pixel,(int)(50+25*cos(i)),COLOR);
pixel+=1;
}
ReleaseDC(myconsole, mydc);
cin.ignore();
return 0;
}
I would like to get the RGB values of a pixel at different x, y coordinates on the screen.
How would I go about this in C++?
I'm trying to create my own gaussian blur effect.
This would be in Windows 7.
Edit
What libraries need to be included for this to run?
What I have going:
#include <iostream>
using namespace std ;
int main(){
HDC dc = GetDC(NULL);
COLORREF color = GetPixel(dc, 0, 0);
ReleaseDC(NULL, dc);
cout << color;
}
You can use GetDC on the NULL window to get a device context for the whole screen, and can follow that up with a call to GetPixel:
HDC dc = GetDC(NULL);
COLORREF color = GetPixel(dc, x, y);
ReleaseDC(NULL, dc);
Of course, you'd want to only acquire and release the device context once while doing all the pixel-reading for efficiency.
As mentioned in a previous post, you want the GetPixel function from the Win32 API.
GetPixel sits inside gdi32.dll, so if you have a proper environment setup, you should be able to include windows.h (which includes wingdi.h) and you should be golden.
If you have a minimal environment setup for whatever reason, you could also use LoadLibrary on gdi32.dll directly.
The first parameter to GetPixel is a handle to the device context, which can be retrieved by calling the GetDC function(which is also available via <windows.h>).
A basic example that loads GetPixel from the dll and prints out the color of the pixel at your current cursor position is as follows.
#include<windows.h>
#include<stdio.h>
typedef WINAPI COLORREF (*GETPIXEL)(HDC, int, int);
int main(int argc, char** argv)
{
HINSTANCE _hGDI = LoadLibrary("gdi32.dll");
if(_hGDI)
{
while(true) {
GETPIXEL pGetPixel = (GETPIXEL)GetProcAddress(_hGDI, "GetPixel");
HDC _hdc = GetDC(NULL);
if(_hdc)
{
POINT _cursor;
GetCursorPos(&_cursor);
COLORREF _color = (*pGetPixel) (_hdc, _cursor.x, _cursor.y);
int _red = GetRValue(_color);
int _green = GetGValue(_color);
int _blue = GetBValue(_color);
printf("Red: 0x%02x\n", _red);
printf("Green: 0x%02x\n", _green);
printf("Blue: 0x%02x\n", _blue);
}
FreeLibrary(_hGDI);
}
}
return 0;
}