Click here to Skip to main content
65,938 articles
CodeProject is changing. Read more.
Articles
(untagged)

Screen and Form capture with Managed C++

0.00/5 (No votes)
3 Jul 2002 1  
This is further demonstration of MC++ IJW capabilities

Introduction

I was continuing to play around with IJW. And when I realized that there was no direct Screen Capture method in .NET, I thought I might as well try to make a simple Screen Capturing program. Of course since I knew as much about GDI programming as some of these Canadians seem to know about Cricket, you might as well have guessed that I found things more than a little difficult. Luckily Tom Archer, James Johnson and Neil Van Note promptly answered my questions.

The sample program allows you to capture the entire screen or only the current form and also allows you to save the image, though here I have hard coded the filename and path, as I was too lazy to show a save dialog. If you repeatedly click on the capture buttons, you get a weird effect of cascading screen captures, one on top of the other as shown in the screenshot of the sample program.

Quick Rundown

Using GetDC or GetWindowDC depending on whether we are capturing the screen or the current window, we get our HDC using which we call CreateCompatibleDC to create our memory DC. Using a call to GetSystemMetrics we get the size of the screen and then call CreateCompatibleBitmap to create a bitmap. Then we select this bitmap into the memory DC using a call to SelectObject and save the old bitmap. Using BitBlt we copy the screen or window contents into this bitmap which is now selected into the memory DC. Then we select the original bitmap back into the DC, delete the memory DC, release the DC, and delete the bitmap. To be frank, there is not much .NET stuff involved in the code, but imagine if you had to write all this using C#. Think of all the P/Invoke stuff you'd have to do and the ugly declarations you'd have had to write for the API calls. With MC++ you are saved all that bother.

Full Source Listing

The source code is self-explanatory. Notice how I have written my WinMain. This is because I am including windows.h where WinMain is already prototyped in that fashion.

#include "stdafx.h"


#using <mscorlib.dll>
#using <System.dll>
#using <System.Drawing.dll>
#using <System.Windows.Forms.dll>

#include <tchar.h>

#include <windows.h>


using namespace System;
using namespace System::ComponentModel;
using namespace System::Drawing;
using namespace System::Windows::Forms;

public __gc class NForm : public Form
{
public:
    Button *btn;
    Button *btn2;
    Button *btn3;
    PictureBox *pbox;
    void btn_Click(Object *sender, System::EventArgs* e); 
    void CaptureScreen(bool FullScreen);

    NForm()
    {
        this->StartPosition = FormStartPosition::CenterScreen;
        this->Text = "Capture screen - Nish for CodeProject -IJW";
        this->Size = Drawing::Size(750,550);
        this->FormBorderStyle = System::Windows::Forms::FormBorderStyle::FixedDialog; 

        btn=new Button();
        btn->Text="Capture Screen";
        btn->Location=Point(5,5);
        btn->Size=Drawing::Size(100,30);
        btn->add_Click(new EventHandler(this,&NForm::btn_Click)); 
        this->Controls->Add(btn);

        btn2=new Button();
        btn2->Text="Capture Form";
        btn2->Location=Point(125,5);
        btn2->Size=Drawing::Size(100,30);
        btn2->add_Click(new EventHandler(this,&NForm::btn_Click)); 
        this->Controls->Add(btn2);

        btn3=new Button();
        btn3->Text="Save C:\\Z.Jpg";
        btn3->Location=Point(250,5);
        btn3->Size=Drawing::Size(100,30);
        btn3->add_Click(new EventHandler(this,&NForm::btn_Click)); 
        this->Controls->Add(btn3);

        pbox = new PictureBox();
        pbox->Location=Point(5,45);
        pbox->Size=Drawing::Size(730,470);
        pbox->BorderStyle = BorderStyle::Fixed3D;
        pbox->SizeMode = PictureBoxSizeMode::StretchImage;


        this->Controls->Add(pbox);
    }
};

int __stdcall WinMain(HINSTANCE,HINSTANCE,LPSTR,int)
{
    Application::Run(new NForm());
    return 0;
}

void NForm::btn_Click(Object *sender, System::EventArgs *e)
{
    if(sender->Equals(btn))
        CaptureScreen(true);
    else
    {
        if(sender->Equals(btn2))
            CaptureScreen(false);
        else
            pbox->Image->Save("C:\\Z.Jpg",
                Drawing::Imaging::ImageFormat::Jpeg);
    }
}

void NForm::CaptureScreen(bool FullScreen)
{
    HDC hDC;
    if(FullScreen)
        hDC = GetDC(NULL); //Get full screen

    else
    {
        HWND hWnd=(HWND)this->Handle.ToInt32(); //Get HWND of form

        hDC = GetWindowDC(hWnd); //Now get it's DC handle 

    }
    HDC hMemDC = CreateCompatibleDC(hDC);
    RECT r;
    GetWindowRect((HWND)this->Handle.ToInt32(),&r); //need this for Form

    SIZE size;
    if(FullScreen)
    { 
        size.cx = GetSystemMetrics(SM_CXSCREEN);
        size.cy = GetSystemMetrics(SM_CYSCREEN);
    } 
    else
    {
        size.cx = r.right-r.left;
        size.cy = r.bottom-r.top;
    }

    //most of the remaining code is normal GDI stuff


    HBITMAP hBitmap = CreateCompatibleBitmap(hDC, size.cx, size.cy);
    if (hBitmap)
    {
        HBITMAP hOld = (HBITMAP) SelectObject(hMemDC, hBitmap);
        BitBlt(hMemDC, 0, 0, size.cx, size.cy, hDC, 0, 0, SRCCOPY);
        SelectObject(hMemDC, hOld);
        DeleteDC(hMemDC);
        ReleaseDC(NULL, hDC);
        pbox->Image = Image::FromHbitmap(hBitmap); 
        DeleteObject(hBitmap);
    }
}

History

  • May 7 02 - Posted the article
  • Jul 4 02 - Added sample VC++ project

License

This article has no explicit license attached to it but may contain usage terms in the article text or the download files themselves. If in doubt please contact the author via the discussion board below.

A list of licenses authors might use can be found here