I created edge browser window using CreateCoreWebView2Host() method. This method is takes parent window handle and creates child window in which we can navigate the web page. After I am done with navigation I need to return my window handle, which I believe I am failed to return.
On Spy++ I see "Chrome_WidgetWin_0", "Chrome_WidgetWin_1", "Intermediate D3D Window" as child windows to my parent window. which one is the child window handle?I thought I am creating one child window.
I tried by fetching window handles using FindWindowEx() passing above mentioned class names. But still not getting expected results in my project. So I doubt if I am passing correct handle.
Now the question is , How to get the window handle(HWND) for the window created by CreateCoreWebView2Host?
You can get the edge window handle by using GetWindow, passing Handle of WebView2 control to it as the first argument and GW_CHILD as second argument. For exaample:
public const uint GW_CHILD = 5;
[DllImport("user32.dll")]
public static extern IntPtr GetWindow(IntPtr hWnd, uint uCmd);
[DllImport("user32.dll")]
public static extern IntPtr SetFocus(IntPtr hWnd);
WebView2 webView21 = new Microsoft.Web.WebView2.WinForms.WebView2();
private async void Form1_Load(object sender, EventArgs e)
{
webView21.Dock = DockStyle.Fill;
this.Controls.Add(webView21);
await webView21.EnsureCoreWebView2Async();
webView21.Source = new Uri("https://bing.com");
webView21.NavigationCompleted += WebView21_NavigationCompleted;
}
private void WebView21_NavigationCompleted(
object sender, CoreWebView2NavigationCompletedEventArgs e)
{
var child = GetWindow(webView21.Handle, GW_CHILD);
SetFocus(child);
}
Note: You may find this GitHub thread useful. This is the trick that I've used to focus the browser in Windows Forms as well.
The WebView2 SDK doesn't provide this HWND because exactly how the WebView2 connects up to the provided parent HWND is intended to be an implementation detail that could change as the underlying Edge browser or WebView2 Runtime is updated even when you stay on the same version of the WebView2 SDK. We're relying in large part on the browser's logic for rendering and so this may change in the future.
Instead of providing an HWND, the CoreWebView2Host (its been renamed to CoreWebView2Controller in more recent SDK releases) provides various methods for you to focus, zoom, and so on. What are you trying to do with the HWND?
Related
In visual studio,all opened editor windows are tabbed in the workspace.when you right click one,there is a "float" option in the menu.if you float a window,the window is detached from the tab and you can drag it anywhere and after all you can dock it back to the tab.
i created a test mdi project,it's not a default behavior.
i overrided the ID_FILE_NEW command and followed the OpenDocumentFile routine,never found out where the framework add the newly created child frame window to the tab.
in the OpenDocumentFile routine,i bumped into a CMFCTabCtrl class and there is a AddTab function.
now the AddTab function:
virtual void AddTab(
CWnd* pTabWnd,
LPCTSTR lpszTabLabel,
UINT uiImageId = (UINT)-1,,
BOOL bDetachable = TRUE
);
virtual void AddTab(
CWnd* pTabWnd,
UINT uiResTabLabel,
UINT uiImageId = (UINT)-1,
BOOL bDetachable = TRUE
);
the Remarks:
If pTabWnd points to an object that is not derived from the CDockablePane Class and
if bDetachable is TRUE, the framework automatically creates a wrapper for the pTabWnd object.
The wrapper makes the pTabWnd object detachable.
By default, the wrapper is an instance of the CDockablePaneAdapter Class.
If the functionality offered by the default wrapper is unacceptable,
use the CMFCBaseTabCtrl::SetDockingBarWrapperRTC method to specify a different wrapper.
still no luck.
i did try to on the fly create a CView and attach it to the CDocument and put that CView in a CDockablePane.so i could switch the old tabbed one to the new dockable one. it's not working yet.
So any thoughts about floating the tabbed CMDIChildWndEx window from CMDIFrameWndEx?
Is there a way I can get a HWND by it's property "name"? I know that every IDE has its own properties for HWND elements but those properties are applied to the HWND.
I'm not working in Visual Studio, this is just a case. I want to get HWNDs by Name in C++ without VS Libraries.
For example:
HWND button = GetHwndByName("button1"); //Example
Property "name" is button1
I'm going to assume you're either trying to access your GUI controls in code or some other program's GUI controls.
As some people have mentioned, the (name) property in the properties editor is just the variable name used for that control. Your screenshot shows Visual Studio editing a .net program. In the case of .net, the (name) field is the name of the class member of the window class that represents the control. So if (name) is button1 then Visual Studio might generate code like
// pseudo-C++/C#-like
class Form1 : public System.Windows.Forms.Form {
private:
System.Windows.Forms.Button *button1;
...
};
The idea here is that you would have event handlers as part of your Form1 class:
void Form1::onButton1Clicked(void)
{
this->button1->SetText("You clicked me!");
}
As such, the (name) is not an intrinsic property of the window from Windows's point of view.
I don't know what CA Plex's GUI editor looks like, but I would assume, given you said you were using C++, that it either
a) produces a class like the one I pasted above, in which case you would just use the (name) directly as members, or
b) produces a header file with each of those control names as global HWND variables
Either way, you can just use them directly from within your code. Perhaps have something like
void doToAllButtons(void (*f)(HWND, LPARAM), LPARAM lParam)
{
(*f)(button1, lParam);
(*f)(button2, lParam);
(*f)(button3, lParam);
}
and simply write an appropriate function to call via this one.
If you need to interface with another program and want to use its variable names, then you're out of luck. You'll need to find the windows you want some other way, such as with FindWindow().
I have followed this link to get the window handle of a ActiveX control
Sample Code from microsoft's site
// The following code should return the actual parent window of the ActiveX control.
HWND CMyOleControl::GetActualParent()
{
HWND hwndParent = 0;
// Get the window associated with the in-place site object,
// which is connected to this ActiveX control.
if (m_pInPlaceSite != NULL)
m_pInPlaceSite->GetWindow(&hwndParent);
return hwndParent; // Return the in-place site window handle.
}
But in my case I keep finding that "m_pInPlaceSite" is always NULL. I'm trying to run this code in my controls FinalConstruct. Is there something else I need to implement for the m_pInPlaceSite to be given a value? Or do I need to Query to get the value.
Thanks
FinalConstruct is way too early. In FinalConstruct your class is just being created and is not yet initialized. There is no "in place" site, there is no yet site at all.
Your control will be called by its owner, it will be given its site, then activated - only then you will possibly have m_pInPlaceSite available.
I've seen that some apps (maybe not .NET apps) that have an extra button on the left from the minimize button on the form's title bar? How can I achieve this in C#?
UPDATE: Added a solution that will work with Aero enabled for Windows Vista and Windows 7
***Non-Aero Solution***
The non-client area of a window interaction is managed by a series of non-client specfic messages. For example WM_NCPAINT message is sent to the window procedure to paint the non-client area.
I have never done this from .NET, but I suspect you can overide the WndProc and handle the WM_NC* messages to achieve what you want.
Update: Since I never tried this from .NET I got a few minutes and thought I would give it a quick try.
Trying this on Windows 7, I found that I needed to disable the Themes for the Window if I wanted to OS to do the base rendering of the non-client area. So here is a short test. I used GetWindowDC to get the DC of the entire window rather than GetDCEx, that was just because I could interop that from memory and did not have lookup all the flag constants for GetDcEx. And of course the code could do with more error checking.
using System;
using System.Drawing;
using System.Windows.Forms;
using System.Runtime.InteropServices;
namespace WindowsFormsApplication1
{
public partial class CustomBorderForm : Form
{
const int WM_NCPAINT = 0x85;
[DllImport("user32.dll", SetLastError = true)]
public static extern IntPtr GetWindowDC(IntPtr hwnd);
[DllImport("user32.dll", SetLastError = true)]
public static extern int ReleaseDC(IntPtr hwnd, IntPtr hdc);
[DllImport("user32.dll", SetLastError = true)]
public static extern void DisableProcessWindowsGhosting();
[DllImport("UxTheme.dll", SetLastError = true, CharSet = CharSet.Unicode)]
public static extern IntPtr SetWindowTheme(IntPtr hwnd, string pszSubAppName, string pszSubIdList);
public CustomBorderForm()
{
// This could be called from main.
DisableProcessWindowsGhosting();
InitializeComponent();
}
protected override void OnHandleCreated(EventArgs e)
{
SetWindowTheme(this.Handle, "", "");
base.OnHandleCreated(e);
}
protected override void WndProc(ref Message m)
{
base.WndProc(ref m);
switch (m.Msg)
{
case WM_NCPAINT:
{
IntPtr hdc = GetWindowDC(m.HWnd);
using (Graphics g = Graphics.FromHdc(hdc))
{
g.FillEllipse(Brushes.Red, new Rectangle((Width-20)/2, 8, 20, 20));
}
ReleaseDC(m.HWnd, hdc);
}
break;
}
}
}
}
Btw. I called DisableProcessWindowsGhosting, this will stop the OS from drawing the non-client area if the application takes too long to respond to windows messages. If you do not do this, then in some situations the border will be renderd but your adornments will not be shown. So that depends on your requirements it that is right for you or not.
***Aero supported solution***
Prompted by the comment from #TheCodeKing, I thought I would take another look at this. It turns out this can be done in a fully documented way while supporting Aero. But it is not for the faint of heart. I will not provide a complete solution here, there are still some kinks to workout, but it does the basics.
This code/solution is based off the Win32 example which can be found at the following location
http://msdn.microsoft.com/en-us/library/bb688195(VS.85).aspx
In principal what you need to do is the following.
Extend the client area of the window to cover the Frame. This is done by handling the WM_NCCALCSIZE message and returning 0. This gives the Non-Client area a size of 0 and therefore the client area now covers the entire window.
Extend the Frame into the client area using DwmExtendFrameIntoClientArea. This gets the OS to render the Frame over the client area.
The above steps will give you a windows with the standard glass frame excluding the system menu (Window Icon) and the title. The minimize, maximize and close buttons will still be drawn and will work. What you will not be able to do is drag or resize the window, this is because the frame is not really there, remember the client area covers the whole window, we have just asked the OS to draw the frame onto the client area.
Now you can draw on the window as normal, even on top of the frame. You can even put controls in the caption area.
Finally, allow the DWM to handle hit-testing for you, by calling DwmDefWindowProc from your WndProc (before you've processed it). It returns a boolean indicating whether the DWM handled the message for you.
Simple Solution:
Step 1: Create a Windows Form (this will be your custom title bar)
-Set Form Border Style to None
-Add whatever controls you would like to this
-I will name this custom form "TitleBarButtons"
Step 2. In the from that you want to use this custom control in add
titleBarBtn = new TitleBarButtons();
titleBarBtn.Location = new Point(this.Location.X + 100, this.Location.Y+5);
titleBarBtn.Show();
titleBarBtn.Owner = this;
To your constructor... you can mess with the offsets this just fit in a nice position for my app
Step 3. Add the move event to your main form
private void Form14_Move(object sender, EventArgs e)
{
titleBarBtn.Location = new Point(this.Location.X + 100, this.Location.Y+5);
}
Please let me know if you would like a better explanation of any of the above code.
I think a way to do this would be to handle WM_NCPAINT message (non-client paint) to draw the button, and to handle non-client mouse clicks to know someone clicked on the "button".
I am creating an MFC application which will be launched on click on Explorer Context (Rightclick) menu.
But I need to launch only single instance of the application. For that I have to use FindWindow and AfxRegisterClass
I tried to register the class in my MFC app as below:
BOOL CNDSClientDlg::InitInstance()
{
//Register Window Updated on 16th Nov 2010, #Subhen
// Register our unique class name that we wish to use
WNDCLASS wndcls;
memset(&wndcls, 0, sizeof(WNDCLASS));
//Class name for using FindWindow later
wndcls.lpszClassName = _T("NDSApp");
// Register new class and exit if it fails
if(!AfxRegisterClass(&wndcls)) // [C]
{
return FALSE;
}
}
and called the method in the constructor of the MFC class. I verified that the class is being registered while I am starting the application.
Now in my shell Extension I am trying to find the Class registered in my MFC as below:
CWnd *pWndPrev = NULL;
pWndPrev = CWnd::FindWindow(_T("NDSApp"),NULL);
if(pWndPrev != NULL)
pWndPrev->BringWindowToTop();
But I am not able to get the CWnd to Window. Not able to figure it out. Please let me know if I am missing something or doing something wrong.
FindWindow finds window instances not window classes. In your app which registers the class you need to actually create a window so that the extension can find that window.
(Finding the window by class name is fine; the problem is you haven't actually created anything to find.)
Also, I suspect if you try to create a window based on the window-class you've registered it will fail because you've left most of the WNDCLASS structure null. See the example you linked to for better default values. (e.g. You must provide a wndproc and hinstance.)