Re: How can I write a program in VC++6 such that only ONE instance of - C / C++ / MFC Discussion Boards

Read my article:
How to get CPU usage by performance counters

Can someone tell me how to make my program switch between monitors, i.e. make a new monitor the active one? Also how do I split one monitor into two spaces, then swap between them. I would think all this is related. Splitting up the desktop, then activating a part of it.

chris

Anyone got an idea how to do a line counter such as we can see on this screenshot:

http://www.htmledit.com/images/htmledit_logiciel.gif[^]

Thanks a lot!

Can I block keyboard events such as keypress or key down of any key in a program?

The program will be controlled by a mouse only.

Thanks

JW

DJ

You can try one of these methods :-

(1) Handle all the keyboard messages and dont call the default handler

or

(2) Set a local keyboard hook ( WH_KEYBOARD ) and dont call the next hook; and return a non-zero value

Nish

Author of the romantic comedy

Summer Love and Some more Cricket [New Win]

Review by Shog⁹
Click here for review[NW]

this program can be run at any time?

I customized media Player based on Windows Midia player control. However, the media player could not be run exclusively, that means if I try to run two of them, each one will open a media file and play simultaneously.

How can I program so that only one media player program can run at any time? Like using Windows MEdia Player, if you try to run a second movie file, the Media Player will close the first media file and turn to start running the second media file?

The platforms are VC++6 and WIndows 2000 Pro.

Appreciate your help!

JW

DJ

See the following articles:

http://www.codeproject.com/useritems/Single_Instance.asp

http://www.codeproject.com/cpp/singletonapp.asp

http://www.codeproject.com/cpp/csingleinst.asp

http://www.codeproject.com/cpp/avoidmultinstance.asp

http://www.codeproject.com/csharp/cssingprocess.asp

Excerpt from MSDN:
You will find this in the file msdn_mfcfaq50.htm (MFC FAQ).

---------------
How do I limit my MFC application to one instance?

Look at the Microsoft C++ sample ONETIME.

In brief:

const char* MyMainWndClassName = "MyMainWndXQW"
BOOL CMyApp::InitApplication()
{
// Call base class. Default version does nothing.
CWinApp::InitApplication();
WNDCLASS wndcls;
// Start with NULL defaults.
memset(&wndcls, 0, sizeof(WNDCLASS));
// Get class information for default window class.
::GetClassInfo(AfxGetInstanceHandle(),"AfxFrameOrView",&wndcls);
// Substitute unique class name for new class.
wndcls.lpszClassName = MyMainWndClassName;
// Register new class and return the result code.
return ::RegisterClass(&wndcls);
}

And:

BOOL CMyApp::FirstInstance()
{
CWnd *PrevCWnd, *ChildCWnd;
// Determine if another window with our class name exists.
PrevCWnd = CWnd::FindWindow(MyMainWndClassName, NULL);
if (PrevCWnd != NULL)
{
// If so, does it have any pop-ups?
ChildCWnd=PrevCWnd->GetLastActivePopup();
// Bring the main window to the top.
PrevCWnd->BringWindowToTop();
// If iconic, restore the main window.
if (PrevCWnd->IsIconic())
PrevCWnd->ShowWindow(SW_RESTORE);
// If there are pop-ups, bring them along too!
if (PrevCWnd != ChildCWnd)
ChildCWnd->BringWindowToTop();
// Return FALSE. This isn't the first instance
// and we are done activating the previous one.
return FALSE;
}
else
// First instance. Proceed as normal.
return TRUE;
}
CMyApp::InitInstance()
{
if (!FirstInstance())
return FALSE;
// ...
}

null@diku.dk, programmer.tools, 6/19/95
See also Win32 SDK Knowledge Base article Q124134 ("Allowing Only One Application Instance on Win32s") and Jeffrey Richter's Advanced Windows NT, chapter 7, "Prohibiting Multiple Instances of an Application from Running: The MultInst Sample Application" (available on the MSDN Library CD).

null@diku.dk, email, 8/8/95
Update—these were posted to mfc-l:

I have each InitApplication() create a semaphore. If GetLastError() returns ERROR_ALREADY_EXISTS then I know that some other application is already running and has gotten that far, so I bail.

Yourapp::InitInstance()
{
hMutexOneInstance =
CreateMutex(NULL,TRUE,_T("PreventSecondInstance"));
if(GetLastError() == ERROR_ALREADY_EXISTS)
bFound = TRUE;
if(hMutexOneInstance)
ReleaseMutex(hMutexOneInstance);
return (bFound == TRUE) ? FALSE : TRUE;
}

mcontest@universal.com
There is a nice section in Jeffrey Richter's book Advanced Windows NT about this. Essentially, it uses shared data segments between processes.

In your main file, add:
#pragma data_seg(".SharedData")
LONG nUsageCount = -1;
#pragma data_seg()

In your Application's InitInstance(), call:
InterlockedIncrement ( &nUsageCount );

This function returns the incremented value of the variable. If it is nonzero, you know that you are not the first App.

In your Application's ExitInstance(), call:
InterlockedDecrement( &nUsageCount );

In your .DEF file, have the following lines (note that the segment name you give here should match the one in the application's main file):
SEGMENTS
.SharedData shared

abalakri@us.oracle.com
You'd better use one of the built-in synchronization methods. See the Win32 Knowledge Base article Q124134, "Allowing Only One Application Instance on Win32s," for a sample of using a memory-mapped file for synchronization. It doesn't include starting the previous instance, but if you detect that you're not the only one running, it should be pretty simple: If CreateFileMapping fails, try to find the previous instance from the window class name. If it's not found, sleep for a while and start over (with CreateFileMapping). In this way, either you will find the other instance when it gets around to creating its window or CreateFileMapping will eventually succeed. The advantage of using CreateFileMapping instead of CreateObject is that it also works on Win32s.

nuj@kruger.dk
Note There's a sample of this that was contributed by john@jing.com (John Xu) called onetime4.zip that is in the MFC FAQ archive.

-----
We are what we repeatedly do. Excellence, then, is not an act, but a habit.

how to implement or integrate them into an existing VC++6 Diaglogue based program so that I can use some of the ActiveX Controls?

THanks

JW

DJ

Really appreciate your help!

JW

DJ

I want to determine if the mouse is in a certain region, I am using GetCursorPos to get the mouse cursor position and GetWindowRgn to get window region.

I know that the former is in screen coordinates and the latter in window coordinates. Which function should I use to convert one of the two values to make them common so PtInRegion will work?

ScreenToClient & ClientToScreen

I don't quite understand your answer but I have tried those methods and they don't work, I think it's because the region is in WINDOW coordinates and not client as stated in MSDN.

I would use DPtoLP() for which you can find more
information here:
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/gdi/cordspac_6rqo.asp[^]

Thanks, can you post some example code, I still don't understand which coordinates need to be converted and the relationship between logical coordinates and screen/client coordinates.

The a preview of my functions:

<br />
GetCursorPos(&pt);<br />
<br />
GetWindowRgn(hwnd, hRgn);<br />
<br />
if ( PtInRegion(hRgn, pt.x, pt.y) != 0 )<br />
   ...<br />
else<br />
   ...<br />

When you do any type of drawing, for example LineTo(), MoveTo(), etc... that involves the device context (DC) - these functions assume you are passing logical coordinates. These functions are members of the device context (DC) which has its own set of coordinates (the logical coordinates). Your mapping mode (a property of the DC) determines the units of measurement when you draw something.
The coordinate data received from the mouse messages is, however, not in logical coordinate form. Thus, points passed to OnLButtonDown(), OnMouseMove(), etc... are in DEVICE UNITS, that is pixels. These are measured relative to the top left corner of the client area. These are called client coordinates. When you call InvalidateRect() the rectangle is assumed to be defined in terms of client coordinates.
If your mapping mode is MM_TEXT, then client coordinates and logical coordinates in the device context are both in units of pixels and so they are the SAME, as long as you don't scroll the window.
Thus to properly convert between coordinates there are two things you need to do:

1. Convert client coordinates that you get from the mouse to logical coordinates

2. Convert any bounding rectangles (regions, etc...) back to client coordinates (if, for example you want to call InvalidateRect())

Thus you could do something like this in OnLButtonDown() or OnMouseMove() handler:

<br />
void CMyView::OnLButtonDown()<br />
{<br />
     CClientDC aDC(this);    // Create a DC<br />
     OnPrepareDC(&aDC);      // Get origin adjusted<br />
     aDC.DPtoLP(&point);     // convert point to logical<br />
<br />
     // your code here<br />
}<br />

In the above, you obtain a DC for the current view by creating a CClientDC object and passing 'this' to the constructor. The advantage of CClientDC is that windows automatically releases the DC when the object goes out of scope. This is necessary because there are a liminted number of DC's in windows.

For a geometric shape (Rgn, Rect) I suppose you could try the following:

<br />
CClientDC aDC(this);<br />
OnPrepareDC(&aDC);<br />
<br />
CRect aRect = /* Code to construct rect goes here */;<br />
aDC.LPtoDP(aRect);<br />
InvalidateRect(aRect);<br />

OR

<br />
CRgn aRgn = /* Code to create region */;<br />
aDC.LPtoDP(aRgn);<br />
<br />
// Your code here....<br />
if ( PtInRegion(hRgn, pt.x, pt.y) != 0 )<br />
   ...<br />
else<br />
   ...<br />

Thus, the coordinates should be properly converted...

Hope this helps.

Ok, explanation understood...I'm coding in Win32 so the code I didn't quite get fully. (Sorry for not stating this fact)

Just wanted to let you know I solved my problem and all it took was a ScreenToClient and one more little thing that MSDN forgot to mention thus leading to wasting of 15 hours of Yahooing and countless cigarettes:

GetWindowRgn(hRgn, pt.x, pt.y) is supposed to obtain a copy of a region but nowhere does it state that you should have created some kind of region with this handle already!!!!!!!!!!!!

Luckily I came across this on some website with some code example in a programming language I have never seen before that said to create a region first (doesn't matter what kind) and then call the function...I am really curious as to why MSDN thought this to be unimportant enough to mention???

Now does this create a resource leak? Who knows with this region stuff?

Thanks again for the help.

Glad you figured it out.

Good luck.

Hi, I'm trying to develop a program which saves/loads MS Outlook Express setting information, including user preference configuration data, POP3/SMTP account(s) information etc. By knowing that OE stores all those info in Windows registry, it's quite easy to save/load the particular REG keys, and I was actually doing well on saving/loading user preference configuration data.

The problem is lying on POP3/SMTP account(s) information, at first I found out the keys that stores account(s) info was under "HKEY_CURRENT_USER\Software\Microsoft\Internet Account Manager\Accounts" and wrote my program by that way, and it worked, but while I was doing more tests by creating some dummy users on my Win2000 and logging on as them and running the application, I saw the oddness, it sometimes stores account(s) info under "HKEY_CURRENT_USER\Identities\{########-####-####-####-############}\Software\Microsoft\Internet Account Manager\Accounts" (substituting the "##..." with user main identity), so I'm confused.

I did some more tests and found out that the key location is actually random between "HKEY_CURRENT_USER\Software\Microsoft\Internet Account Manager\Accounts" and "HKEY_CURRENT_USER\Identities\{########-####-####-####-############}\Software\Microsoft\Internet Account Manager\Accounts", say, in one user's registry it's here, and in the other user's registry it's there, and both users have exactly identical privilege and other settings, why is that?

My question is, is there any way I can programatically determine which one of the above two keys stores current user's POP3/SMPT account(s) information? Thanks a lot.

Hi, everyone!

Are there some free tools that can calculate the running time
of each function of a project? My project is a C/C++ project
that runs on WIN32 platform.

Thank in advance,
George

I have added a new entry to my menu and MFC has generated the ON_COMMAND entry for the item in its message map. Thats fine..
Now, how do I post a message in my application so that I can cause the same function to be called at my will ?.
I dont want to make I direct call as I dont want any return.

Thanks in advance.

Either SendMessage/PostMessage
YourWnd->SendMessage(WM_COMMAND, ID_MY_COMMAND, NULL);

Thanks very much..it was the WM_COMMAND bit I was missing..
Cheers

I have been having some trouble changing the parent of a CView object. I have a program which utilizes the CSplitterWnd to have two separate views. My intention is for users to undock a splitter view and make it a free floating window. The problem comes when I try and assign ownership of a CView class to another window.

I am using both a CView(when in a splitter) and a CWnd(when a toolbox) to wrap the main CView class containg application information. When i try and assign the CView class a new parent SetParent(wndToolBox) it works fine. However when it comes to sending click messages and such it crashes :/

in CView::OnMouseActivate

fails evaluating
ASSERT(pParentFrame == pDesktopWnd || pDesktopWnd->IsChild(pParentFrame));

It seems that the the CView class which has had a new parent assigned has not been removed from the notification queue of its previous parent.

Is it even possible to dynamically reasign a CView class to another parent window without such errors occuring ??

Any help would be GREATLY appreciated Smile | :)

Thanks