ATL / WTL / STL

Re: Failure to Recognise Base Class

20-Aug-11 23:32

I have always found these MFC macros more than somewhat confusing - great if they work, but almost impossible to debug.

Unrequited desire is character building. OriginalGriff

«_Superman_»22-Aug-11 13:14

«_Superman_»

22-Aug-11 13:14

Is CProgressDlg inside any namespace that you need to use?

«_Superman_»
I love work. It gives me something to do between weekends.

Microsoft MVP (Visual C++)

Polymorphism in C

Common Controls Issue

Tracy Software16-Aug-11 2:55

Tracy Software

16-Aug-11 2:55

How to add Common Controls 6.0 to DLL / ATL ?
I don't have the source code of the EXE, I'm writing only the DLL linked to it.

If I put this line in the stdafx.h of the DLL it doesn't work:

#pragma comment(linker,"/manifestdependency:\"type='win32' name='Microsoft.Windows.Common-Controls' version='6.0.0.0' processorArchitecture='x86' publicKeyToken='6595b64144ccf1df' language='*'\"")

InitCommonControls(); / InitCommonControlsEx();
is not working neither,
I need Common Controls 6.0 to show my tooltips controls and they work only with the manifestdependency line at the exe file / or manifest file to exe

Thanks for anyone who can help,
Mithrill

iterator default value [modified]

liquid_5-Aug-11 1:34

liquid_

5-Aug-11 1:34

How can I check whether an iterator was assigned a value or not? In other words, if I create an iterator with default constructor what value will it present which I could check against?
It is not the trivial case such this:

C++

list<int>::iterator it; //what does it contains?

but this:

C++

map<AClass *,list<int>::iterator> aci;
AClass *ptr=someinitializedaclassptr;
if(aci[ptr]==UNINITIALIZED_ITERATOR)
{
   dosomething();
}

In above example issuing aci[ptr] creates already an entry in the map with default constructed value list<int>::iterator if there isn't a key of value ptr.

modified on Friday, August 5, 2011 7:48 AM

Re: iterator default value

Stephen Hewitt6-Aug-11 20:40

Stephen Hewitt

6-Aug-11 20:40

Read this thread on (ahem) stackoverflow: http://stackoverflow.com/questions/3395180/what-is-an-iterators-default-value[^]

Steve

Re: iterator default value

liquid_7-Aug-11 9:18

liquid_

7-Aug-11 9:18

Well, it is obvious it has no special value as int val;. The problem is how to check this. I'll try the below code:

C++

#include "list"
#include "map"
#include "iostream"

using namespace std;

int main()
{
    list<int>::iterator it,itnull;
    map<int,list<int>::iterator> imap;

    it = imap[1]; //map creates entry with key '1' and default constructed iterator
    if(it==itnull) //checking with other default constructed iterator
        cout<<"it unitialized"<<endl; //the right answer
    else
        cout<<"it has other value"<<endl;
    return 0;
}

SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\

VeganFanatic24-Jul-11 8:44

VeganFanatic

24-Jul-11 8:44

So I am digging around deep in the registry

So far so good, but I need to access specifics for each subkey

for (i=0; i<cSubKeys; i++)  {
    cbName = MAX_KEY_LENGTH;
    retCode = RegEnumKeyEx(hKey, i, achKey, &cbName, 0, 0, 0, &ftLastWriteTime);
    if (retCode == ERROR_SUCCESS) {
        RegQueryValueEx(hKey,L"DisplayName",NULL,NULL,(LPBYTE) lpData, &buffersize);
        applist.push_back(lpData);
    }
}

Anyway all I get are a bunch of "-" so am I getting the right type etc?

http://www.contract-developer.tk

Re: SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\

Richard MacCutchan24-Jul-11 21:09

Re: SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\

24-Jul-11 21:09

What are the types of lpData and applist?

The best things in life are not things.

VeganFanatic25-Jul-11 2:25

VeganFanatic

25-Jul-11 2:25

I added another layer of hKey

lpData is char*
applist is CString

I got it figured out with a lot more code needed

http://www.contract-developer.tk

Re: SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\

Richard MacCutchan25-Jul-11 3:33

Re: SOFTWARE\\Microsoft\\Windows\\CurrentVersion\\Uninstall\\

25-Jul-11 3:33

Are you sure that lpData is big enough to hold the returned value? What is the value of buffersize when you call the query function? Also, I do not see a push_back() method for either CString or CStringT classes; are you sure this is what you mean?

The best things in life are not things.

MicroVirus6-Sep-11 2:18

MicroVirus

6-Sep-11 2:18

I'm guessing both lpData as applist need to be wide (unicode). You are using the wide version of RegQueryValueEx so for string values you are (I think) returned the unicode/wide version.

Help regarding method WNetCancelConnection2

tasumisra18-Jul-11 23:42

tasumisra

18-Jul-11 23:42

Hi Guys,

I am having one application(Windows service) that was working fine with VS 2003,i am compiling this application with Visual Studio pro 2008 and getting linker error for WNetCancelConnection2.

To supress linker error i have added MPR.LIB,it all work fine but whe its been deployed to server it was giving error
"Dependent Assembly Microsoft.VC80.MFC could not be found and Last Error was
The referenced assembly is not installed on your system ".

can somebody help me here.

thanks

vikas da

Re: Help regarding method WNetCancelConnection2

Richard MacCutchan19-Jul-11 0:35

Problems with Docking Toolbars

19-Jul-11 0:35

You need to get the runtime libraries installed on your server. See the link in this message[^].

The best things in life are not things.

Syamlal S Nair18-Jul-11 17:59

Syamlal S Nair

18-Jul-11 17:59

Hi,

I have been trying to adopt certain new MFC classes(version-10.0) into my application (written based on MFC 7.0 originally).
A brief of the context is as below.

I have an application with the main window class, CMainFrame derived from CMDIFrameWnd, and the toolbars created from CToolBar.

These ToolBars are fitted into objects of a class called CMyDialogBar. CMyDialogBar is derived from CDialogBar (which is a child of CControlBar, as you know).

In order to serialize the dock information, in one of the Dlls of my application, CDockState object is used in the code. A pointer to the application's main window is retrieved using AfxGetMainWnd() and dynamic casted to CFrameWnd, and methods like GetDockState() etc. are called against it. Also, CControlBarInfo pointer is used to get each control bar information from CDockState object, and remove a couple of them from the list based on some specific logic before
saving it for persisting.

Now starts the real problem.

I need to incorporate CMFCToolBar instead of CToolBar, and the application's main window is changed to derive from CMDIFrameWndEx instead of CMDIFrameWnd.CMyDialogBar class is now altered to derive from 'CDockablePane' (instead of CDialogBar) as I found in some forums.

I get debug assertion failures, wherever CControlBarInfo is used, and wherever GetDockState() or SetDockState() is called since those methods refer to CControlbar objects/pointers internally.

Sample code from dll, is as shown here:

void CDialogBarService::RestoreState()
{
	CFrameWnd* pWnd = DYNAMIC_DOWNCAST( CFrameWnd, OldStyleAfxGetMainWnd());
	pWnd->PostMessage( WM_STATUS_UPDATE, 0, 0);
	
	ASSERT( pWnd != NULL);

	// Restore from user object
	//
	CDockState state;

	CMyContext context;
	CMwSession session;
	CProxyUser*	pUser = session.GetUserProxy();
	if ( pUser != NULL && pUser->GetInstrProfileContext( NAME_LAYOUT_DIALOG_BAR_SERVICE, context) )
	{
		if ( context.Exists( NAME_TOOLBAR_STATE) )
		{
			CByteArray data;
			if ( context[NAME_TOOLBAR_STATE].Get( data) )
			{
				// create a RawByteFile with a pointer to the byte array
				// reqd as the archive object requires a file not an array
				//
				CRawByteFile file( data.GetSize(), data.GetData());

				// create an archive object, set to LOAD
				//
				CArchive ar( &file, CArchive::load);

				// extract the toolbar state from the archive
				//
				state.Serialize( ar);
			}
		}

		// Do some pre-processing to ensure that we get no errors
		//
		// If our toolbars are hidden they are not restored and so we can get
		// error conditions in SetDockState.  This code just purges
		// the saved state of any toolbars that don't exist at this time
		//
		for ( int i = state.m_arrBarInfo.GetUpperBound(); i >= 0 ; i-- )
		{
			CControlBarInfo* pInfo = reinterpret_cast< CControlBarInfo*>( state.m_arrBarInfo[i]);
			ASSERT( pInfo != NULL);
			pInfo->m_pBar = pWnd->GetControlBar( pInfo->m_nBarID);
			if ( pInfo->m_pBar == NULL )
			{
				CControlBarInfo* pState = reinterpret_cast< CControlBarInfo*>( state.m_arrBarInfo[i]);
				state.m_arrBarInfo.RemoveAt( i);
				delete pState;
			}
		}

		// Finally set toolbar state		
		pWnd->SetDockState( state);
		
	}
}

In the above sample code, i even tried to replace CFrameWnd pointer to CMDIFrameWndEx pointer, but the method call SetDockState() was giving compilation error, since it is made protected in the new MFC class hierarchy. Since i am standing in a dll, i cannot have my CMainFrame class instantiated here, to call a public method or overridden member function, which in turn can invoke the protected member.

Questions-

1. Which class i can use for replacing CControlBarInfo in the context of DockablePanes?

2. Why those methods like SetDockState(), GetDockingManager() etc. made protected for CMDIFrameWndEx class in new hierarchy?

3. Is there any other mechanism/classes using which i could achieve the above persistence?

Thanks in advance.

Regards
Syamlal

SYAMLAL

Compiling with VS2010

Kyudos14-Jul-11 13:39

14-Jul-11 13:39

I recently switched to VS2010 and its a big pain. I understand it is more standards compliant etc., but it also seems incredibly fragile. The linker crashes often, the thing that compiled fine yesterday won't compile today, despite me not changing anything. If I leave my compilation running while locking the computer VS2010 crashes. All in all the whole product seems flaky.

Anyway to specifics:

How can I get rid of the MSB8012 warning? Why does MS insist on thinking it knows what I want better than I do? Output a DLL to a non-standard location and it starts freaking out. Also doesn't seem to like relative paths much in these settings.

Is there a way to turn off the use of the path macros?

I also keep getting:

error LNK1318: Unexpected PDB error; ACCESS_DENIED (15)

Even though the PDB in question is for a different product and nothing is using it and it isn't read-only.

Any one know anything about that?

Bring back VS6! Roll eyes | :rolleyes:

Albert Holguin15-Jul-11 3:54

Re: Compiling with VS2010 [modified]

15-Jul-11 3:54

Make sure you install the service packs (I do believe SP1 is out for VS2010), I use VS2008 and have not major complaints after the first service pack.

Kyudos wrote:
How can I get rid of the MSB8012 warning?

Use a #pragma warning( disable: WARNING_NUMBER ) statement

Kyudos wrote:
Output a DLL to a non-standard location and it starts freaking out.

I do this all the time in VS2008 and it has no problem with it, wouldn't imagine they changed that very much from one release to another.

Kyudos wrote:
error LNK1318: Unexpected PDB error; ACCESS_DENIED (15)

These are the type of random errors you sometimes experience pre-service packs... Laugh | :laugh:

I hope your experience with VS2010 gets better, I've seen a lot of improvements from the usability stand point and the better standard C++ compliance is certainly a great step forward for MS. Good luck! Smile | :)

Kyudos17-Jul-11 17:50

17-Jul-11 17:50

Thanks Albert - I downloaded SP1 and now have a different set of problems!

At first my application would crash the development environment every time I tried to debug (F5),
so I decided to do a full rebuild.

Now one of my Fortran libraries won't compile, because it appears to be looking for a dependent library in a weird place:

fatal error LNK1181: cannot open input file 'C:\Program Files\Microsoft Visual Studio 10.0\Common7\...'

I can't seem to find anything that would make it look there for lib files.

Any ideas?

EDIT: Ah - Fortran problem!

Solution[^]

modified on Sunday, July 17, 2011 11:56 PM

Albert Holguin18-Jul-11 4:33

18-Jul-11 4:33

So you got your linker problems resolved?

Kyudos18-Jul-11 11:13

18-Jul-11 11:13

Yeah, finally, thanks. Will just have to wait and see if SP1 improves the general stability - I've had the dev-env crash a couple of times already though.

Albert Holguin18-Jul-11 11:19

18-Jul-11 11:19

They usually do, what OS are you running on?

Kyudos19-Jul-11 10:40

19-Jul-11 10:40

XP SP3 at the moment, but we'll be looking to move to Win7 in the near future.

Albert Holguin19-Jul-11 11:32