C / C++ / MFC

Re: Error - windows.h already included

10-Sep-04 7:59

skg747 wrote:
I checked in the Tools - Options - Directories - Include files and it looks fine.

What about the INCLUDE environment variable?

It's a long shot, but have you looked through the code for any #error directives?

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

skg74710-Sep-04 10:12

skg747

10-Sep-04 10:12

This is the only #error directive I see (in VC98\MFC\Include\afxv_w32.h):

#ifdef _WINDOWS_
#error WINDOWS.H already included. MFC apps must not #include <windows.h>
#endif

Exception "The parameter is incorrect." on CArchive

Ed K9-Sep-04 11:24

Ed K

9-Sep-04 11:24

I'm trying to save a file to a server and getting a "The parameter is incorrect." exception.

<code>
CFile file;
CFileException ex;
if( file.Open(path, CFile::modeWrite | CFile::shareExclusive | CFile::modeCreate, &ex) )
{
	try
	{
		CArchive ar(&file, CArchive::store);
		ploan->Serialize(ar);

		ar.Close();
		file.Close();
	
	}
	catch(CFileException* pEx)
	{
		// not caught here
	}
	catch(CArchiveException* pEx)
	{
		// not caught here
	}
	catch(CException* pEx)
	{
		// "The parameter is incorrect." caught here
	}
}
</code>

It creates a zero length file and fails. Any ideas??

Thanks!

ed

Re: Exception "The parameter is incorrect." on CArchive

David Crow10-Sep-04 3:27

Re: Exception "The parameter is incorrect." on CArchive

10-Sep-04 3:27

Does it happen if path points to some local (as opposed to a server) folder? Have you stepped into the CArchive constructor or the Serialize() method to see where the exception gets thrown? While the other two Close() methods can also throw an exception, I would rule them out for the time being.

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Ed K10-Sep-04 6:04

Ed K

10-Sep-04 6:04

Thanks!!!

This doesn't happen on the local drive. Only on network drives. And it doesn't happen to everyone either. I have 3 clients out of 1000+ that have this issue.

ed

Re: Exception "The parameter is incorrect." on CArchive

David Crow10-Sep-04 7:52

Re: Exception "The parameter is incorrect." on CArchive

10-Sep-04 7:52

I'm leaning towards a permission-related issue. To verify, dummy up a little program that does nothing but create and write to a file on the network. Does that uncover anything?

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Ed K4-Oct-04 15:49

Ed K

4-Oct-04 15:49

Took a while but wrote a little app for them. There are 3 locations this happens at so far. The test app had the same result. It wasn't an CFile or CArchive exception rather it was caught in the CException catch block.

They all have the problem when they write to the server. When they change the path so that it writes to their C:\ drive there isn't a problem.

Thanks for the bandwidth!

ed

The absence of evidence is not evidence of absence.

Owner Draw List Box Scrolling Problem - ?XP bug

Michael Gunlock9-Sep-04 11:05

Michael Gunlock

9-Sep-04 11:05

I have noticed a "visual scrolling" problem with owner-draw list boxes (both mine and others) in XP. This occurs with both stand-alone list boxes and ones associated with a combo box. The problem occurs if the My Computer | Properties | Performance Options | Visual Effects | Smooth-scroll List Box option is checked. The animated scrolling generated by windows is okay unless you click on the scroll bar to generate page-up/page-down messages. With this, the scrolling animation visually appears to be going in the opposite direction as expected...but when the animation terminates, the appropriate update has occurred (e.g. the expected items are visual). This is quite obvious when the list box size is large enough and the items are different in appearance. This does not appear to occur with non owner-draw list boxes.

I cannot find any reference to this bug but it is clearly reproducible. Is anyone aware of a fix?

Sorting Recordset

Ouenstreet9-Sep-04 10:49

9-Sep-04 10:49

Can someone help me. I have two recordsets that are opened (ADO)
AdRecordSet and AdTimesRecordSet.

The data is then combined by the field AdID into a class / recordset SiteAdInfoSet (the instance is called CAddInfoSet)

I would like to sort CSiteAdInfoSet by m_NrPlays (int) in descending order (field came from AdTimesRecordSet.. I want to re-store the data back into CsiteAdInfoSet as to keep changes to a minimum.

I hope this is clear
Thanks.

TomH

David Crow9-Sep-04 10:57

9-Sep-04 10:57

Does the recordset have an SQL query (e.g., SELECT <field> FROM WHERE ...) associated with it? If so, just add ORDER BY to it.

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Ouenstreet10-Sep-04 4:05

10-Sep-04 4:05

No is does not have an SQL query.

TomH

Ouenstreet10-Sep-04 10:55

10-Sep-04 10:55

Any help out there???
Thanks
TomH

TomH

Ouenstreet13-Sep-04 9:56

13-Sep-04 9:56

Any help out there?

Thanks.
TomH

TomH

Help with Printing

Eversman9-Sep-04 10:01

Eversman

9-Sep-04 10:01

Hey all,

I need a little help with printing. I have it all ready to print and it print fines, but I coded it stupidly. I hard coded the coordinates of where the text is printed, so it only prints correctly on certain printers, on others it prints all messed up. So now I am trying to change it so that it works right with all printers. This is what I have done so far.

<br />
<br />
CPrintDialog dlg(FALSE);     <br />
	CDC dc; <br />
	#if 0     <br />
	if (dlg.DoModal() == IDCANCEL)       <br />
		return;   <br />
	dc.Attach(dlg.GetPrinterDC());<br />
	#else     <br />
	PRINTDLG prtDlg;    <br />
	AfxGetApp()->GetPrinterDeviceDefaults(&prtDlg);     <br />
	dlg.m_pd.hDevMode = prtDlg.hDevMode;    <br />
	dlg.m_pd.hDevNames = prtDlg.hDevNames;    <br />
	dc.Attach(dlg.CreatePrinterDC());<br />
	#endif   <br />
	dc.m_bPrinting = TRUE;   <br />
<br />
<br />
	CRect printArea;<br />
printArea.SetRect(printArea.left/2, printArea.top/2, dc.GetDeviceCaps(HORZRES), dc.GetDeviceCaps(VERTRES));<br />
<br />
CString str("Delayed Entry");<br />
dc.DrawText(str, printArea, DT_NOPREFIX | DT_WORDBREAK);<br />
<br />

When I try that out it compiles fine, but when I print it it doesn't work, can anyone help me out here?

Thanks

David Crow9-Sep-04 10:38

9-Sep-04 10:38

Eversman wrote:
...it doesn't work, can anyone help me out here?

What doesn't work? What is the code (not) doing? What are you wanting it to do? Be a little more specific instead of just "it doesn't work" and you'll get much more help.

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Eversman9-Sep-04 10:43

Eversman

9-Sep-04 10:43

Well it doesn't print, it prints the old stuff that I had hard coded in, but when I change it to the new way I just showed, it doesn't print, well maybe it does but it would be off the page when it printed.

I am not sure if I am using the GetDeviceCaps and that such correctly.

David Crow9-Sep-04 11:00

9-Sep-04 11:00

Eversman wrote:
CRect printArea;
printArea.SetRect(printArea.left/2, printArea.top/2, dc.GetDeviceCaps(HORZRES), dc.GetDeviceCaps(VERTRES));

This should raise a red flag. You have a variable printArea that has not been initialized/assigned a value, yet you are trying to use its left and top members. Is that intentional?

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Eversman9-Sep-04 11:26

Eversman

9-Sep-04 11:26

Well, I am using this example that I found in the printing section. This is what the tutorial has it as.

<br />
<br />
   CRect rect;<br />
   rect.left = (JRECT.Width() - size.cx)/2;<br />
   rect.right = rect.left + size.cx;<br />
   rect.top = (JRECT.Height() - size.cy)/2;<br />
   rect.bottom = rect.top + size.cy;<br />
    <br />
   JDC.DrawText(str, &rect, DT_SINGLELINE);<br />
<br />

so I kinda modified it to fit what I already had. No where in the example that I see is that stuff initialized. But I need to initialize it to the printer page resolution right? So would I have to initialize it to HORZRES and VERTRES?

David Crow10-Sep-04 2:11

Win32 control compositing

10-Sep-04 2:11

Eversman wrote:
No where in the example that I see is that stuff initialized.

You're kidding, right? Two-thirds of the lines you've shown are initializing variables. When DrawText() is called, rect has all the right numbers.

"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen

Jim Crafton9-Sep-04 9:47

Jim Crafton

9-Sep-04 9:47

During porting the VCF to osx I came across a distinction that Apples uses - the notion of a composited window, where a window is the popup window where all the child controls live. While I was debugging something on OSX I found that if a window was marked as being composited ALL the child controls used the same graphics device to paint on during their paint event.
So, while I don't know how Apple does this, I started to wonder about this on Win32. What would happen if you maintained a map of popup windows and bitmap hdc's such that for each popup window you had one in-memory bitmap DC that was the height/width of the popup window. When the window was resized so would the bitmap. Then, for all the child control drawing, they would draw onto a section of this main DC, and then just bitblt the relevant part back to the HDC passed in to the WM_PAINT.
This would give you:
- automatic double buffering without each control having to maintain a separate bitmap/memory dc
- the ability to draw "outside" the edges of your control, for example if you wanted a "glow" effect on the edges of your control this would make this possible.
- possibly more efficient transparency effects that because you are reusing the same DC to paint on.

I would be curious to hear what others think. Has this been done before? If so did it work very well? Am I totally out of my mind?

Cheers

Jim

¡El diablo está en mis pantalones! ¡Mire, mire!

Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)!

SELECT * FROM User WHERE Clue > 0
0 rows returned

Ryan Binns9-Sep-04 18:46

Ryan Binns

9-Sep-04 18:46

Sounds interesting. The child controls would need to be either rewritten or subclasses to use the new DC as the WM_PAINT message doesn't include a DC. Other than that, I think it has potential. I guess you'd BitBlt() the memory DC to the screen when the parent processes the WM_PAINT message as the children are always drawn first.

Ryan

"Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact"

Jim Crafton10-Sep-04 5:25

Jim Crafton

10-Sep-04 5:25

WM_PAINT message doesn't include a DC

Actually it sort of does. According to MSDN "For some common controls, the default WM_PAINT message processing checks the wParam parameter. If wParam is non-NULL, the control assumes that the value is an HDC and paints using that device context."[^]

¡El diablo está en mis pantalones! ¡Mire, mire!

Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)!

SELECT * FROM User WHERE Clue > 0
0 rows returned

Ryan Binns10-Sep-04 22:29

Ryan Binns

10-Sep-04 22:29

Yeah, I've seen that before too Smile | :)

. And it would be good if it worked, but as it says, some common controls check this parameter, not all. And Windows itself never sets the wParam parameter; the WM_PAINT documentation states that wParam is unused. I've always thought this lack of DC was a very bad oversight.

But still, it could be useful Smile | :)

Ryan

"Punctuality is only a virtue for those who aren't smart enough to think of good excuses for being late" John Nichol "Point Of Impact"