Hello,

this situation is a little strange ...

The program terminates when oDList:create() is executed. If I use XppDbg to
singlestep through the lines, everything is fine. Running with F5 or
running the program without the debugger crashes when creating the object:

FATAL ERROR LOG 
Stack Overflow
SYS Thread-ID: 632 
Module: EH
Error Codes: EH: 12 Sub: 0(0) OS: 0 XPP: 16
Call Stack of Thread 1 (440):
CONFIGURATIONWIZARD(67)
(B)GenerateMenu(89)
MAIN(63)
Call Stack of Thread 2 (632):
Call Stack of Thread 3 (920):
File: E:\Projects\PC.Calc\gdb.exe
TimeStamp: 20071129 14:53
End of FATAL ERROR LOG.

Here the code are where the problem arises:

   aPosR[1] += aSizeR[1] + (4 * nDistanceLeftBorder) + 10
   aPosR[2] := 10
   aSizeR[2] := (3 * nRowHeight) + 10
   oDList := XbpComboBox():new(oXbp, oXbp, aPosR, aSizeR)
   oDList:tabStop := .T.
   oDList:type := XBPCOMBO_SIMPLE
   oDList:create()

   WHILE !oCursor:eof
      cItem := oCursor:FieldGet(1)
      oDList:addItem(cItem)
      oCursor:skip(1)
   END

   oDList:itemMarked := {|uNil1, uNil2, self| CWHousingData(self, oCursor,
oXbp)}

Any idea why this happens? This is the second case where a program crashes
without debugger ... Maybe it is the way I program ...

Kind regards,

Georg

This is usually caused by a recursive error condition. Are you using a
custom error handler? If so you may have an error in the error handler.

Jim

Georg Lorrig wrote:
> Hello,
> 
> this situation is a little strange ...
> 
> The program terminates when oDList:create() is executed. If I use XppDbg to
> singlestep through the lines, everything is fine. Running with F5 or
> running the program without the debugger crashes when creating the object:
> 
> FATAL ERROR LOG 
> Stack Overflow
> SYS Thread-ID: 632 
> Module: EH
> Error Codes: EH: 12 Sub: 0(0) OS: 0 XPP: 16
> Call Stack of Thread 1 (440):
> CONFIGURATIONWIZARD(67)
> (B)GenerateMenu(89)
> MAIN(63)
> Call Stack of Thread 2 (632):
> Call Stack of Thread 3 (920):
> File: E:\Projects\PC.Calc\gdb.exe
> TimeStamp: 20071129 14:53
> End of FATAL ERROR LOG.
> 
> Here the code are where the problem arises:
> 
>    aPosR[1] += aSizeR[1] + (4 * nDistanceLeftBorder) + 10
>    aPosR[2] := 10
>    aSizeR[2] := (3 * nRowHeight) + 10
>    oDList := XbpComboBox():new(oXbp, oXbp, aPosR, aSizeR)
>    oDList:tabStop := .T.
>    oDList:type := XBPCOMBO_SIMPLE
>    oDList:create()
> 
>    WHILE !oCursor:eof
>       cItem := oCursor:FieldGet(1)
>       oDList:addItem(cItem)
>       oCursor:skip(1)
>    END
> 
>    oDList:itemMarked := {|uNil1, uNil2, self| CWHousingData(self, oCursor,
> oXbp)}
> 
> Any idea why this happens? This is the second case where a program crashes
> without debugger ... Maybe it is the way I program ...
> 
> Kind regards,
> 
> Georg

Hi, Jim -


no, I use the standard errorsys.prg. Just for curiosity I added a
ConfirmBox() at the very beginning of the error routine - it does not pop
up.

After replacing the XbpComboBox with a XbpListBox the error is gone.

Still puzzled and looking for an explanation...

Georg

On Thu, 29 Nov 2007 15:56:55 -0500, James Loughner wrote:

> This is usually caused by a recursive error condition. Are you using a
> custom error handler? If so you may have an error in the error handler.
> 
> Jim
> 
> Georg Lorrig wrote:
>> ...
>> FATAL ERROR LOG 
>> Stack Overflow
>> SYS Thread-ID: 632 
>> Module: EH
>> Error Codes: EH: 12 Sub: 0(0) OS: 0 XPP: 16
>> Call Stack of Thread 1 (440):
>> CONFIGURATIONWIZARD(67)
>> (B)GenerateMenu(89)
>> MAIN(63)
>> Call Stack of Thread 2 (632):
>> Call Stack of Thread 3 (920):
>> File: E:\Projects\PC.Calc\gdb.exe
>> TimeStamp: 20071129 14:53
>> End of FATAL ERROR LOG.

Georg
I had a similar situation when I started using 1.9 I would get a fatal error
caused by a stack overflow. Tech support sent me a new version of XPPUI1.DLL
The information about the DLL is as follows:Hotfix for PDR 5562, ActiveX
Control Refresh, PDR 5532, 5564, _WM_REDRAW mgmt., PDR 5567, PDR 5181, PDR
5569, 5578, PDR 5589, GraSegSave()/GraSegLoad(), PDR 5595, PDR 5598, PDR
5599, PDR 5600, 5622, 5571, 5568, 5614, 5634, HIGH_PRECISION, 5689, 5690,
XbpPS:RefDevice
Perhaps this may help you.
Doug
"Georg Lorrig" <glorrig@fastmail.fm> wrote in message
news:8k0dvplhil1i$.1k9ngqihpwupo$.dlg@40tude.net...
> Hi, Jim -
>
>
> no, I use the standard errorsys.prg. Just for curiosity I added a
> ConfirmBox() at the very beginning of the error routine - it does not pop
> up.
>
> After replacing the XbpComboBox with a XbpListBox the error is gone.
>
> Still puzzled and looking for an explanation...
>
> Georg
>
> On Thu, 29 Nov 2007 15:56:55 -0500, James Loughner wrote:
>
> > This is usually caused by a recursive error condition. Are you using a
> > custom error handler? If so you may have an error in the error handler.
> >
> > Jim
> >
> > Georg Lorrig wrote:
> >> ...
> >> FATAL ERROR LOG
> >> Stack Overflow
> >> SYS Thread-ID: 632
> >> Module: EH
> >> Error Codes: EH: 12 Sub: 0(0) OS: 0 XPP: 16
> >> Call Stack of Thread 1 (440):
> >> CONFIGURATIONWIZARD(67)
> >> (B)GenerateMenu(89)
> >> MAIN(63)
> >> Call Stack of Thread 2 (632):
> >> Call Stack of Thread 3 (920):
> >> File: E:\Projects\PC.Calc\gdb.exe
> >> TimeStamp: 20071129 14:53
> >> End of FATAL ERROR LOG.