Using QT SDK from trolltech site ./demoxbp worked as expected.
Then I've installed distro's KDE and qt-dev and I always get at start:
Program received signal SIGSEGV, Segmentation fault.
0x006a9163 in strlen () from /lib/libc.so.6
...
(gdb) bt
#0 0x006a9163 in strlen () from /lib/libc.so.6
#1 0x
Revision: 12400
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=12400&view=rev
Author: druzus
Date: 2009-09-03 23:44:49 + (Thu, 03 Sep 2009)
Log Message:
---
2009-09-04 01:44 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/ChangeLog
Hi
Here is the DllCall() usage in Xbase++:
#define SW_HIDE 0
#define SW_NORMAL 1
#define SW_SHOWMINIMIZED2
#define SW_SHOWMAXIMIZED3
#define SW_MAXIMIZE 3
#define SW_SHOWNOACTIVATE 4
#define SW_SHOW 5
#define SW_MINIMIZ
Revision: 12399
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=12399&view=rev
Author: druzus
Date: 2009-09-03 21:55:01 + (Thu, 03 Sep 2009)
Log Message:
---
2009-09-03 23:54 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/contrib/hbwin
On Thu, 03 Sep 2009, Mindaugas Kavaliauskas wrote:
Hi,
> In such a case we can add additional parameter:
> __AXREGISTERHANDLER( pDisp, bHandler [, cIID ] ) --> pDisp
> cIID is default to IID_IDispatch.
>
> This extension will allow to use any interface, f.e., you can use:
> __AXREGISTERHANDLE
Mindaugas Kavaliauskas wrote:
This makes a little more light (to me) on the subject. Is this
"dispinterface ICoBpTiEvents { ... }" a C++ syntax?
"OLE View" is an application supplied with Visual Studio 6 which displays type
library information, I guess it might correspond to how the interface
This is all well and good again, but begs another question: "How does a
client know the IID it wants to connect to?" Excellent question! There
are two possible answers:
Actually, the first method is known and implemented, but the second is
not clear for me.
From OLE View bundled with VB6:
Mindaugas Kavaliauskas wrote:
Can you enumerate and print all available BPTI connection interfaces and
try if these interfaces are available via FindConnectionPoint()?
I'm a bit out of my depth here, for me it's kind of "monkey see, monkey do".
Do you have any simple MSCAL test code, I can co
Przemyslaw Czerpak wrote:
So there is sth wrong with FindConnectionPoint() and probably due to
wrong interface ID but I think we should verify it.
And of course we should check what interface ID we are accepting now.
Alex?
I am not sure what to do? Check it for equality with IID_IDispatch?
Reg
My BPTI ActiveX uses EnumConnectionPoints and GetConnectionInterface as
FindConnectionPoint fails. It seems to work fine, I get events that I
expect. But, like MSCAL it GPF's on exit.
Can you enumerate and print all available BPTI connection interfaces and
try if these interfaces are available
Mindaugas Kavaliauskas wrote:
That's why I simplified code and used FindConnectionPoint().
Does this from MSDN help explain:
===
This is all well and good again, but begs another question: "How does a client
know the IID it wants to connect to?" Excellent question! There are two possible
ans
Mindaugas Kavaliauskas wrote:
... and yet somehow, it works ... I just tested with this rather weird
EFT ActiveX I have been given to talk to, and Przemyslaws new code.
Unless the GPF you warn of is on the last event :)
Can you debug: what interface you connect to (value of rriid after
connec
On Thu, 03 Sep 2009, Mindaugas Kavaliauskas wrote:
> But that is done by :FindConnectionPoint() method. It accepts interface
> parameter and returns a required connect point if it exists, instead of
> evaluation over all available interfaces.
>
> It's a kind of:
>pConnectionPoint := hConnPoin
If I understand you well then we should remove the rriid ISink member
and modifiy this code:
for( ;; )
{
lOleError = HB_VTBL( pEnumCPs )->Next( HB_THIS_( pEnumCPs ) 1,
&pCP, NULL );
if( lOleError != S_OK )
break;
if( HB
On Thu, 03 Sep 2009, Mindaugas Kavaliauskas wrote:
Hi,
> Connection point can be implemented using different interfaces. Let us
> assume, Internet Explorer notifies a new URL is loaded using UrlLoaded
> method of IExplorerNotify interface:
> void IExplorerNotify:UrlLoaded( char* pszURL )
> Su
... and yet somehow, it works ... I just tested with this rather weird
EFT ActiveX I have been given to talk to, and Przemyslaws new code.
Unless the GPF you warn of is on the last event :)
Can you debug: what interface you connect to (value of rriid after
connection point enumeration loop)? W
On Thu, 03 Sep 2009, Alex Strickland wrote:
> ActiveX code, so perhaps there is some change in the general ole code that
> is exploited by the ActiveX. The last "good" version of Harbour which I
> have is "10786 2009-04-06 13:35:09Z druzus", which was roughly 1.1.0.
> Perhaps I will try some cut
Mindaugas Kavaliauskas wrote:
Connection point can be implemented using different interfaces. Let us
assume, Internet Explorer notifies a new URL is loaded using UrlLoaded
method of IExplorerNotify interface:
void IExplorerNotify:UrlLoaded( char* pszURL )
Such interface is good for C code, b
Hi,
Looking at the code I see one important difference between our new AX
code and GTWVG/HWGUI.
New AX code calls
lOleError = HB_VTBL( pCPC )->FindConnectionPoint( HB_THIS_( pCPC )
HB_ID_REF( IID_IDispatch ), &pCP );
to take connection point but GTWVG
Przemyslaw Czerpak wrote:
so it's not the exact reason of the problem.
No. Note that this same behavior (GPF) occurs with new Harbour and HWGUI ActiveX
code, so perhaps there is some change in the general ole code that is exploited
by the ActiveX. The last "good" version of Harbour which I h
On Thu, 03 Sep 2009, Alex Strickland wrote:
> static void hb_oleAxExit( void* cargo )
> {
>HB_SYMBOL_UNUSED( cargo );
>if( s_hLib )
>{
> s_pAtlAxGetControl = NULL;
> MessageBox(NULL, "FreeLibrary", "", MB_OK);
> //FreeLibrary( s_hLib );
> s_hLib = NULL;
>}
> }
> Th
Hi
If I write:
MsgInfo( "Activate" )
ACTIVATE WINDOW oMainWindow
MsgInfo( "Setting oMSCAL to nil" )
oMSCAL := nil
MsgInfo( "Setting oBPTI to nil" )
oBPTI := nil
Then I see "Activate" but not "Setting oMSCAL to nil". It is when the container
window is destroyed that the
Przemyslaw Czerpak wrote:
Please comment in contrib/hbwin/axcore.c[78] this line:
FreeLibrary( s_hLib );
and check if it resolves the problem.
static void hb_oleAxExit( void* cargo )
{
HB_SYMBOL_UNUSED( cargo );
if( s_hLib )
{
s_pAtlAxGetControl = NULL;
MessageBox(NULL,
On Thu, 03 Sep 2009, Alex Strickland wrote:
>> I can only guess that you sync SVN after few days and you haven't made
>> clean build. Please recompile whole Harbour code, i.e.
> You were correct.
Fine,
> The bad news is that on exit, the application GPFs. I have done a build
> with debugging ena
Too quick, the full log:
'hellox.exe': Loaded 'C:\Users\Alex\ECR6.4\tests\hellox.exe', Symbols loaded.
'hellox.exe': Loaded 'C:\Windows\System32\ntdll.dll'
'hellox.exe': Loaded 'C:\Windows\System32\kernel32.dll'
'hellox.exe': Loaded 'C:\Windows\System32\user32.dll'
'hellox.exe': Loaded 'C:\Window
Revision: 12398
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=12398&view=rev
Author: druzus
Date: 2009-09-03 12:02:22 + (Thu, 03 Sep 2009)
Log Message:
---
2009-09-03 14:02 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/compil
Przemyslaw Czerpak wrote:
I can only guess that you sync SVN after few days and you haven't made
clean build. Please recompile whole Harbour code, i.e.
You were correct.
The good news is that both activex controls load without error, I have done no
very deep tests but things seem fine.
The
Hello Fernando,
When can we start testing ViaNemo ?
Regards,
Rossine.
--
View this message in context:
http://www.nabble.com/SF.net-SVN%3A-harbour-project%3A-12382--trunk-harbour-tp25231702p25274557.html
Sent from the Harbour - Dev mailing list archive at Nabble.com.
__
HOW TO BUILD AND INSTALL HARBOUR FROM SOURCE AND CROSS COMPILE HARBOUR
FOR WIN ON LINUX
===
*
( Tested on Ubuntu 9.04 and Ubuntu 9.10 )
*
Fernando Athayde wrote:
why memvar and public in same time?
you should to put only public
I like to compile with full warnings so I always put
memvar x
public x
If you mean why am I mixing memvars and statics, only because I was trying a few
things to make sense of the crazy compiler messag
why memvar and public in same time?
you should to put only public
regards,
Fernando
De: Alex Strickland
Para: Harbour Project Main Developer List.
Enviadas: Quinta-feira, 3 de Setembro de 2009 8:13:11
Assunto: Re: [Harbour] ActiveX
Przemyslaw Czerpak wrote:
On Thu, 03 Sep 2009, Alex Strickland wrote:
> I Didn't get near ActiveX! When I compile I get :
>
> Compiling 'hellox.prg'...
> hellox.prg(0) Warning W0001 Ambiguous reference 'OMAINWINDOW'
> hellox.prg(0) Warning W0002 Ambiguous reference, assuming memvar
> 'OMAINWINDOW'
> hellox.prg(0) Wa
Przemyslaw Czerpak wrote:
I can add such method to our new AX code if FindConnectionPoint() fails.
Anyhow I would like to ask Mindaugas to accept it and you and Alex to make
tests.
I Didn't get near ActiveX! When I compile I get :
Compiling 'hellox.prg'...
hellox.prg(0) Warning W0001 Am
Hi Fernando
Did you consider applying your efforts to HWGUI or HMG? I know nothing of HMG,
but I think HWGUI would benefit from some internal changes, and yet you would
still have something of a community which already exists. The command syntax is
quite similar to FiveWin.
Why did you decid
Przemyslaw Czerpak wrote:
I can add such method to our new AX code if FindConnectionPoint() fails.
Anyhow I would like to ask Mindaugas to accept it and you and Alex to make
tests.
I will test asap, thank you for your efforts. Really, sometimes reading mails on
this group is like getting Chri
Hello Przemek,
>>Thank you very much for great contribution.
>>I think that it's the answer for many user requests:
>>GUI library which is compatible with FWH, works on desktop windows
>>and with WinCE and can be ported to other OS-es, maybe even integrated
>>with our HBQT keeping the compatibili
Hello Massimo,
>>Is possible to have compatibility with MiniGUI extended also ?
I don't know the MINIGUI sintax, but ViaNemo is opensource and full source
code,
Any user can expand for user other GUI's sintax.
Best Regards,
Fernando Mancera
http://blog.viaopen.com
37 matches
Mail list logo
