gamedevlists-windows — List for discussing game development on the Windows platform

RE: [GD-Windows] Call stack

[Mat N. \(BUNGIE\) <mat...@mi...>]

If you want to view the stack in a debugger later, you can also generate
a minidump via MiniDumpWriteDump.

MSN
-----Original Message-----
From: gam...@li...
[mailto:gam...@li...] On Behalf Of
Jon Watte
Sent: Wednesday, February 09, 2005 9:24 AM
To: gam...@li...
Subject: RE: [GD-Windows] Call stack


In debug mode, you can walk the EBP frames pretty easily. Use=20
some inline assembly or setjmp() to read the value of the register=20
and then treat it like a pointer-to-void*, and you're good to go.

In release mode, it might still be possible to do, assuming you=20
don't turn on the "omit frame pointer" optimization. Sadly, that=20
optimization gives you another whole register to play with, so it's=20
usually something you'll want for your shipping builds. At that=20
point, I've had reasonable success with scanning the stack for=20
addresses that point right after a CALL or JMP REG instruction=20
(JMP REG is used for DLL/vectored calls).

In debug mode, it looks like this:


#include <stdio.h>


__declspec( naked ) char ** get_ebp() {
  __asm {
    mov eax, ebp
    ret
  }
}

// the stack root needs to be thread-local for multi-threaded
// applications
char ** root =3D 0;
void init_callstack( void * ptr ) {
  root =3D (char **)ptr;
}

void callstack( char const * name ) {
  char ** ptr =3D get_ebp();
  while( ptr < root ) {
    printf( "%s: frame: %08x caller: %08x\n", name, *ptr, ptr[1] );
    ptr =3D *(char ***)ptr;
  }
  printf( "\n" );
}


void func3() {
  callstack( "func3" );
}

void func2() {
  callstack( "func2" );
  func3();
}

void func1() {
  func3();
  func2();
}

int main( int argc, char * argv[] ) {
  int temp;
  init_callstack( &temp );
  func1();
  return 0;
}


C:\temp>stack.exe
func3: frame: 0012ff6c caller: 0040106d
func3: frame: 0012ff74 caller: 00401091
func3: frame: 0012ff80 caller: 004010ad

func2: frame: 0012ff6c caller: 0040107f
func2: frame: 0012ff74 caller: 00401096
func2: frame: 0012ff80 caller: 004010ad

func3: frame: 0012ff64 caller: 0040106d
func3: frame: 0012ff6c caller: 00401087
func3: frame: 0012ff74 caller: 00401096
func3: frame: 0012ff80 caller: 004010ad


Turning the addresses to actual function named needs the use of the=20
DbgHelp library, which can open and use pdb files for you.

Cheers,

				/ h+

-----Original Message-----
From: gam...@li...
[mailto:gam...@li...]On Behalf Of
Diogo de Andrade
Sent: Wednesday, February 09, 2005 5:43 AM
To: gam...@li...
Subject: [GD-Windows] Call stack


Hey all!

This is a long shot, but is there any way to output the call stack in
MSVC 6.0?
More precisely, I'm having some problems with the recording/playback in
my engine, and I tracked it to the random number generated (when on
playback, it is being called more often than on recording, which is a
strange behavior). Anyway, I'm logging the output of the random number
generator (this is how I found out the asynch), but it would be really
helpful if I could output to the log file the call stack just to see
exactly WHO is calling the random number generator out of turn...
Is there any "simple" way of doing this (just for debug purposes)?

Thanks in advance (and sorry if this is OT, but I'm not sure)

Diogo de Andrade
Creative and Technical Director
Spellcaster Studios
dio...@sp...






-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick
_______________________________________________
Gamedevlists-windows mailing list
Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=3D555



-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick
_______________________________________________
Gamedevlists-windows mailing list
Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=3D555

RE: [GD-Windows] Call stack

[Grills, J. <jg...@so...>]

I have very similar code for walking the stack.  My code works on many
versions of windows, but it has stopped working as of Windows XP SP2.  I
even did a test on a machine where my test program worked, I upgraded to
SP2, and the test program stopped working, so I know it had something to
do with the SP2 upgrade.  StackWalk won't walk the whole stack - it just
gives up after returning 2 call stack frames.  I've ended up having to
rewrite StackWalk myself (which isn't very hard, if you disable the
"Frame pointer omission" optimization, since you can use the EBP
register to walk up the stack).  I've complained to Microsoft about
this, but they haven't been very quick to try and fix it.  Anyone else
run into the same problem on XP SP2?  If it's just me, then I'll start
comparing my code against other implementations (like the one below) to
see what I may doing wrong.  If it's not just me, I'd be happy to share
my stack walking code so that others can work around the issue..
=20
One thing I do differently than most other implementations I've seen is
that I don't do any symbol lookup while walking the stack.  I have
plenty of cases in the engine where I capture a call stack for debugging
later if something else goes wrong, and the symbol lookup may not be
necessary.  For instance, our memory manager can grab a stack for every
allocation, and at application exit time we can report the call stack
for any memory leaks.  Typically all allocations are properly freed, so
looking up the symbols when capturing the call stack during an
allocation is unnecessary and also seriously impacts performance.  In
the rare case of a memory leak (most often discovered in new code that
hasn't been submitted), very few addresses need to be translated into
symbols.
=20
Jeff Grills=20
Senior Technical Director, Austin Studio=20
Sony Online Entertainment=20

	-----Original Message-----
	From: gam...@li...
[mailto:gam...@li...] On Behalf Of
Simon O'Connor
	Sent: Wednesday, February 09, 2005 7:58 AM
	To: gam...@li...
	Subject: Re: [GD-Windows] Call stack
=09
=09

=09
	I wrote some code a while back that sounds like what you're
after - or at least should point you in the right direction:

=09
http://www.sc3d.com/thought/snippets/debugging/GetFunctionCaller/

	NB: It's not quite what I'd class as "production" quality - it
was written for debugging purposes only :o)

	Cheers,
=09
	Simon O'Connor
	Programmer @ Reflections Interactive
	& Microsoft DirectX MVP
=09
=09

RE: [GD-Windows] Call stack

[Jon W. <hp...@mi...>]

In debug mode, you can walk the EBP frames pretty easily. Use 
some inline assembly or setjmp() to read the value of the register 
and then treat it like a pointer-to-void*, and you're good to go.

In release mode, it might still be possible to do, assuming you 
don't turn on the "omit frame pointer" optimization. Sadly, that 
optimization gives you another whole register to play with, so it's 
usually something you'll want for your shipping builds. At that 
point, I've had reasonable success with scanning the stack for 
addresses that point right after a CALL or JMP REG instruction 
(JMP REG is used for DLL/vectored calls).

In debug mode, it looks like this:


#include <stdio.h>


__declspec( naked ) char ** get_ebp() {
  __asm {
    mov eax, ebp
    ret
  }
}

// the stack root needs to be thread-local for multi-threaded
// applications
char ** root = 0;
void init_callstack( void * ptr ) {
  root = (char **)ptr;
}

void callstack( char const * name ) {
  char ** ptr = get_ebp();
  while( ptr < root ) {
    printf( "%s: frame: %08x caller: %08x\n", name, *ptr, ptr[1] );
    ptr = *(char ***)ptr;
  }
  printf( "\n" );
}


void func3() {
  callstack( "func3" );
}

void func2() {
  callstack( "func2" );
  func3();
}

void func1() {
  func3();
  func2();
}

int main( int argc, char * argv[] ) {
  int temp;
  init_callstack( &temp );
  func1();
  return 0;
}


C:\temp>stack.exe
func3: frame: 0012ff6c caller: 0040106d
func3: frame: 0012ff74 caller: 00401091
func3: frame: 0012ff80 caller: 004010ad

func2: frame: 0012ff6c caller: 0040107f
func2: frame: 0012ff74 caller: 00401096
func2: frame: 0012ff80 caller: 004010ad

func3: frame: 0012ff64 caller: 0040106d
func3: frame: 0012ff6c caller: 00401087
func3: frame: 0012ff74 caller: 00401096
func3: frame: 0012ff80 caller: 004010ad


Turning the addresses to actual function named needs the use of the 
DbgHelp library, which can open and use pdb files for you.

Cheers,

				/ h+

-----Original Message-----
From: gam...@li...
[mailto:gam...@li...]On Behalf Of
Diogo de Andrade
Sent: Wednesday, February 09, 2005 5:43 AM
To: gam...@li...
Subject: [GD-Windows] Call stack


Hey all!

This is a long shot, but is there any way to output the call stack in
MSVC 6.0?
More precisely, I'm having some problems with the recording/playback in
my engine, and I tracked it to the random number generated (when on
playback, it is being called more often than on recording, which is a
strange behavior). Anyway, I'm logging the output of the random number
generator (this is how I found out the asynch), but it would be really
helpful if I could output to the log file the call stack just to see
exactly WHO is calling the random number generator out of turn...
Is there any "simple" way of doing this (just for debug purposes)?

Thanks in advance (and sorry if this is OT, but I'm not sure)

Diogo de Andrade
Creative and Technical Director
Spellcaster Studios
dio...@sp...






-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Gamedevlists-windows mailing list
Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=555

Re: [GD-Windows] Call stack

[Alen L. <ale...@cr...>]

Start here: http://www.microsoft.com/msj/0497/hood/hood0497.aspx . Matt
Pietrek wrote some articles and excellent examples on how to walk stack both
manually and using dbghelp.dll. I prefer using manual walk and recording
just the plain addresses in runtime and then later when dumping decode them
to actual symbols.

YMMV,
Alen

----- Original Message -----
From: "Diogo de Andrade" <Dio...@ne...>
To: <gam...@li...>
Sent: Wednesday, February 09, 2005 13:43
Subject: [GD-Windows] Call stack


> Hey all!
>
> This is a long shot, but is there any way to output the call stack in
> MSVC 6.0?
> More precisely, I'm having some problems with the recording/playback in
> my engine, and I tracked it to the random number generated (when on
> playback, it is being called more often than on recording, which is a
> strange behavior). Anyway, I'm logging the output of the random number
> generator (this is how I found out the asynch), but it would be really
> helpful if I could output to the log file the call stack just to see
> exactly WHO is calling the random number generator out of turn...
> Is there any "simple" way of doing this (just for debug purposes)?
>
> Thanks in advance (and sorry if this is OT, but I'm not sure)
>
> Diogo de Andrade
> Creative and Technical Director
> Spellcaster Studios
> dio...@sp...
>
>
>
>
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> Gamedevlists-windows mailing list
> Gam...@li...
> https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
> Archives:
> http://sourceforge.net/mailarchive/forum.php?forum_id=555
>

RE: [GD-Windows] Call stack

[Diogo de A. <Dio...@ne...>]

Is there something I should enable/include/etc to make this work? I've
tried creating a simple console application and added the files
available in the URL you provided, but it just failed. Debug output:
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
GetFunctionCaller() : SymFromAddr() failed. GetLastError()=3D=3D487
GetFunctionCaller() : SymGetLineFromAddr64() failed. =
GetLastError()=3D=3D487
=20
(0) : ()
=20
The thread 0xED8 has exited with code 0 (0x0).
=20
Maybe I forgot some flag in the compile settings.
Anyway, thanks for the help.
Diogo de Andrade
Director Criativo e de Tecnologia
Spellcaster Studios
dio...@sp...
-----Original Message-----
From: gam...@li...
[mailto:gam...@li...] On Behalf Of
Simon O'Connor
Sent: quarta-feira, 9 de Fevereiro de 2005 13:58
To: gam...@li...
Subject: Re: [GD-Windows] Call stack
=20

I wrote some code a while back that sounds like what you're after - or
at least should point you in the right direction:
http://www.sc3d.com/thought/snippets/debugging/GetFunctionCaller/
NB: It's not quite what I'd class as "production" quality - it was
written for debugging purposes only :o)
Cheers,

Simon O'Connor
Programmer @ Reflections Interactive
& Microsoft DirectX MVP

--- "Diogo de Andrade" <Dio...@ne...> wrote:

From: "Diogo de Andrade" <Dio...@ne...>
Date: Wed, 9 Feb 2005 13:43:16 -0000
To: <gam...@li...>
Subject: [GD-Windows] Call stack

Hey all!=20

This is a long shot, but is there any way to output the call stack in=20
MSVC 6.0?=20
More precisely, I'm having some problems with the recording/playback in=20
my engine, and I tracked it to the random number generated (when on=20
playback, it is being called more often than on recording, which is a=20
strange behavior). Anyway, I'm logging the output of the random number=20
generator (this is how I found out the asynch), but it would be really=20
helpful if I could output to the log file the call stack just to see=20
exactly WHO is calling the random number generator out of turn...=20
Is there any "simple" way of doing this (just for debug purposes)?=20

Thanks in advance (and sorry if this is OT, but I'm not sure)=20

Diogo de Andrade=20
Creative and Technical Director=20
Spellcaster Studios=20
dio...@sp...=20






-------------------------------------------------------=20
SF email is sponsored by - The IT Product Guide=20
Read honest & candid reviews on hundreds of IT Products from real users.

Discover which products truly live up to the hype. Start reading now.=20
http://ads.osdn.com/?ad_id=3D6595&alloc_id=3D14396&op=3Dclick=20
_______________________________________________=20
Gamedevlists-windows mailing list=20
Gam...@li...=20
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows=20
Archives:=20
http://sourceforge.net/mailarchive/forum.php?forum_id=3D555
------------------------------------------------------- SF email is
sponsored by - The IT Product Guide Read honest & candid reviews on
hundreds of IT Products from real users. Discover which products truly
live up to the hype. Start reading now.
http://ads.osdn.com/?ad_ide95&alloc_id=14396&op=3Dick
_______________________________________________ Gamedevlists-windows
mailing list Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives: http://sourceforge.net/mailarchive/forum.php?forum_idU5

Re: [GD-Windows] Call stack

[Simon O'C. <si...@sc...>]

<html><body><P><FONT face=3DArial size=3D2><BR>I wrote some code a while ba=
ck that sounds like what you're after - or at least&nbsp;should point you i=
n the right direction:</FONT></P>
<P><FONT style=3D"FONT-FAMILY: arial,helvetica,sans-serif" size=3D2><A href=
=3D"http://www.sc3d.com/thought/snippets/debugging/GetFunctionCaller/">http=
://www.sc3d.com/thought/snippets/debugging/GetFunctionCaller/</A></FONT></P>
<P><FONT style=3D"FONT-FAMILY: arial,helvetica,sans-serif" size=3D2>NB: It'=
s not quite what I'd class as "production" quality - it was written&nbsp;fo=
r debugging purposes&nbsp;only :o)</P>
<P>Cheers,<BR><BR>Simon O'Connor<BR>Programmer @ Reflections Interactive<BR=
>&amp; Microsoft DirectX MVP<BR><BR>--- "Diogo de Andrade" &lt;Diogo.Andrad=
e...@ne...&gt; wrote:<BR><BR>From: "Diogo de Andrade" &lt;Diogo.Andrade@=
netvisao.pt&gt;<BR>Date: Wed, 9 Feb 2005 13:43:16 -0000<BR>To: &lt;gamedevl=
ist...@li...&gt;<BR>Subject: [GD-Windows] Call stack<=
BR><BR>Hey all! <BR><BR>This is a long shot, but is there any way to output=
 the call stack in <BR>MSVC 6.0? <BR>More precisely, I'm having some proble=
ms with the recording/playback in <BR>my engine, and I tracked it to the ra=
ndom number generated (when on <BR>playback, it is being called more often =
than on recording, which is a <BR>strange behavior). Anyway, I'm logging th=
e output of the random number <BR>generator (this is how I found out the as=
ynch), but it would be really <BR>helpful if I could output to the log file=
 the call stack just to see <BR>exactly WHO is calling the random number ge=
nerator out of turn... <BR>Is there any "simple" way of doing this (just fo=
r debug purposes)? <BR><BR>Thanks in advance (and sorry if this is OT, but =
I'm not sure) <BR><BR>Diogo de Andrade <BR>Creative and Technical Director =
<BR>Spellcaster Studios <BR>dio...@sp... <BR><BR><B=
R><BR><BR><BR><BR>------------------------------------------------------- <=
BR>SF email is sponsored by - The IT Product Guide <BR>Read honest &amp; ca=
ndid reviews on hundreds of IT Products from real users. <BR>Discover which=
 products truly live up to the hype. Start reading now. <BR>http://ads.osdn=
.com/?ad_id=3D6595&amp;alloc_id=3D14396&amp;op=3Dclick <BR>________________=
_______________________________ <BR>Gamedevlists-windows mailing list <BR>G=
ame...@li... <BR>https://lists.sourceforge.net=
/lists/listinfo/gamedevlists-windows <BR>Archives: <BR>http://sourceforge.n=
et/mailarchive/forum.php?forum_id=3D555<BR></P></FONT></body></html>

[GD-Windows] Call stack

[Diogo de A. <Dio...@ne...>]

Hey all!

This is a long shot, but is there any way to output the call stack in
MSVC 6.0?
More precisely, I'm having some problems with the recording/playback in
my engine, and I tracked it to the random number generated (when on
playback, it is being called more often than on recording, which is a
strange behavior). Anyway, I'm logging the output of the random number
generator (this is how I found out the asynch), but it would be really
helpful if I could output to the log file the call stack just to see
exactly WHO is calling the random number generator out of turn...
Is there any "simple" way of doing this (just for debug purposes)?

Thanks in advance (and sorry if this is OT, but I'm not sure)

Diogo de Andrade
Creative and Technical Director
Spellcaster Studios
dio...@sp...

RE: [GD-Windows] STL Containers and __declspec(export/import)

[Jon W. <hp...@mi...>]

You don't want to be exporting classes with implementation details, 
anyway. Especially the STL isn't binary compatible at all between 
revisions. Could you re-formulate the interface for your exported 
class using a pure abstract base class, and export that using a 
factory function? Vtable layout is a pretty safe linkage dependency.

Cheers,

			/ h+

-----Original Message-----
From: gam...@li...
[mailto:gam...@li...]On Behalf Of
Chris Raine
Sent: Sunday, February 06, 2005 2:26 PM
To: gam...@li...
Subject: [GD-Windows] STL Containers and __declspec(export/import)


Hi, 

If you stick a STL container to a class with __declspec(export) vcc will
whine about the STL container not being declared with __declspec(export)
attribute. What I found after a bit of doc-digging was that MSDN
suggests that you explicitly instantiate the STL container with
'extern'. 

RE: [GD-Windows] STL Containers and __declspec(export/import)

[Vasco L. <va...@kr...>]

At 00:47 07.02.2005, you wrote:
>Your guess is correct.
>
>Exporting the STL containers is a Bad Thing(tm).
>
>Implementations of the STL are not necessarily binary compatible.  There is
>no convenient way to translate between different implementations, so if the
>DLL uses a different version than your executable, the objects will differ
>and your program will fail.

afaik this is a general problem of c++ (no general C++ ABI, see also 
http://aegisknight.org/cppinterface.html ), so in principle the same 
applies to normal c++ objects.

>Vectors are a special case.  The latest versions of the C++ standard have a
>guarantee that vectors are contiguous, and all major implementations before
>that guarantee also have that behavior, so passing vectors around as general
>arrays can be done safely.
>
>I seem to remember also that the VC++ 7.1 uses local storage in the
>implementation, so in that case, even moving between the same implementation
>would fail if you crossed EXE/DLL boundaries.

Not if you link all your modules to the DLL runtime, again all with the 
same version.

I use lots of stl and boost with my DLLs. As long as you make sure that all 
components are compiled with same compiler (+ same version) and linked with 
the runtime DLL it works without any problems. Otherwise it really gets 
involved: only C functions and pure abstract C++ types allowed and 
seperated memory managment between modules.

Vasco  

RE: [GD-Windows] STL Containers and __declspec(export/import)

[Bryan W. <br...@xm...>]

Your guess is correct.

Exporting the STL containers is a Bad Thing(tm).

Implementations of the STL are not necessarily binary compatible.  There =
is
no convenient way to translate between different implementations, so if =
the
DLL uses a different version than your executable, the objects will =
differ
and your program will fail.

Vectors are a special case.  The latest versions of the C++ standard =
have a
guarantee that vectors are contiguous, and all major implementations =
before
that guarantee also have that behavior, so passing vectors around as =
general
arrays can be done safely.

I seem to remember also that the VC++ 7.1 uses local storage in the
implementation, so in that case, even moving between the same =
implementation
would fail if you crossed EXE/DLL boundaries.

Hope that explains things a bit.

Bryan.

-----Original Message-----
From: gam...@li...
[mailto:gam...@li...] On Behalf Of =
Chris
Raine
Sent: Sunday, February 06, 2005 3:26 PM
To: gam...@li...
Subject: [GD-Windows] STL Containers and __declspec(export/import)

Hi,=20

If you stick a STL container to a class with __declspec(export) vcc will
whine about the STL container not being declared with __declspec(export)
attribute. What I found after a bit of doc-digging was that MSDN
suggests that you explicitly instantiate the STL container with
'extern'.=20

Having done that and still having no success with getting around the
annoying warning, I found out that the same MSDN-document had a very
small note stating :

<quote>=20
"The only STL container that can currently be exported is vector. The
other containers (that is map, set,queue, list, deque) all contain
nested class and [therefore] cannot be exported."
<quote/>

The document I'm referring to is entitled "Howto: Exporting STL
Components Inside & Outside of a Class" and was last updated on
5-7-2001. Has there been any update on this topic for Studio .Net 2003
(which is what I am using) that I could not find?

Am I correct with guessing that using the STL container classes in
external header files of a dll is a no-no and should be avoided?
Ignoring that warning would be dangerous (for I do not like the concept
of fixing problems by disabling the warnings they create).

confused and thanking for any insight given regarding this topic,=20
Chris=20





-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Gamedevlists-windows mailing list
Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=3D555

[GD-Windows] STL Containers and __declspec(export/import)

[Chris R. <c....@gm...>]

Hi, 

If you stick a STL container to a class with __declspec(export) vcc will
whine about the STL container not being declared with __declspec(export)
attribute. What I found after a bit of doc-digging was that MSDN
suggests that you explicitly instantiate the STL container with
'extern'. 

Having done that and still having no success with getting around the
annoying warning, I found out that the same MSDN-document had a very
small note stating :

<quote> 
"The only STL container that can currently be exported is vector. The
other containers (that is map, set,queue, list, deque) all contain
nested class and [therefore] cannot be exported."
<quote/>

The document I'm referring to is entitled "Howto: Exporting STL
Components Inside & Outside of a Class" and was last updated on
5-7-2001. Has there been any update on this topic for Studio .Net 2003
(which is what I am using) that I could not find?

Am I correct with guessing that using the STL container classes in
external header files of a dll is a no-no and should be avoided?
Ignoring that warning would be dangerous (for I do not like the concept
of fixing problems by disabling the warnings they create).

confused and thanking for any insight given regarding this topic, 
Chris 

[GD-Windows] How to use the interface DWebBrowserEvents2 only in Win32 ?

[<ge...@fr...>]

Hello,

Another new question about plain Win32 programming ... I cannot use MFC, =
or any=20
other kind of library.

So here is my question, for one of my developement i need to open an =
internet=20
explorer instance. All is good for now, the html page is loading, we can =
browse it,
but if the user quit the window i don't get the notification of this =
close so i cannot=20
release my object cleanly.=20

After searching on MSDN i found the interface DWebBrowserEvents2, but =
sadly there=20
is almost no information about how to instantiate it and how to make it =
cooperate=20
with me IWebBrowser2 ...=20

Is anyone out there that already did that ?=20

Fabien.

RE: [GD-Windows] mutiple cameras

[Andrew G. <ag...@cl...>]

You just render to different portions of the same render target, which will
normally be your backbuffer.

A.
 

> -----Original Message-----
> From: gam...@li... 
> [mailto:gam...@li...] On 
> Behalf Of Brett Bibby
> Sent: Sunday, January 30, 2005 4:12 PM
> To: Gam...@li...
> Subject: [GD-Windows] mutiple cameras
> 
> What is the recommended way to handle multiple cameras/views 
> per "window" in both DirectX and OpenGL (e.g. a car racing 
> game with a rear view mirror, or split screen 2 player)?  For 
> example, for DX is it normal to use additional swap chains 
> for each camera viewport or to use render targets for each 
> camera and then copy the surface to the framebuffer when 
> done?  Or is there another way?  How about OpenGL?
> Cheers,
> -bb
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive 
> Reporting Tool for open source databases. Create drag-&-drop 
> reports. Save time by over 75%! Publish reports on the web. 
> Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Gamedevlists-windows mailing list
> Gam...@li...
> https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
> Archives:
> http://sourceforge.net/mailarchive/forum.php?forum_id=555
> 

[GD-Windows] mutiple cameras

[Brett B. <res...@ga...>]

What is the recommended way to handle multiple cameras/views per "window" in
both DirectX and OpenGL (e.g. a car racing game with a rear view mirror, or
split screen 2 player)?  For example, for DX is it normal to use additional
swap chains for each camera viewport or to use render targets for each
camera and then copy the surface to the framebuffer when done?  Or is there
another way?  How about OpenGL?
Cheers,
-bb

Re: [GD-Windows] ActiveX Control question

[Brian H. <bri...@gm...>]

> That was sort of my original question, stated differently =)  I'm just
> not sure where long-term work can be stored, all the examples I see
> are basically focused on property exchange, event reaction, and doing
> things in OnDraw().

What I ended up doing, and I think this is okay, is having an
OnCreate() for the control which turns around and does a
SendMessage()...then I turn around and have an OnCommand() which calls
"Go()" which spawns the worker thread and goes from there.  Whew. 
Ugly, but it seems to work.

Re: [GD-Windows] ActiveX Control question

[Brian H. <bri...@gm...>]

> Can you do your setup work, including launching a thread, in the constructor?

That was sort of my original question, stated differently =)  I'm just
not sure where long-term work can be stored, all the examples I see
are basically focused on property exchange, event reaction, and doing
things in OnDraw().

Re: [GD-Windows] ActiveX Control question

[Ivan-Assen I. <iva...@gm...>]

> I looked for references to the OnLoad event -- where is this
> documented?  I didn't see it under COleControl.

Errrr, sorry to have mislead you... OnLoad is the .NET event. 

COleControl is an MFC class, right? 
Can you do your setup work, including launching a thread, in the constructor?

Re: [GD-Windows] ActiveX Control question

[Brian H. <bri...@gm...>]

I looked for references to the OnLoad event -- where is this
documented?  I didn't see it under COleControl.

RE: [GD-Windows] ActiveX Control question

[Alastair P. <apa...@sa...>]

Hi Brian,

I don't know exactly what you're doing but, I have written some browser
plugins before and I found the Netscape plugin API to be much easier to
use for simple stuff than ActiveX. Last time I checked, Netscape plugins
also worked with IE and Mozilla on multiple platforms.
http://www.mozilla.org/projects/plugins/.

You might also consider Java applets, although unfortunately these no
longer work "out of the box" with IE.

Al


-----Original Message-----
From: gam...@li...
[mailto:gam...@li...] On Behalf Of
Brian Hook
Sent: Tuesday, January 11, 2005 11:02 AM
To: gam...@li...
Subject: Re: [GD-Windows] ActiveX Control question

> Assuming you get the COM threading models right, I believe the
"proper" way
> would the to launch a separate thread in the OnLoad event. Have you
tried that?

No, but I will now =3D)  The amount of information to digest for COM +
OLE + MFC + ActiveX + the whole ATL distraction thing is just, well,
immense.  Just figuring out where to start has been difficult, most
people who have been doing this stuff have been doing it for 7 years.

Thanks,

Brian


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
_______________________________________________
Gamedevlists-windows mailing list
Gam...@li...
https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_id=3D555

Re: [GD-Windows] ActiveX Control question

[Brian H. <bri...@gm...>]

> Assuming you get the COM threading models right, I believe the "proper" way
> would the to launch a separate thread in the OnLoad event. Have you tried that?

No, but I will now =)  The amount of information to digest for COM +
OLE + MFC + ActiveX + the whole ATL distraction thing is just, well,
immense.  Just figuring out where to start has been difficult, most
people who have been doing this stuff have been doing it for 7 years.

Thanks,

Brian

RE: [GD-Windows] ActiveX Control question

[Ivan-Assen I. <as...@ha...>]

> I also have a specific question:  is there a proper way to implement a
> control that is completely passive, i.e. accepts no input and just
> does something.  Imagine an application that allows you to download an
> OCX to perform a bunch of calculations and then upload the results
> (kind of like SETI at home, but as an explicit Web page control, not
> as an explicit downloadable).

Assuming you get the COM threading models right, I believe the "proper" way
would the to launch a separate thread in the OnLoad event. Have you tried that?
The main thread of a "proper" Windows application with a user interface 
(including an ActiveX control) should ideally do nothing but react to events.

[GD-Windows] ActiveX Control question

[Brian H. <bri...@gm...>]

I've been reading up on ActiveX control development, and I'm looking
for a good book that covers ActiveX in depth (as opposed to the books
out there that spend 1000 pages on MFC and then devote 50 pages to an
ActiveX clock example written using MFC/ATL).  Any suggestions?

I also have a specific question:  is there a proper way to implement a
control that is completely passive, i.e. accepts no input and just
does something.  Imagine an application that allows you to download an
OCX to perform a bunch of calculations and then upload the results
(kind of like SETI at home, but as an explicit Web page control, not
as an explicit downloadable).

The typical control reacts to events, e.g. with OnXXXX response. 
Should I just setup a timer that periodically calls a callback
function, or is there a better/cleaner weay?

Re: [GD-Windows] useful VS.NET plugins

[Chris S. <cs...@st...>]

   A couple of plugins I find useful are Fast Solution Build and Project Line 
Counter.  Fast Solution Build makes building projects in .NET much faster by 
skipping redundant dependency checks.  Project Line Counter gives you statistics 
on line counts, comment percentages, etc.

Fast Solution Build: http://workspacewhiz.com/OtherAddins.html
Project Line Counter: http://www.wndtabs.com/plc/

There's also source code available for both plugins:
http://www.codeproject.com/macro/fastsolutionbuild.asp
http://www.codeproject.com/macro/linecount.asp

Chris

Daniel Vogel wrote:
>>missing from .NET as I always thought it was 
>>everything-including-the-kitchen-sink IDE.
> 
> 
> Not out of the box though at least VS.NET 2005 is a big step in the
> right direction.
> 
> I couldn't live without Workspace Whiz and Visual Assist to supplement
> VS.NET 2003's IDE.
> 
> http://www.workspacewhiz.com/
> http://www.wholetomato.com/ 
> 
> Can anyone else recommend any other useful plugins? I wouldn't mind
> finding one with more control over syntax highlighting.
> 
> -- Daniel, Epic Games Inc.

[GD-Windows] useful VS.NET plugins

[Daniel V. <Dan...@ep...>]

> missing from .NET as I always thought it was=20
> everything-including-the-kitchen-sink IDE.

Not out of the box though at least VS.NET 2005 is a big step in the
right direction.

I couldn't live without Workspace Whiz and Visual Assist to supplement
VS.NET 2003's IDE.

http://www.workspacewhiz.com/
http://www.wholetomato.com/=20

Can anyone else recommend any other useful plugins? I wouldn't mind
finding one with more control over syntax highlighting.

-- Daniel, Epic Games Inc.

> -----Original Message-----
> From: gam...@li...=20
> [mailto:gam...@li...] On=20
> Behalf Of Brett Bibby
> Sent: Saturday, January 08, 2005 11:39 PM
> To: gam...@li...
> Subject: Re: [GD-Windows] declspec?
>=20
> > I usually just re-open the solution and do a full rebuild=20
> whenever I=20
> > get really bizarre problems that *shouldn't* happen :)
> >
>  I'm new to .NET and I thought when I switched from=20
> CodeWarrior bizarre problems would go away :-)  sigh.  I've=20
> been shocked though at how many basic, simple features are=20
> missing from .NET as I always thought it was=20
> everything-including-the-kitchen-sink IDE.
> --bb
>=20
>=20
>=20
> -------------------------------------------------------
> The SF.Net email is sponsored by: Beat the post-holiday blues=20
> Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
> It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
> _______________________________________________
> Gamedevlists-windows mailing list
> Gam...@li...
> https://lists.sourceforge.net/lists/listinfo/gamedevlists-windows
> Archives:
> http://sourceforge.net/mailarchive/forum.php?forum_id=3D555
>=20

Re: [GD-Windows] declspec?

[Brett B. <res...@ga...>]

> I usually just re-open the solution and do a full rebuild whenever I get
> really bizarre problems that *shouldn't* happen :)
>
 I'm new to .NET and I thought when I switched from CodeWarrior bizarre
problems would go away :-)  sigh.  I've been shocked though at how many
basic, simple features are missing from .NET as I always thought it was
everything-including-the-kitchen-sink IDE.
--bb