mingw-users Mailing List for MinGW - Minimalist GNU for Windows

On 2006-5-31 4:42 UTC, Michael Gerdau wrote:
> 
> http://blogs.msdn.com/calvin_hsia/archive/2005/01/24/359713.aspx
> (Admittedly one source is sufficient :-)
> 
> I'm not sure this counts as an official M$ source but since this
> seems to be a M$ blog I suppose we are safe.

The "Terms of Use" link at the bottom of this page
  http://blogs.msdn.com/
links to
  http://www.microsoft.com/info/cpyright.mspx
which says
  "You may not modify, copy, distribute, transmit, display, perform,
  reproduce, publish, license, create derivative works from, transfer,
  or sell any information, software, products or services obtained
  from the Services."

It looks like anyone can sign up and blog there; if so, then this
would seem little different from newsgroups hosted on a ms server,
which often contain proprietary information--e.g.,
  http://groups.google.com/group/microsoft.public.dotnet.languages.csharp/msg/2b4c737823066c77

Quoting Michael Gerdau <mg...@te...>:

>
> From the suggested 7 sources at most 1 might be valid:
> http://blogs.msdn.com/calvin_hsia/archive/2005/01/24/359713.aspx
> (Admittedly one source is sufficient :-)
>

Actually, this reference is stated from Visual FoxPro sources so it is 
also a "no way".  The Intel reference is the one I thought seemed to be 
most like a usable resource.

Earnie Boyd

http://shop.siebunlimited.com

> Web pages where the definition of PBT_APMRESUMEAUTOMATIC can be found:
> You can decide if any of these are "good enough" documentation.

Ok, lets have a look at all of these:

> http://www.intel.com/cd/ids/developer/asmo-na/eng/76359.htm?prn=3DY

Does claim PBT_APMAUTOMATICRESUE has a value of 0x12. The text gives
hints that PBT_APMAUTOMATICRESUE actually might mean PBT_APMAUTOMATICRESUME
and the former is a typo, but that is speculation. Both of theses names
are not known by MSDN AFAIK.

One might speculate wether PBT_APMAUTOMATICRESUME actually should
have been PBT_APMRESUMEAUTOMATIC.

> http://www.spampal.org.uk/cgi-bin/viewcvs/viewcvs.cgi/spampal/spampal.c?r=
ev=3D1.12

This explicitly cites the file pbt.h as source and is therefor no way.

> http://blogs.msdn.com/calvin_hsia/archive/2005/01/24/359713.aspx

This does give indeed the values for some if not most of the PBT_*
values and seems to be a valid source.

> http://blog.kellybrownsberger.com/archive/2005/05/11/293.aspx

I can't open that page (error HTTP 404: The resource cannot be found.)

> http://www.eggheadcafe.com/aspnet_answers/NETFrameworkdrawing/Jan2006/pos=
t25815483.asp

Does give quite a few values without actually naming the source. From
the rest of the page I'd assume it comes from the PSDK (i.e. no way).

> http://blog.hexun.com/isea/1935895_d.html

This page does not provide where the info comes from, or at least
I'm not reading japaneese and thus am not able to decide that.

> http://weblogs.asp.net/ralfw/archive/2003/09/09/26908.aspx

This page gives the PSDK as source (i.e. no way).


=46rom the suggested 7 sources at most 1 might be valid:
http://blogs.msdn.com/calvin_hsia/archive/2005/01/24/359713.aspx
(Admittedly one source is sufficient :-)

I'm not sure this counts as an official M$ source but since this
seems to be a M$ blog I suppose we are safe.

Best,
Michael
=2D-=20
 Vote against SPAM - see http://www.politik-digital.de/spam/
 Michael Gerdau       email: mg...@te...
 GPG-keys available on request or at public keyserver

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3757330
By: wszafran

Actually it's the Lesser GPL but that probably makes no difference.

There's no way out then?  All publicly available sources of this information
are either copyrighted or licensed in some way.

One has to wonder how did the other values make it into the pbt.h header in
the first place, as they don't seem to be explicitly documented by MSDN either.

Best regards.


______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3757221
By: earnie

WINE headers have a copyright that belongs to the WINE project and are subject
to GPL.  We cannot plagiarize them and we cannot create modified works based
on them.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3757199
By: wszafran

I agree, I must admit I don't know what I was thinking, my suggestion was plain
silly, thanks for pointing that out.

The only reference the MSDN is giving is the note that the values for the power
events are defined in the pbt.h header.  I looked around on the Web a bit and
found several other references listing the proper value as 18 (or 0x0012).

If the definition seen in one of the Wine project headers is good enough as
a reference basis, then here's the link to the full listing of their winuser.h
header, which contains values for the power events:

http://source.winehq.org/source/include/winuser.h

Best regards.


______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Good grief.

Web pages where the definition of PBT_APMRESUMEAUTOMATIC can be found:
You can decide if any of these are "good enough" documentation.

http://www.intel.com/cd/ids/developer/asmo-na/eng/76359.htm?prn=Y
http://www.spampal.org.uk/cgi-bin/viewcvs/viewcvs.cgi/spampal/spampal.c?rev=1.12
http://blogs.msdn.com/calvin_hsia/archive/2005/01/24/359713.aspx
http://blog.kellybrownsberger.com/archive/2005/05/11/293.aspx
http://www.eggheadcafe.com/aspnet_answers/NETFrameworkdrawing/Jan2006/post25815483.asp
http://blog.hexun.com/isea/1935895_d.html
http://weblogs.asp.net/ralfw/archive/2003/09/09/26908.aspx

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3757105
By: dessent

The values associated with these #defines are *not* arbitrary, and there is
no way just to pick them at will.  They are defined precisely by the API/ABI,
as dictated solely by Microsoft and the PSDK.  Mingw has absolutely no choice
in this matter, because the code that you are calling that consumes them is
Microsoft's.  If someone were to just make up arbitrary values and stick them
in the header file, it would have disastrous results.

The problem is that we as an open source project cannot use Microsoft's headers
as a basis for any work, so as to remove any question of taint.  This means
that these numbers have to be gathered from other information sources, which
usually means somewhere on the msdn site.  All of the existing headers were
created in this way.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3757075
By: wszafran

> Ir0nh34d's point was that *you* have to give a link to documentation for the
values you need.

I got the point, thank you.  Although I don't entirely agree with it, let's
put that aside now.

> If you can find it at msdn.microsoft.com and point us to it then we can do
something about it.

If only it were so straightforward.  MSDN doesn't mention actual numerical values
for the power events.  The MinGW-supplied pbt.h header already defines values
for several other events, except the one in question.  Someone (Win API maintainer?)
assigned the numeric values to the symbolic names in the past.  I have no way
of knowing the needed value, which I thought I made clear before, and that's
why I made the suggestion how to proceed about it.  But the decision about what
value should actually be used has to be made not by me but rather by either
the Win API maintainer or the project leader.  I think it's a reasonable
expectation.

Best regards.


______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3756166
By: earnie

Ir0nh34d's point was that *you* have to give a link to documentation for the
values you need.  If you can find it at msdn.microsoft.com and point us to it
then we can do something about it.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

In addition to my previous post:

> According to the errormsg you pasted you have a type mismatch in a
> functioncall in file frmlogin.ui.h, line 28.
>=20
> > In file included from frmlogin.cpp:2:
> > frmlogin.ui.h: In member function `int FrmLogin::StartButton_Show_Hide(=
BOOL)':
> > frmlogin.ui.h:28: error: cannot convert `const char*' to `const WCHAR*'=
 for argu
> > ment `1' to `HWND__* FindWindowW(const WCHAR*, const WCHAR*)'
>=20
> This errormsg says you are providing a const single byte character
> pointer (const char*) where the function expects a const 2-byte character
> pointer (const WCHAR*). It happens with the first argument in the call to
> FindWindowW, all on line 28 in file frmlogin.ui.h
>=20
> I don't know what you expect the compiler to do short of fixing
> the bug in your code. How detailed should the errormsg be ?
>=20
> Note that for writing the above I didn't even look into your code !

Meanwhile I looked into your sourcecode. Line 28 in frmlogin.ui.h says
hWnd =3D GetDlgItem(FindWindow(TASKBAR, NULL), ID_STARTBUTTON);

The first parameter to FindWindow (which the compiler complains
about) is the constant TASKBAR which is defined in line 6 of the
same file as
#define     TASKBAR         "Shell_TrayWnd"		// Taskbar class name

=46rom the errormsg I deduce you are compiling a unicode enabled app.
However TASKBAR seems to have been defined as a singlebyte char constant.
and that triggers the error.


In your mail you write you get the above error when compiling from the Qt4
commandprompt. What is the Qt4 command prompt ?

HTH,
Michael
=2D-=20
 Vote against SPAM - see http://www.politik-digital.de/spam/
 Michael Gerdau       email: mg...@te...
 GPG-keys available on request or at public keyserver

> In this example I try to hide the task bar, but I have an error
> with the NULL value this is the error:=20

???

An error with the NULL value ?

According to the errormsg you pasted you have a type mismatch in a
functioncall in file frmlogin.ui.h, line 28.

> In file included from frmlogin.cpp:2:
> frmlogin.ui.h: In member function `int FrmLogin::StartButton_Show_Hide(BO=
OL)':
> frmlogin.ui.h:28: error: cannot convert `const char*' to `const WCHAR*' f=
or argu
> ment `1' to `HWND__* FindWindowW(const WCHAR*, const WCHAR*)'

This errormsg says you are providing a const single byte character
pointer (const char*) where the function expects a const 2-byte character
pointer (const WCHAR*). It happens with the first argument in the call to
=46indWindowW, all on line 28 in file frmlogin.ui.h

I don't know what you expect the compiler to do short of fixing
the bug in your code. How detailed should the errormsg be ?

Note that for writing the above I didn't even look into your code !

> In file included from frmlogin.cpp:2:
> frmlogin.ui.h:36:2: warning: no newline at end of file

This is not an error but a warning which the compiler issues because
apparently the last in frmlogin.ui.h is not properly terminated by
a newline character.

> This compiling with Qt4 Command prompt,but when I Compiled with mingw lik=
e this,
>=20
> g++ -o mainmingw.exe -c mainmingw.cpp
>=20
> I don't have any problem.
>=20
> What's wrong?.......

See above.

HTH,
Michael
=2D-=20
 Vote against SPAM - see http://www.politik-digital.de/spam/
 Michael Gerdau       email: mg...@te...
 GPG-keys available on request or at public keyserver

In this example I try to hide the task bar, but I have an error with the NULL value this is the error:

In file included from frmlogin.cpp:2:
frmlogin.ui.h: In member function `int FrmLogin::StartButton_Show_Hide(BOOL)':
frmlogin.ui.h:28: error: cannot convert `const char*' to `const WCHAR*' for argu
ment `1' to `HWND__* FindWindowW(const WCHAR*, const WCHAR*)'
In file included from frmlogin.cpp:2:
frmlogin.ui.h:36:2: warning: no newline at end of file
mingw32-make[1]: *** [debug\frmlogin.o] Error 1
mingw32-make[1]: Leaving directory `C:/CNet/Cliente'
mingw32-make: *** [debug] Error 2

This compiling with Qt4 Command prompt,but when I Compiled with mingw like this,

g++ -o mainmingw.exe -c mainmingw.cpp

I don't have any problem.

What's wrong?.......

Ivan Kolev wrote:
> 

> How can I change that, i.e. use MSVCRT 7.1? (I can only see libmsvcr70.a and libmsvcr71.a, no 8.0) I tried to change -lmsvcrt to -lmsvcr71 in the lines
> 
> *libgcc:
> %{mthreads:-lmingwthrd} -lmingw32 -lgcc -lmoldname -lmingwex -lmsvcrt
> 
> in the file \MinGW\lib\gcc\mingw32\3.4.5\specs , but the result is that although the program is linked successfully, I get an error "the procedure _ctype was not found in msvcr71.dll". Also, the Dependency Walker shows that Test.exe still depends on msvcrt.dll, although it also depends on msvcr71.dll.

I don't know.  I would have thought that some combination of editing the
specs file and/or including -lmscvrt71 explicitly on the link command
would have worked.  You might also try invoking gcc with -v to see all
the options being silently passed to the subprocesses.

Brian

Brian, many thanks for the detailed explanations.

> user's system.  And this means shipping Microsoft redistributable code
> with your product, which of course is permitted to licensees of the
> Visual Studio product, but I certainly don't know the implications of
> this for someone using only a mingw toolchain.  So for this reason and

Right. I'm not sure about VC 2005 Express, but the free VC 2003 Toolkit *does not* include the dynamic versions of the MS runtime library, only the static ones. This means that MS do not allow free usage of MSVCRT 7.0 or later. I doubt that this has changed in VC 2005 Express.
In that case if I insist to use GCC on Windows, my choices remain the ones already mentioned, either avoiding the supposed bug in MSVCRT by reading text streams as binary (quite a bad idea, it's stupid to make non-trivial changes in a working design because of some bug in the lower level libraries), or switching to Cygwin. The reason I prefer MinGW to Cygwin is that I have no experience with Cygwin (e.g., not sure if I could use an IDE like Code::Blocks with it) and that AFAIK applications built with it will require it to be installed.

> for the reason of wanting to avoid needing to distribute extra things
> with generated binaries, the mingw gcc specs file contains only
> "-lmsvcrt", so even if a newer version is present it won't be used.

How can I change that, i.e. use MSVCRT 7.1? (I can only see libmsvcr70.a and libmsvcr71.a, no 8.0) I tried to change -lmsvcrt to -lmsvcr71 in the lines

*libgcc:
%{mthreads:-lmingwthrd} -lmingw32 -lgcc -lmoldname -lmingwex -lmsvcrt

in the file \MinGW\lib\gcc\mingw32\3.4.5\specs , but the result is that although the program is linked successfully, I get an error "the procedure _ctype was not found in msvcr71.dll". Also, the Dependency Walker shows that Test.exe still depends on msvcrt.dll, although it also depends on msvcr71.dll.

> I don't know why your test case works fine with plain C FILE streams but
> fails with STL.  But as I said earlier I suspect it might be luck,

I'd still want to understand what's going on inside libstdc++. Since I cannot debug it (not sure why, is it Code::Blocks, gdb, or some compilation options), I'm trying to add printf's inside it, but that doesn't change much - it seems that some pre-compiled version is used and any changes in the headers have no effect. How can I disable that, i.e. make GCC re-compile the STL each time?

Thanks,
Ivan

I'm attempting to port an Open Source Linux application to Windows using=
 mingw32 and msys.  It uses memalign.  I checked the mailing list archiv=
es and noticed mingw supplies _mingw_aligned_malloc and _aligned_malloc.=
  Tried both of these (one at a time) in the program in place of the mem=
align call.  Both ran out of memory (returned 0) after 3 calls.  I tried=
 -Wl,--heap and set the heap to some arbitrarily large number during the=
 link phase.  I didn't notice any difference.  Any suggestions?  I'm rea=
lly stuck.  =


While I'm asking for advice, I've been using msys as the build environme=
nt for the program I'm porting.  What's the best practice for setting CP=
ATH so mingw will know where the includes are?  I tried setting it in th=
e control panel as an environment variable, but I used the actual (DOS s=
tyle as in c:\mingw\msys\local\include) path to the includes.  Should I =
be using /usr/local/include instead?  Is there a better locaton to set t=
his (such as a settings file) or is the control panel's environment vari=
ables section the right location to use for setting environment variable=
s used by mingw.

Thanks.
Laura
http://www.distasis.com/cpp


_____________________________________________________________________
Call Anyone, Anytime, Anywhere in the World - FREE!
Free Internet calling from NetZero Voice
Visit http://www.netzerovoice.com today!

Hi !

> Hi everyone I would like to use the windows api functions with my Qt proj=
ect.

That is no problem -- I know it works.

> I  know that I must include the windows.h file

You might also have to include other files from w32api, depending
on the windows api functions you use...

> but when I try to=20
> write some api functions the compiler said that the NULL value is
> diferent that NULL MinGw or something like that. =20

That probably means there is an error in your code or something like
that.

Sigh. Please provide the *EXACT* errormsg.

Providing a selfcontained example together with the exact commands
used, versions of the tools you used, environment you are running
under etc. etc. are helpful as well.

> On the other hand, I found a dll library
> (http://www.codeproject.com/win32/AntonioWinLock.asp).

I can't download the source.

> This dll lock windows desktop (Ctrl+Alt+Del and others),
> and it work well with vc++ and vb, I need to call this dll into my
> Qt project. I wrote into my .pro file the next opctions:   =20
>=20
> LIBS+:=3Dpathtothedll'\winlock.lib .
>=20
> And into frmloging.h #include "winlockdll.h", then I can do that
>=20
> Class FrmLogin : public QDialog
>         {
>            public :
>            ....
>            int CtrlAltDel_Enable_Disable(BOOL bEnableDisable);

Here you declare a memberfunction CtrlAltDel_Enable_Disable which
I deduce from the rest of your writing you don't want. You most likely
should just delete that line.

>             .....
>=20
> }
>=20
> void FrmLogin::myFunction(){
> CtrlAltDel_Enable_Disable(false);
>=20
> }
> but when I try to Call this function I have an error.
> (Undefined the definition from CtrlAltDel_Enable_Disable).=20
>=20
>=20
> What must I do.

Get a book on C++ !!!

You must either provide a memberfunction CtrlAltDel_Enable_Disable:

int FrmLogin::CtrlAltDel_Enable_Disable(BOOL b){
CtrlAltDel_Enable_Disable(false);
    <add your code here>
}

or (and that's what I assume you wish to do) delete it's declaration
from the above classfile.

You definitely should read in the mingwiki http://www.mingw.org/MinGWiki/
the paragraphs on creating a DLL. These do describe in detail all
that is required to get your example working. Also you might want
to check out the FAQ section.

HTH,
Michael
=2D-=20
 Vote against SPAM - see http://www.politik-digital.de/spam/
 Michael Gerdau       email: mg...@te...
 GPG-keys available on request or at public keyserver

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3754335
By: wszafran

I actually can't look at an MS-supplied header because I don't have one.

The MinGW-supplied header in question only contains a handful of definitions,
all with consecutive ordinal numbers.  I don't know how critical the actual
numbers are but I suppose that as long as they are unique, it doesn't matter
that much, so the next available number seems a logical choice for the missing
value.

Best regards.


______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Hi everyone I would like to use the windows api functions with my Qt project.

I  know that I must include the windows.h file but when I try to write some api functions the compiler said that the NULL value is diferent that NULL MinGw or something like that.


On the other hand, I found a dll library (http://www.codeproject.com/win32/AntonioWinLock.asp). This dll lock windows desktop (Ctrl+Alt+Del and others),  and it work well with vc++ and vb, I need to call this dll into my Qt project. I wrote into my .pro file the next opctions:

LIBS+:=pathtothedll'\winlock.lib .

And into frmloging.h #include "winlockdll.h", then I can do that

Class FrmLogin : public QDialog
        {
           public :
           ....
           int CtrlAltDel_Enable_Disable(BOOL bEnableDisable);
            .....

}

void FrmLogin::myFunction(){
CtrlAltDel_Enable_Disable(false);

}
but when I try to Call this function I have an error. (Undefined the definition from CtrlAltDel_Enable_Disable).


What must I do.

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3754276
By: ir0nh34d

It could be added if you know (without looking at a Microsoft supplied pbt.h)
what the value of define should be.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3754269
By: wszafran

Hello,

I was wondering if a missing definition of the PBT_APMRESUMEAUTOMATIC event
could be added to the pbt.h header.

Thanks in advance, best regards.


______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286641

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3754140
By: lmichaels

There are some Linux command line video programs such as mplex, mpgtx, dvdauthor
that are very popular for use when authoring DVDs.  There are  ports of many
of these applications to Cygwin.  Was interested in figuring out if they'd port
to mingw (with msys) as well.  I've been working on getting mjpegtools and mpgtx
to compile.  I've also been trying to get libdvdread to compile so I can build
dvdauthor.  Was just wondering if anyone else has tried this or is in the process
of doing so or is interested in it.  Would like to compare notes on the process.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=290275

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3753218
By: rookie_one

Gah. That's what you get when a noob tries to do something. ;) Thanks!

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286529

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3753205
By: dessent

The name is mangled because that is how C++ requires things to work, and because
you named your file with a .cpp extension and compiled with g++.  If you named
it .c and compiled with gcc it would be compiled as a C file with no name mangling.
If you want to export a name in a C++ file that is callable from C then you
need to use 'extern "C"'.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286529

Read and respond to this message at: 
https://sourceforge.net/forum/message.php?msg_id=3753186
By: rookie_one

I have a function like this:

__declspec(dllexport) game_export_t* GetGameAPI( game_import_t *import );

MSVC exports it without any problem:

Exports from gamex86.dll
 1 exported name(s), 1 export addresse(s).  Ordinal base is 1.
 Sorted by Name:
   RVA      Ord. Hint Name
   -------- ---- ---- ----
   00003904    1 0000 GetGameAPI

But MinGW with the following makefile:

all: quakecode gamex86 clientcmds admin
g++ -shared -Wl,--out-implib,gamex86.a -o gamex86.dll quakecode.o gamex86.o
clientcmds.o admin.o

quakecode: quakecode.cpp quakecode.h
g++ -c -Wall -O2 quakecode.cpp

gamex86: gamex86.cpp gamex86.h quakecode
g++ -c -Wall -O2 gamex86.cpp

clientcmds: clientcmds.cpp gamex86 quakecode
g++ -c -Wall -O2 clientcmds.cpp

admin: admin.cpp gamex86 quakecode
g++ -c -Wall -O2 admin.cpp

clean:
del quakecode.o
del gamex86.o
del clientcmds.o
del admin.o
del gamex86.a

Does this

Exports from gamex86.dll
 1 exported name(s), 1 export addresse(s).  Ordinal base is 1.
 Sorted by Name:
   RVA      Ord. Hint Name
   -------- ---- ---- ----
   00002550    1 0000 _Z10GetGameAPIP13game_import_t

WTF? I've been trying to get it right for like 5 hours now, I badly need this
to be compilable under mingw.

______________________________________________________________________
You are receiving this email because you elected to monitor this forum.
To stop monitoring this forum, login to SourceForge.net and visit: 
https://sourceforge.net/forum/unmonitor.php?forum_id=286529