qzdoom/src/win32/i_main.cpp

1099 lines
30 KiB
C++
Raw Normal View History

// Emacs style mode select -*- C++ -*-
//-----------------------------------------------------------------------------
//
// $Id:$
//
// Copyright (C) 1993-1996 by id Software, Inc.
//
// This source is available for distribution and/or modification
// only under the terms of the DOOM Source Code License as
// published by id Software. All rights reserved.
//
// The source is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// FITNESS FOR A PARTICULAR PURPOSE. See the DOOM Source Code License
// for more details.
//
// $Log:$
//
// DESCRIPTION:
// Main program, simply calls D_DoomMain high level loop.
//
//-----------------------------------------------------------------------------
#define WIN32_LEAN_AND_MEAN
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#define _WIN32_WINNT 0x0501
#include <windows.h>
#include <mmsystem.h>
#include <objbase.h>
#include <commctrl.h>
//#include <wtsapi32.h>
#define NOTIFY_FOR_THIS_SESSION 0
#include <malloc.h>
#include "m_alloc.h"
#ifdef _MSC_VER
#include <eh.h>
#include <new.h>
#include <crtdbg.h>
#endif
#include "resource.h"
#include <stdio.h>
#include <stdarg.h>
#include <math.h>
#define USE_WINDOWS_DWORD
#include "doomerrors.h"
#include "hardware.h"
#include "doomtype.h"
#include "m_argv.h"
#include "d_main.h"
#include "i_system.h"
#include "c_console.h"
#include "version.h"
#include "i_video.h"
#include "i_sound.h"
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#include "i_input.h"
#include "autosegs.h"
#include "w_wad.h"
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#include "templates.h"
#include "stats.h"
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#include "st_start.h"
#include <assert.h>
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#define WINDOW_TITLE GAMESIG " " DOTVERSIONSTR " (" __DATE__ ")"
LRESULT CALLBACK WndProc (HWND, UINT, WPARAM, LPARAM);
void CreateCrashLog (char *custominfo, DWORD customsize);
extern void DisplayCrashLog ();
extern EXCEPTION_POINTERS CrashPointers;
// Will this work with something besides VC++?
// Answer: yes it will.
// Which brings up the question, what won't it work with?
//
//extern int __argc;
//extern char **__argv;
DArgs Args;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
const char WinClassName[] = "ZDoomMainWindow";
HINSTANCE g_hInst;
DWORD SessionID;
HANDLE MainThread;
DWORD MainThreadID;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
// The main window
HWND Window;
// The subwindows used for startup and error output
HWND ConWindow, GameTitleWindow;
HWND ErrorPane, ProgressBar, NetStartPane;
HFONT GameTitleFont;
LONG GameTitleFontHeight;
HMODULE hwtsapi32; // handle to wtsapi32.dll
extern UINT TimerPeriod;
extern HCURSOR TheArrowCursor, TheInvisibleCursor;
#define MAX_TERMS 32
void (*TermFuncs[MAX_TERMS])(void);
static int NumTerms;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// atterm
//
// Our own atexit because atexit can be problematic under Linux, though I
// forget the circumstances that cause trouble.
//
//===========================================================================
void atterm (void (*func)(void))
{
// Make sure this function wasn't already registered.
for (int i = 0; i < NumTerms; ++i)
{
if (TermFuncs[i] == func)
{
return;
}
}
if (NumTerms == MAX_TERMS)
{
func ();
I_FatalError ("Too many exit functions registered.\nIncrease MAX_TERMS in i_main.cpp");
}
TermFuncs[NumTerms++] = func;
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// popterm
//
// Removes the most recently register atterm function.
//
//===========================================================================
void popterm ()
{
if (NumTerms)
NumTerms--;
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// call_terms
//
//===========================================================================
static void STACK_ARGS call_terms (void)
{
while (NumTerms > 0)
{
TermFuncs[--NumTerms]();
}
}
#ifdef _MSC_VER
static int STACK_ARGS NewFailure (size_t size)
{
I_FatalError ("Failed to allocate %d bytes from process heap", size);
return 0;
}
#endif
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// UnCOM
//
// Called by atterm if CoInitialize() succeeded.
//
//===========================================================================
static void UnCOM (void)
{
CoUninitialize ();
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// UnWTS
//
// Called by atterm if RegisterSessionNotification() succeeded.
//
//===========================================================================
static void UnWTS (void)
{
if (hwtsapi32 != 0)
{
typedef BOOL (WINAPI *ursn)(HWND);
ursn unreg = (ursn)GetProcAddress (hwtsapi32, "WTSUnRegisterSessionNotification");
if (unreg != 0)
{
unreg (Window);
}
FreeLibrary (hwtsapi32);
hwtsapi32 = 0;
}
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// LayoutErrorPane
//
// Lays out the error pane to the desired width, returning the required
// height.
//
//===========================================================================
static int LayoutErrorPane (HWND pane, int w)
{
HWND ctl;
RECT margin, rectc;
int cxicon, cyicon;
const char *text;
int textheight, textwidth, padding;
// Determine margin sizes.
SetRect (&margin, 7, 7, 0, 0);
MapDialogRect (pane, &margin);
// Get size of icon.
cxicon = GetSystemMetrics (SM_CXICON);
cyicon = GetSystemMetrics (SM_CYICON);
// Determine size of text control.
text = (const char *)(LONG_PTR)GetWindowLongPtr (pane, DWLP_USER);
textwidth = w - margin.left*3 - cxicon;
textheight = 0;
if (text != NULL)
{
HWND ctl;
HDC dc;
HGDIOBJ oldfont;
int len;
ctl = GetDlgItem (pane, IDC_ERRORTEXT);
rectc.top = rectc.left = 0;
rectc.right = textwidth;
rectc.bottom = 0;
len = (int)strlen(text);
dc = GetDC (pane);
oldfont = SelectObject (dc, (HFONT)SendMessage (ctl, WM_GETFONT, 0, 0));
DrawText (dc, text, len, &rectc, DT_CALCRECT | DT_EDITCONTROL | DT_NOPREFIX | DT_EXPANDTABS | DT_WORDBREAK | DT_NOCLIP);
SelectObject (dc, oldfont);
ReleaseDC (pane, dc);
textheight = rectc.bottom;
}
// Fill the text box to the determined size. If it is shorter than the icon,
// center it vertically.
padding = MAX (cyicon - textheight, 0);
ctl = GetDlgItem (pane, IDC_ERRORTEXT);
MoveWindow (ctl, margin.left*2 + cxicon, margin.top + padding/2, textwidth, textheight, TRUE);
InvalidateRect (ctl, NULL, TRUE);
textheight += padding;
// Center the Okay button horizontally, just underneath the text box.
ctl = GetDlgItem (pane, IDOK);
GetClientRect (ctl, &rectc); // Find out how big it is.
MoveWindow (ctl, (w - rectc.right) / 2, margin.top*2 + textheight, rectc.right, rectc.bottom, TRUE);
InvalidateRect (ctl, NULL, TRUE);
// Return the needed height for this layout
return margin.top*3 + textheight + rectc.bottom;
}
//===========================================================================
//
// LayoutNetStartPane
//
// Lays out the network startup pane to the specified width, returning
// its required height.
//
//===========================================================================
int LayoutNetStartPane (HWND pane, int w)
{
HWND ctl;
RECT margin, rectc;
int staticheight, barheight;
// Determine margin sizes.
SetRect (&margin, 7, 7, 0, 0);
MapDialogRect (pane, &margin);
// Stick the message text in the upper left corner.
ctl = GetDlgItem (pane, IDC_NETSTARTMESSAGE);
GetClientRect (ctl, &rectc);
MoveWindow (ctl, margin.left, margin.top, rectc.right, rectc.bottom, TRUE);
// Stick the count text in the upper right corner.
ctl = GetDlgItem (pane, IDC_NETSTARTCOUNT);
GetClientRect (ctl, &rectc);
MoveWindow (ctl, w - rectc.right - margin.left, margin.top, rectc.right, rectc.bottom, TRUE);
staticheight = rectc.bottom;
// Stretch the progress bar to fill the entire width.
ctl = GetDlgItem (pane, IDC_NETSTARTPROGRESS);
barheight = GetSystemMetrics (SM_CYVSCROLL);
MoveWindow (ctl, margin.left, margin.top*2 + staticheight, w - margin.left*2, barheight, TRUE);
// Center the abort button underneath the progress bar.
ctl = GetDlgItem (pane, IDCANCEL);
GetClientRect (ctl, &rectc);
MoveWindow (ctl, (w - rectc.right) / 2, margin.top*3 + staticheight + barheight, rectc.right, rectc.bottom, TRUE);
return margin.top*4 + staticheight + barheight + rectc.bottom;
}
//===========================================================================
//
// LayoutMainWindow
//
// Lays out the main window with the game title and log controls and
// possibly the error pane and progress bar.
//
//===========================================================================
void LayoutMainWindow (HWND hWnd, HWND pane)
{
RECT rect;
int errorpaneheight = 0;
int bannerheight = 0;
int progressheight = 0;
int netpaneheight = 0;
int w, h;
GetClientRect (hWnd, &rect);
w = rect.right;
h = rect.bottom;
if (pane != NULL)
{
errorpaneheight = LayoutErrorPane (pane, w);
SetWindowPos (pane, HWND_TOP, 0, 0, w, errorpaneheight, 0);
}
if (DoomStartupInfo != NULL)
{
bannerheight = GameTitleFontHeight + 5;
MoveWindow (GameTitleWindow, 0, errorpaneheight, w, bannerheight, TRUE);
InvalidateRect (GameTitleWindow, NULL, FALSE);
}
if (ProgressBar != NULL)
{
// Base the height of the progress bar on the height of a scroll bar
// arrow, just as in the progress bar example.
progressheight = GetSystemMetrics (SM_CYVSCROLL);
MoveWindow (ProgressBar, 0, h - progressheight, w, progressheight, TRUE);
}
if (NetStartPane != NULL)
{
netpaneheight = LayoutNetStartPane (NetStartPane, w);
SetWindowPos (NetStartPane, HWND_TOP, 0, h - progressheight - netpaneheight, w, netpaneheight, SWP_SHOWWINDOW);
RECT foo;
GetClientRect (NetStartPane, &foo);
foo.bottom = foo.bottom;;
}
// The log window uses whatever space is left.
MoveWindow (ConWindow, 0, errorpaneheight + bannerheight, w,
h - bannerheight - errorpaneheight - progressheight - netpaneheight, TRUE);
}
//===========================================================================
//
// LConProc
//
// The main window's WndProc during startup. During gameplay, the WndProc
// in i_input.cpp is used instead.
//
//===========================================================================
LRESULT CALLBACK LConProc (HWND hWnd, UINT msg, WPARAM wParam, LPARAM lParam)
{
HWND view;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
HDC hdc;
HBRUSH hbr;
HGDIOBJ oldfont;
RECT rect;
int titlelen;
SIZE size;
LOGFONT lf;
TEXTMETRIC tm;
HINSTANCE inst = (HINSTANCE)(LONG_PTR)GetWindowLongPtr(hWnd, GWLP_HINSTANCE);
HFONT font;
DRAWITEMSTRUCT *drawitem;
switch (msg)
{
case WM_CREATE:
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
// Create game title static control
memset (&lf, 0, sizeof(lf));
hdc = GetDC (hWnd);
lf.lfHeight = -MulDiv(12, GetDeviceCaps(hdc, LOGPIXELSY), 72);
lf.lfCharSet = ANSI_CHARSET;
lf.lfWeight = FW_BOLD;
lf.lfPitchAndFamily = VARIABLE_PITCH | FF_ROMAN;
strcpy (lf.lfFaceName, "Trebuchet MS");
GameTitleFont = CreateFontIndirect (&lf);
font = GameTitleFont != NULL ? GameTitleFont : (HFONT)GetStockObject (DEFAULT_GUI_FONT);
oldfont = SelectObject (hdc, font);
GetTextMetrics (hdc, &tm);
SelectObject (hdc, oldfont);
ReleaseDC (hWnd, hdc);
GameTitleFontHeight = tm.tmHeight;
// Create log read-only edit control
view = CreateWindowEx (WS_EX_NOPARENTNOTIFY | WS_EX_CLIENTEDGE, "EDIT", NULL,
WS_CHILD | WS_VISIBLE | WS_VSCROLL |
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
ES_LEFT | ES_MULTILINE | WS_CLIPSIBLINGS,
0, 0, 0, 0,
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
hWnd, NULL, inst, NULL);
if (view == NULL)
{
return -1;
}
SendMessage (view, WM_SETFONT, (WPARAM)GetStockObject (DEFAULT_GUI_FONT), FALSE);
SendMessage (view, EM_SETREADONLY, TRUE, 0);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
SendMessage (view, EM_SETLIMITTEXT, 0, 0);
ConWindow = view;
view = CreateWindowEx (WS_EX_NOPARENTNOTIFY, "STATIC", NULL, WS_CHILD | WS_VISIBLE | WS_CLIPSIBLINGS | SS_OWNERDRAW, 0, 0, 0, 0, hWnd, NULL, inst, NULL);
if (view == NULL)
{
return -1;
}
GameTitleWindow = view;
return 0;
case WM_SIZE:
if (wParam != SIZE_MAXHIDE && wParam != SIZE_MAXSHOW)
{
LayoutMainWindow (hWnd, ErrorPane);
}
return 0;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
case WM_DRAWITEM:
if (DoomStartupInfo != NULL)
{
const PalEntry *c;
// Draw the game title strip at the top of the window.
drawitem = (LPDRAWITEMSTRUCT)lParam;
// Draw the background.
rect = drawitem->rcItem;
rect.bottom -= 1;
c = (const PalEntry *)&DoomStartupInfo->BkColor;
hbr = CreateSolidBrush (RGB(c->r,c->g,c->b));
FillRect (drawitem->hDC, &drawitem->rcItem, hbr);
DeleteObject (hbr);
// Calculate width of the title string.
SetTextAlign (drawitem->hDC, TA_TOP);
oldfont = SelectObject (drawitem->hDC, GameTitleFont != NULL ? GameTitleFont : (HFONT)GetStockObject (DEFAULT_GUI_FONT));
titlelen = (int)strlen (DoomStartupInfo->Name);
GetTextExtentPoint32 (drawitem->hDC, DoomStartupInfo->Name, titlelen, &size);
// Draw the title.
c = (const PalEntry *)&DoomStartupInfo->FgColor;
SetTextColor (drawitem->hDC, RGB(c->r,c->g,c->b));
SetBkMode (drawitem->hDC, TRANSPARENT);
TextOut (drawitem->hDC, rect.left + (rect.right - rect.left - size.cx) / 2, 2, DoomStartupInfo->Name, titlelen);
SelectObject (drawitem->hDC, oldfont);
return TRUE;
}
return FALSE;
case WM_CLOSE:
PostQuitMessage (0);
break;
case WM_CTLCOLORSTATIC:
return (LRESULT)GetStockObject (WHITE_BRUSH);
case WM_DESTROY:
if (GameTitleFont != NULL)
{
DeleteObject (GameTitleFont);
}
break;
}
return DefWindowProc (hWnd, msg, wParam, lParam);
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// ErrorPaneProc
//
// DialogProc for the error pane.
//
//===========================================================================
INT_PTR CALLBACK ErrorPaneProc (HWND hDlg, UINT msg, WPARAM wParam, LPARAM lParam)
{
HWND ctl;
switch (msg)
{
case WM_INITDIALOG:
SetWindowLongPtr (hDlg, DWLP_USER, lParam);
// Set the static control to the error text.
ctl = GetDlgItem (hDlg, IDC_ERRORTEXT);
SetWindowText (ctl, (LPCSTR)lParam);
// Set the icon to the system default error icon.
ctl = GetDlgItem (hDlg, IDC_ICONPIC);
SendMessage (ctl, STM_SETICON, (WPARAM)LoadIcon (NULL, IDI_ERROR), 0);
// Appear in the main window.
LayoutMainWindow (GetParent (hDlg), hDlg);
// Make sure the last line of output is visible in the log window.
SendMessage (ConWindow, EM_LINESCROLL, 0, SendMessage (ConWindow, EM_GETLINECOUNT, 0, 0) -
SendMessage (ConWindow, EM_GETFIRSTVISIBLELINE, 0, 0));
return TRUE;
case WM_COMMAND:
// There is only one button, and it's "Ok" and makes us quit.
if (HIWORD(wParam) == BN_CLICKED)
{
PostQuitMessage (0);
return TRUE;
}
break;
}
return FALSE;
}
//===========================================================================
//
// I_SetWndProc
//
// Sets the main WndProc, hides all the child windows, and starts up
// in-game input.
//
//===========================================================================
void I_SetWndProc()
{
if (GetWindowLongPtr (Window, GWLP_USERDATA) == 0)
{
SetWindowLongPtr (Window, GWLP_USERDATA, 1);
SetWindowLongPtr (Window, GWLP_WNDPROC, (WLONG_PTR)WndProc);
ShowWindow (ConWindow, SW_HIDE);
ShowWindow (GameTitleWindow, SW_HIDE);
I_InitInput (Window);
}
}
//===========================================================================
//
// RestoreConView
//
// Returns the main window to its startup state.
//
//===========================================================================
void RestoreConView()
{
// Make sure the window has a frame in case it was fullscreened.
SetWindowLongPtr (Window, GWL_STYLE, WS_VISIBLE|WS_OVERLAPPEDWINDOW);
if (GetWindowLong (Window, GWL_EXSTYLE) & WS_EX_TOPMOST)
{
SetWindowPos (Window, HWND_BOTTOM, 0, 0, 512, 384,
SWP_DRAWFRAME | SWP_NOCOPYBITS | SWP_NOMOVE);
SetWindowPos (Window, HWND_TOP, 0, 0, 0, 0, SWP_NOCOPYBITS | SWP_NOMOVE | SWP_NOSIZE);
}
else
{
SetWindowPos (Window, NULL, 0, 0, 512, 384,
SWP_DRAWFRAME | SWP_NOCOPYBITS | SWP_NOMOVE | SWP_NOZORDER);
}
SetWindowLongPtr (Window, GWLP_WNDPROC, (WLONG_PTR)LConProc);
ShowWindow (ConWindow, SW_SHOW);
ShowWindow (GameTitleWindow, SW_SHOW);
// Make sure the progress bar isn't visible.
ST_Done();
}
//===========================================================================
//
// ShowErrorPane
//
// Shows an error message, preferably in the main window, but it can
// use a normal message box too.
//
//===========================================================================
void ShowErrorPane(const char *text)
{
// Ensure that the network pane is hidden.
ST_NetDone();
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
ErrorPane = CreateDialogParam (g_hInst, MAKEINTRESOURCE(IDD_ERRORPANE), Window, ErrorPaneProc, (LPARAM)text);
if (ErrorPane == NULL)
{
MessageBox (Window, text,
"ZDOOM Fatal Error", MB_OK|MB_ICONSTOP|MB_TASKMODAL);
}
else
{
BOOL bRet;
MSG msg;
SetWindowText (Window, "Fatal Error - " WINDOW_TITLE);
while ((bRet = GetMessage(&msg, NULL, 0, 0)) != 0)
{
if (bRet == -1)
{
MessageBox (Window, text,
"ZDOOM Fatal Error", MB_OK|MB_ICONSTOP|MB_TASKMODAL);
return;
}
else if (!IsDialogMessage (ErrorPane, &msg))
{
TranslateMessage (&msg);
DispatchMessage (&msg);
}
}
}
}
//===========================================================================
//
// DoMain
//
//===========================================================================
void DoMain (HINSTANCE hInstance)
{
LONG WinWidth, WinHeight;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
int height, width, x, y;
RECT cRect;
TIMECAPS tc;
DEVMODE displaysettings;
try
{
#ifdef _MSC_VER
_set_new_handler (NewFailure);
#endif
Args.SetArgs (__argc, __argv);
// Under XP, get our session ID so we can know when the user changes/locks sessions.
// Since we need to remain binary compatible with older versions of Windows, we
// need to extract the ProcessIdToSessionId function from kernel32.dll manually.
HMODULE kernel = GetModuleHandle ("kernel32.dll");
// NASM does not support creating writeable code sections (even though this
// is a perfectly valid configuration for Microsoft's COFF format), so I
// need to make the self-modifying code writeable after it's already loaded.
#ifdef USEASM
{
BYTE *module = (BYTE *)GetModuleHandle (NULL);
IMAGE_DOS_HEADER *dosHeader = (IMAGE_DOS_HEADER *)module;
IMAGE_NT_HEADERS *ntHeaders = (IMAGE_NT_HEADERS *)(module + dosHeader->e_lfanew);
IMAGE_SECTION_HEADER *sections = IMAGE_FIRST_SECTION (ntHeaders);
int i;
LPVOID *start = NULL;
SIZE_T size = 0;
DWORD oldprotect;
for (i = 0; i < ntHeaders->FileHeader.NumberOfSections; ++i)
{
if (memcmp (sections[i].Name, ".rtext\0", 8) == 0)
{
start = (LPVOID *)(sections[i].VirtualAddress + module);
size = sections[i].Misc.VirtualSize;
break;
}
}
// I think these pages need to be mapped PAGE_EXECUTE_WRITECOPY (based on the
// description of PAGE_WRITECOPY), but PAGE_EXECUTE_READWRITE seems to work
// just as well; two instances of the program can be running with different
// resolutions at the same time either way. Perhaps the file mappings for
// executables are created with PAGE_WRITECOPY, so any attempts to give them
// write access are automatically transformed to copy-on-write?
//
// This used to be PAGE_EXECUTE_WRITECOPY until Timmie found out Win9x doesn't
// support it, although the MSDN does not indicate it.
if (!VirtualProtect (start, size, PAGE_EXECUTE_READWRITE, &oldprotect))
{
I_FatalError ("The self-modifying code section code not be made writeable.");
}
}
#endif
// Set the timer to be as accurate as possible
if (timeGetDevCaps (&tc, sizeof(tc)) != TIMERR_NOERROR)
TimerPeriod = 1; // Assume minimum resolution of 1 ms
else
TimerPeriod = tc.wPeriodMin;
timeBeginPeriod (TimerPeriod);
/*
killough 1/98:
This fixes some problems with exit handling
during abnormal situations.
The old code called I_Quit() to end program,
while now I_Quit() is installed as an exit
handler and exit() is called to exit, either
normally or abnormally.
*/
atexit (call_terms);
atterm (I_Quit);
// Figure out what directory the program resides in.
GetModuleFileName (NULL, progdir, 1024);
*(strrchr (progdir, '\\') + 1) = 0;
FixPathSeperator (progdir);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
/*
height = GetSystemMetrics (SM_CYFIXEDFRAME) * 2 +
GetSystemMetrics (SM_CYCAPTION) + 12 * 32;
width = GetSystemMetrics (SM_CXFIXEDFRAME) * 2 + 8 * 78;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
*/
width = 512;
height = 384;
// Many Windows structures that specify their size do so with the first
// element. DEVMODE is not one of those structures.
memset (&displaysettings, 0, sizeof(displaysettings));
displaysettings.dmSize = sizeof(displaysettings);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
EnumDisplaySettings (NULL, ENUM_CURRENT_SETTINGS, &displaysettings);
x = (displaysettings.dmPelsWidth - width) / 2;
y = (displaysettings.dmPelsHeight - height) / 2;
#if _DEBUG
x = y = 0;
#endif
TheInvisibleCursor = LoadCursor (hInstance, MAKEINTRESOURCE(IDC_INVISIBLECURSOR));
TheArrowCursor = LoadCursor (NULL, IDC_ARROW);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
WNDCLASS WndClass;
WndClass.style = 0;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
WndClass.lpfnWndProc = LConProc;
WndClass.cbClsExtra = 0;
WndClass.cbWndExtra = 0;
WndClass.hInstance = hInstance;
WndClass.hIcon = LoadIcon (hInstance, MAKEINTRESOURCE(IDI_ICON1));
WndClass.hCursor = TheArrowCursor;
WndClass.hbrBackground = NULL;
WndClass.lpszMenuName = NULL;
WndClass.lpszClassName = (LPCTSTR)WinClassName;
/* register this new class with Windows */
if (!RegisterClass((LPWNDCLASS)&WndClass))
I_FatalError ("Could not register window class");
/* create window */
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
Window = CreateWindowEx(
WS_EX_APPWINDOW,
(LPCTSTR)WinClassName,
(LPCTSTR)WINDOW_TITLE,
WS_OVERLAPPEDWINDOW | WS_VISIBLE | WS_CLIPCHILDREN,
x, y, width, height,
(HWND) NULL,
(HMENU) NULL,
hInstance,
NULL);
if (!Window)
I_FatalError ("Could not open window");
if (kernel != 0)
{
typedef BOOL (WINAPI *pts)(DWORD, DWORD *);
pts pidsid = (pts)GetProcAddress (kernel, "ProcessIdToSessionId");
if (pidsid != 0)
{
if (!pidsid (GetCurrentProcessId(), &SessionID))
{
SessionID = 0;
}
hwtsapi32 = LoadLibraryA ("wtsapi32.dll");
if (hwtsapi32 != 0)
{
FARPROC reg = GetProcAddress (hwtsapi32, "WTSRegisterSessionNotification");
if (reg == 0 || !((BOOL(WINAPI *)(HWND, DWORD))reg) (Window, NOTIFY_FOR_THIS_SESSION))
{
FreeLibrary (hwtsapi32);
hwtsapi32 = 0;
}
else
{
atterm (UnWTS);
}
}
}
}
GetClientRect (Window, &cRect);
WinWidth = cRect.right;
WinHeight = cRect.bottom;
CoInitialize (NULL);
atterm (UnCOM);
C_InitConsole (((WinWidth / 8) + 2) * 8, (WinHeight / 12) * 8, false);
I_DetectOS ();
D_DoomMain ();
}
catch (class CDoomError &error)
{
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
I_ShutdownGraphics ();
RestoreConView ();
if (error.GetMessage ())
{
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
ShowErrorPane (error.GetMessage());
}
exit (-1);
}
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// DoomSpecificInfo
//
// Called by the crash logger to get application-specific information.
//
//===========================================================================
void DoomSpecificInfo (char *buffer)
{
const char *arg;
int i;
buffer += wsprintf (buffer, "ZDoom version " DOTVERSIONSTR " (" __DATE__ ")\r\n");
buffer += wsprintf (buffer, "\r\nCommand line: %s\r\n", GetCommandLine());
for (i = 0; (arg = Wads.GetWadName (i)) != NULL; ++i)
{
buffer += wsprintf (buffer, "\r\nWad %d: %s", i, arg);
}
if (gamestate != GS_LEVEL && gamestate != GS_TITLELEVEL)
{
buffer += wsprintf (buffer, "\r\n\r\nNot in a level.");
}
else
{
char name[9];
strncpy (name, level.mapname, 8);
name[8] = 0;
buffer += wsprintf (buffer, "\r\n\r\nCurrent map: %s", name);
if (!viewactive)
{
buffer += wsprintf (buffer, "\r\n\r\nView not active.");
}
else
{
buffer += wsprintf (buffer, "\r\n\r\nviewx = %d", viewx);
buffer += wsprintf (buffer, "\r\nviewy = %d", viewy);
buffer += wsprintf (buffer, "\r\nviewz = %d", viewz);
buffer += wsprintf (buffer, "\r\nviewangle = %x", viewangle);
}
}
*buffer++ = '\r';
*buffer++ = '\n';
*buffer++ = '\0';
}
extern FILE *Logfile;
// Here is how the error logging system works.
//
// To catch exceptions that occur in secondary threads, CatchAllExceptions is
// set as the UnhandledExceptionFilter for this process. It records the state
// of the thread at the time of the crash using CreateCrashLog and then queues
// an APC on the primary thread. When the APC executes, it raises a software
// exception that gets caught by the __try/__except block in WinMain.
// I_GetEvent calls SleepEx to put the primary thread in a waitable state
// periodically so that the APC has a chance to execute.
//
// Exceptions on the primary thread are caught by the __try/__except block in
// WinMain. Not only does it record the crash information, it also shuts
// everything down and displays a dialog with the information present. If a
// console log is being produced, the information will also be appended to it.
//
// If a debugger is running, CatchAllExceptions never executes, so secondary
// thread exceptions will always be caught by the debugger. For the primary
// thread, IsDebuggerPresent is called to determine if a debugger is present.
// Note that this function is not present on Windows 95, so we cannot
// statically link to it.
//
// To make this work with MinGW, you will need to use inline assembly
// because GCC offers no native support for Windows' SEH.
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// SleepForever
//
//===========================================================================
void SleepForever ()
{
Sleep (INFINITE);
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// ExitMessedUp
//
// An exception occurred while exiting, so don't do any standard processing.
// Just die.
//
//===========================================================================
LONG WINAPI ExitMessedUp (LPEXCEPTION_POINTERS foo)
{
ExitProcess (1000);
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// ExitFatally
//
//===========================================================================
void CALLBACK ExitFatally (ULONG_PTR dummy)
{
SetUnhandledExceptionFilter (ExitMessedUp);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
I_ShutdownGraphics ();
RestoreConView ();
DisplayCrashLog ();
exit (-1);
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// CatchAllExceptions
//
//===========================================================================
LONG WINAPI CatchAllExceptions (LPEXCEPTION_POINTERS info)
{
#ifdef _DEBUG
if (info->ExceptionRecord->ExceptionCode == EXCEPTION_BREAKPOINT)
{
return EXCEPTION_CONTINUE_SEARCH;
}
#endif
static bool caughtsomething = false;
if (caughtsomething) return EXCEPTION_EXECUTE_HANDLER;
caughtsomething = true;
char *custominfo = (char *)HeapAlloc (GetProcessHeap(), 0, 16384);
CrashPointers = *info;
DoomSpecificInfo (custominfo);
CreateCrashLog (custominfo, (DWORD)strlen(custominfo));
// If the main thread crashed, then make it clean up after itself.
// Otherwise, put the crashing thread to sleep and signal the main thread to clean up.
if (GetCurrentThreadId() == MainThreadID)
{
info->ContextRecord->Eip = (DWORD_PTR)ExitFatally;
}
else
{
info->ContextRecord->Eip = (DWORD_PTR)SleepForever;
QueueUserAPC (ExitFatally, MainThread, 0);
}
return EXCEPTION_CONTINUE_EXECUTION;
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// infiniterecursion
//
// Debugging routine for testing the crash logger.
//
//===========================================================================
#ifdef _DEBUG
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
static void infiniterecursion(int foo)
{
if (foo)
{
infiniterecursion(foo);
}
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#endif
//===========================================================================
//
// WinMain
//
//===========================================================================
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
int WINAPI WinMain (HINSTANCE hInstance, HINSTANCE nothing, LPSTR cmdline, int nCmdShow)
{
g_hInst = hInstance;
InitCommonControls (); // Be pretty under XP
#if !defined(__GNUC__) && defined(_DEBUG)
if (__argc == 2 && strcmp (__argv[1], "TestCrash") == 0)
{
__try
{
*(int *)0 = 0;
}
__except(CrashPointers = *GetExceptionInformation(),
CreateCrashLog (__argv[1], 9), EXCEPTION_EXECUTE_HANDLER)
{
}
DisplayCrashLog ();
exit (0);
}
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
if (__argc == 2 && strcmp (__argv[1], "TestStackCrash") == 0)
{
__try
{
infiniterecursion(1);
}
__except(CrashPointers = *GetExceptionInformation(),
CreateCrashLog (__argv[1], 14), EXCEPTION_EXECUTE_HANDLER)
{
}
DisplayCrashLog ();
exit (0);
}
#endif
MainThread = INVALID_HANDLE_VALUE;
DuplicateHandle (GetCurrentProcess(), GetCurrentThread(), GetCurrentProcess(), &MainThread,
0, FALSE, DUPLICATE_SAME_ACCESS);
MainThreadID = GetCurrentThreadId();
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
#ifndef _DEBUG
if (MainThread != INVALID_HANDLE_VALUE)
{
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
SetUnhandledExceptionFilter (CatchAllExceptions);
}
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
#endif
#if defined(_DEBUG) && defined(_MSC_VER)
// Uncomment this line to make the Visual C++ CRT check the heap before
// every allocation and deallocation. This will be slow, but it can be a
// great help in finding problem areas.
//_CrtSetDbgFlag (_CRTDBG_ALLOC_MEM_DF | _CRTDBG_CHECK_ALWAYS_DF);
_CrtSetDbgFlag (_CrtSetDbgFlag(0) | _CRTDBG_LEAK_CHECK_DF);
#endif
DoMain (hInstance);
CloseHandle (MainThread);
MainThread = INVALID_HANDLE_VALUE;
return 0;
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
//===========================================================================
//
// CCMD crashout
//
// Debugging routine for testing the crash logger.
// Useless in a debug build, because that doesn't enable the crash logger.
//
//===========================================================================
#ifndef _DEBUG
#include "c_dispatch.h"
CCMD (crashout)
{
*(int *)0 = 0;
}
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#endif