gzdoom/src/win32/st_start.cpp

1172 lines
34 KiB
C++
Raw Normal View History

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
/*
** st_start.cpp
** Handles the startup screen.
**
**---------------------------------------------------------------------------
** Copyright 2006-2007 Randy Heit
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
** All rights reserved.
**
** Redistribution and use in source and binary forms, with or without
** modification, are permitted provided that the following conditions
** are met:
**
** 1. Redistributions of source code must retain the above copyright
** notice, this list of conditions and the following disclaimer.
** 2. Redistributions in binary form must reproduce the above copyright
** notice, this list of conditions and the following disclaimer in the
** documentation and/or other materials provided with the distribution.
** 3. The name of the author may not be used to endorse or promote products
** derived from this software without specific prior written permission.
**
** THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
** IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
** OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
** IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
** INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
** NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
** DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
** THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
** (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
** THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
**---------------------------------------------------------------------------
**
*/
// HEADER FILES ------------------------------------------------------------
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_LEAN_AND_MEAN
#define _WIN32_WINNT 0x0501 // required to get the MARQUEE defines
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 <windows.h>
#include <commctrl.h>
#include <stdio.h>
#include "resource.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 USE_WINDOWS_DWORD
#include "st_start.h"
#include "resource.h"
#include "templates.h"
#include "i_system.h"
#include "i_input.h"
#include "hardware.h"
#include "gi.h"
#include "w_wad.h"
#include "s_sound.h"
#include "m_alloc.h"
// MACROS ------------------------------------------------------------------
#define ST_MAX_NOTCHES 32
#define ST_NOTCH_WIDTH 16
#define ST_NOTCH_HEIGHT 23
#define ST_PROGRESS_X 64 // Start of notches x screen pos.
#define ST_PROGRESS_Y 441 // Start of notches y screen pos.
#define ST_NETPROGRESS_X 288
#define ST_NETPROGRESS_Y 32
#define ST_NETNOTCH_WIDTH 4
#define ST_NETNOTCH_HEIGHT 16
#define ST_MAX_NETNOTCHES 8
#define TEXT_FONT_HEIGHT 14
#define TEXT_FONT_NAME "vga-rom-font.14"
#define HERETIC_MINOR_VERSION '3' // Since we're based on Heretic 1.3
#define THERM_X 14
#define THERM_Y 14
#define THERM_LEN 51
#define THERM_COLOR 0xAA // light green
// Text mode color values
#define LO 85
#define MD 170
#define HI 255
// TYPES -------------------------------------------------------------------
// EXTERNAL FUNCTION PROTOTYPES --------------------------------------------
void RestoreConView();
bool ST_Util_CreateStartupWindow ();
void ST_Util_SizeWindowForBitmap ();
BITMAPINFO *ST_Util_CreateBitmap (int width, int height, int color_bits);
BYTE *ST_Util_BitsForBitmap (BITMAPINFO *bitmap_info);
void ST_Util_FreeBitmap (BITMAPINFO *bitmap_info);
void ST_Util_PlanarToChunky4 (BYTE *dest, const BYTE *src, int width, int height);
void ST_Util_DrawBlock (BYTE *dest, const BYTE *src, int bytewidth, int height, int destpitch);
void ST_Util_ClearBlock (BYTE *dest, BYTE fill, int bytewidth, int height, int destpitch);
void ST_Util_InvalidateRect (HWND hwnd, BITMAPINFO *bitmap_info, int left, int top, int right, int bottom);
BYTE *ST_Util_LoadFont (const char *filename, int height);
void ST_Util_FreeFont (BYTE *font);
BITMAPINFO *ST_Util_AllocTextBitmap (const BYTE *font);
void ST_Util_DrawTextScreen (BITMAPINFO *bitmap_info, const BYTE *text_screen, const BYTE *font);
void ST_Util_DrawChar (BITMAPINFO *screen, const BYTE *font, int x, int y, BYTE charnum, BYTE attrib);
// PUBLIC FUNCTION PROTOTYPES ----------------------------------------------
// PRIVATE FUNCTION PROTOTYPES ---------------------------------------------
static INT_PTR CALLBACK NetStartPaneProc (HWND hDlg, UINT msg, WPARAM wParam, LPARAM lParam);
static void ST_Basic_Init ();
static void ST_Basic_Done ();
static void ST_Basic_Progress ();
static void ST_Basic_NetInit (const char *message, int numplayers);
static void ST_Basic_NetProgress (int count);
static void ST_Basic_NetMessage (const char *format, ...);
static void ST_Basic_NetDone ();
static bool ST_Basic_NetLoop (bool (*timer_callback)(void *), void *userdata);
static void ST_Hexen_Init ();
static void ST_Hexen_Done ();
static void ST_Hexen_Progress ();
static void ST_Heretic_Init ();
static void ST_Heretic_Progress ();
// EXTERNAL DATA DECLARATIONS ----------------------------------------------
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
extern HINSTANCE g_hInst;
extern HWND Window, ConWindow, ProgressBar, NetStartPane, StartupScreen, GameTitleWindow;
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
extern void LayoutMainWindow (HWND hWnd, HWND pane);
// PUBLIC DATA DEFINITIONS -------------------------------------------------
void (*ST_Done)();
void (*ST_Progress)();
void (*ST_NetInit)(const char *message, int numplayers);
void (*ST_NetProgress)(int count);
void (*ST_NetMessage)(const char *format, ...);
void (*ST_NetDone)();
bool (*ST_NetLoop)(bool (*timer_callback)(void *), void *userdata);
BITMAPINFO *StartupBitmap;
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
// PRIVATE DATA DEFINITIONS ------------------------------------------------
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
static int MaxPos, CurPos, NotchPos;
static int NetMaxPos, NetCurPos;
static LRESULT NetMarqueeMode;
static int ThermX, ThermY, ThermWidth, ThermHeight;
static const RGBQUAD TextModePalette[16] =
{
{ 0, 0, 0, 0 }, // 0 black
{ MD, 0, 0, 0 }, // 1 blue
{ 0, MD, 0, 0 }, // 2 green
{ MD, MD, 0, 0 }, // 3 cyan
{ 0, 0, MD, 0 }, // 4 red
{ MD, 0, MD, 0 }, // 5 magenta
{ 0, MD, MD, 0 }, // 6 brown
{ MD, MD, MD, 0 }, // 7 light gray
{ LO, LO, LO, 0 }, // 8 dark gray
{ HI, LO, LO, 0 }, // 9 light blue
{ LO, HI, LO, 0 }, // A light green
{ HI, HI, LO, 0 }, // B light cyan
{ LO, LO, HI, 0 }, // C light red
{ HI, LO, HI, 0 }, // D light magenta
{ LO, HI, HI, 0 }, // E yellow
{ HI, HI, HI, 0 }, // F white
};
// Hexen's notch graphics, converted to chunky pixels.
static const BYTE NotchBits[] =
{
0x00, 0x00, 0x00, 0x00, 0x70, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x07, 0x70, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x0c, 0xc0, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x08, 0x80, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x68, 0x86, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0x78, 0x87, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd8, 0x8d, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd8, 0x8d, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd8, 0x8d, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd8, 0x8d, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd8, 0x87, 0x00, 0x00, 0x00,
0x00, 0x00, 0x00, 0xd7, 0x7d, 0x60, 0x00, 0x00,
0x00, 0x66, 0x99, 0x99, 0x96, 0x69, 0x66, 0x00,
0x00, 0x69, 0x96, 0x99, 0x69, 0x96, 0x96, 0x00,
0x06, 0x9d, 0x99, 0x69, 0x96, 0xd9, 0x79, 0x60,
0x06, 0x7d, 0xdd, 0xdd, 0xdd, 0xdd, 0x77, 0x60,
0x06, 0x78, 0x88, 0x88, 0x88, 0x88, 0xd6, 0x60,
0x06, 0x7a, 0xaa, 0xaa, 0xaa, 0xaa, 0xd6, 0x60,
0x06, 0x7a, 0x77, 0x77, 0x77, 0xa7, 0x96, 0x60,
0x06, 0x77, 0xa7, 0x77, 0x77, 0xa7, 0x96, 0x60,
0x06, 0x97, 0xa7, 0x79, 0x77, 0x77, 0x96, 0x60,
0x00, 0x67, 0x79, 0x99, 0x99, 0xd7, 0x96, 0x60,
0x00, 0x69, 0x99, 0x66, 0x69, 0x69, 0x66, 0x00
};
static const BYTE NetNotchBits[] =
{
0x52, 0x20,
0x23, 0x25,
0x33, 0x25,
0x31, 0x35,
0x31, 0x35,
0x31, 0x35,
0x33, 0x35,
0x31, 0x35,
0x31, 0x35,
0x31, 0x25,
0x33, 0x35,
0x31, 0x20,
0x21, 0x35,
0x23, 0x25,
0x52, 0x20,
0x05, 0x50
};
// CODE --------------------------------------------------------------------
//==========================================================================
//
// ST_Init
//
// Initializes the startup screen for the detected game.
// Sets the size of the progress bar and displays the startup screen.
//
//==========================================================================
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
void ST_Init(int maxProgress)
{
MaxPos = maxProgress;
CurPos = 0;
NotchPos = 0;
if (gameinfo.gametype == GAME_Hexen)
{
ST_Hexen_Init ();
}
else if (gameinfo.gametype == GAME_Heretic)
{
ST_Heretic_Init ();
}
else
{
ST_Basic_Init ();
}
}
//==========================================================================
//
// ST_Basic_Init
//
// Shows a progress bar at the bottom of the window.
//
//==========================================================================
static void ST_Basic_Init ()
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
{
ProgressBar = CreateWindowEx(0, PROGRESS_CLASS,
NULL, WS_CHILD | WS_VISIBLE | WS_CLIPSIBLINGS,
0, 0, 0, 0,
Window, 0, g_hInst, NULL);
SendMessage (ProgressBar, PBM_SETRANGE, 0, MAKELPARAM(0,MaxPos));
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
LayoutMainWindow (Window, NULL);
ST_Done = ST_Basic_Done;
ST_Progress = ST_Basic_Progress;
ST_NetInit = ST_Basic_NetInit;
ST_NetProgress = ST_Basic_NetProgress;
ST_NetMessage = ST_Basic_NetMessage;
ST_NetDone = ST_Basic_NetDone;
ST_NetLoop = ST_Basic_NetLoop;
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
}
//==========================================================================
//
// ST_Basic_Done
//
// Called just before entering graphics mode to deconstruct the startup
// screen.
//
//==========================================================================
static void ST_Basic_Done()
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
{
if (ProgressBar != NULL)
{
DestroyWindow (ProgressBar);
ProgressBar = NULL;
LayoutMainWindow (Window, NULL);
}
}
//==========================================================================
//
// ST_Basic_Progress
//
// Bumps the progress meter one notch.
//
//==========================================================================
static void ST_Basic_Progress()
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
{
if (CurPos < MaxPos)
{
CurPos++;
SendMessage (ProgressBar, PBM_SETPOS, CurPos, 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
//
// ST_Basic_NetInit
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
//
// Shows the network startup pane if it isn't visible. Sets the message in
// the pane to the one provided. If numplayers is 0, then the progress bar
// is a scrolling marquee style. If numplayers is 1, then the progress bar
// is just a full bar. If numplayers is >= 2, then the progress bar is a
// normal one, and a progress count is also shown in the pane.
//
//==========================================================================
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
static void ST_Basic_NetInit(const char *message, int numplayers)
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
{
NetMaxPos = numplayers;
if (NetStartPane == NULL)
{
NetStartPane = CreateDialogParam (g_hInst, MAKEINTRESOURCE(IDD_NETSTARTPANE), Window, NetStartPaneProc, 0);
// We don't need two progress bars.
if (ProgressBar != NULL)
{
DestroyWindow (ProgressBar);
ProgressBar = NULL;
}
LayoutMainWindow (Window, NULL);
// 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));
}
if (NetStartPane != NULL)
{
HWND ctl;
SetDlgItemText (NetStartPane, IDC_NETSTARTMESSAGE, message);
ctl = GetDlgItem (NetStartPane, IDC_NETSTARTPROGRESS);
if (numplayers == 0)
{
// PBM_SETMARQUEE is only available under XP and above, so this might fail.
NetMarqueeMode = SendMessage (ctl, PBM_SETMARQUEE, TRUE, 100);
if (NetMarqueeMode == FALSE)
{
SendMessage (ctl, PBM_SETRANGE, 0, MAKELPARAM(0,16));
}
else
{
// If we don't set the PBS_MARQUEE style, then the marquee will never show up.
SetWindowLong (ctl, GWL_STYLE, GetWindowLong (ctl, GWL_STYLE) | PBS_MARQUEE);
}
SetDlgItemText (NetStartPane, IDC_NETSTARTCOUNT, "");
}
else
{
NetMarqueeMode = FALSE;
SendMessage (ctl, PBM_SETMARQUEE, FALSE, 0);
// Make sure the marquee really is turned off.
SetWindowLong (ctl, GWL_STYLE, GetWindowLong (ctl, GWL_STYLE) & (~PBS_MARQUEE));
SendMessage (ctl, PBM_SETRANGE, 0, MAKELPARAM(0,numplayers));
if (numplayers == 1)
{
SendMessage (ctl, PBM_SETPOS, 1, 0);
SetDlgItemText (NetStartPane, IDC_NETSTARTCOUNT, "");
}
}
}
NetMaxPos = numplayers;
NetCurPos = 0;
ST_NetProgress(1); // You always know about yourself
}
//==========================================================================
//
// ST_Basic_NetDone
//
// Removes the network startup pane.
//
//==========================================================================
static void ST_Basic_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
{
if (NetStartPane != NULL)
{
DestroyWindow (NetStartPane);
NetStartPane = NULL;
LayoutMainWindow (Window, NULL);
}
}
//==========================================================================
//
// ST_Basic_NetMessage
//
// Call this between ST_NetInit() and ST_NetDone() instead of Printf() to
// display messages, in case the progress meter is mixed in the same output
// stream as normal messages.
//
//==========================================================================
static void ST_Basic_NetMessage(const char *format, ...)
{
FString str;
va_list argptr;
va_start (argptr, format);
str.VFormat (format, argptr);
va_end (argptr);
Printf ("%s\n", str.GetChars());
}
//==========================================================================
//
// ST_Basic_NetProgress
//
// Sets the network progress meter. If count is 0, it gets bumped by 1.
// Otherwise, it is set to count.
//
//==========================================================================
static void ST_Basic_NetProgress(int count)
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
{
if (count == 0)
{
NetCurPos++;
}
else
{
NetCurPos = count;
}
if (NetStartPane == NULL)
{
return;
}
if (NetMaxPos == 0 && !NetMarqueeMode)
{
// PBM_SETMARQUEE didn't work, so just increment the progress bar endlessly.
SendDlgItemMessage (NetStartPane, IDC_NETSTARTPROGRESS, PBM_SETPOS, NetCurPos & 15, 0);
}
else if (NetMaxPos > 1)
{
char buf[16];
sprintf (buf, "%d/%d", NetCurPos, NetMaxPos);
SetDlgItemText (NetStartPane, IDC_NETSTARTCOUNT, buf);
SendDlgItemMessage (NetStartPane, IDC_NETSTARTPROGRESS, PBM_SETPOS, MIN(NetCurPos, NetMaxPos), 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
//
// ST_Basic_NetLoop
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 timer_callback function is called at least two times per second
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
// and passed the userdata value. It should return true to stop the loop and
// return control to the caller or false to continue the loop.
//
// ST_NetLoop will return true if the loop was halted by the callback and
// false if the loop was halted because the user wants to abort the
// network synchronization.
//
//==========================================================================
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
static bool ST_Basic_NetLoop(bool (*timer_callback)(void *), void *userdata)
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
{
BOOL bRet;
MSG msg;
if (SetTimer (Window, 1337, 500, NULL) == 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
{
I_FatalError ("Could not set network synchronization timer.");
}
while ((bRet = GetMessage(&msg, NULL, 0, 0)) != 0)
{
if (bRet == -1)
{
KillTimer (Window, 1337);
return false;
}
else
{
if (msg.message == WM_TIMER && msg.hwnd == Window && msg.wParam == 1337)
{
if (timer_callback (userdata))
{
KillTimer (NetStartPane, 1);
return true;
}
}
if (!IsDialogMessage (NetStartPane, &msg))
{
TranslateMessage (&msg);
DispatchMessage (&msg);
}
}
}
KillTimer (Window, 1337);
return false;
}
//==========================================================================
//
// NetStartPaneProc
//
// DialogProc for the network startup pane. It just waits for somebody to
// click a button, and the only button available is the abort one.
//
//==========================================================================
static INT_PTR CALLBACK NetStartPaneProc (HWND hDlg, UINT msg, WPARAM wParam, LPARAM 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
{
if (msg == WM_COMMAND && HIWORD(wParam) == BN_CLICKED)
{
PostQuitMessage (0);
return TRUE;
}
return FALSE;
}
//==========================================================================
//
// ST_Hexen_Init
//
// Shows the Hexen startup screen. If the screen doesn't appear to be
// valid, it falls back to ST_Basic_Init.
//
// The startup graphic is a planar, 4-bit 640x480 graphic preceded by a
// 16 entry (48 byte) VGA palette.
//
//==========================================================================
static void ST_Hexen_Init ()
{
int startup_lump = Wads.CheckNumForName ("STARTUP");
if (startup_lump < 0 || Wads.LumpLength (startup_lump) != 153648 || !ST_Util_CreateStartupWindow())
{
ST_Basic_Init ();
return;
}
BYTE startup_screen[153648];
union
{
RGBQUAD color;
DWORD quad;
};
Wads.ReadLump (startup_lump, startup_screen);
color.rgbReserved = 0;
StartupBitmap = ST_Util_CreateBitmap (640, 480, 4);
// Initialize the bitmap palette.
for (int i = 0; i < 16; ++i)
{
color.rgbRed = startup_screen[i*3+0];
color.rgbGreen = startup_screen[i*3+1];
color.rgbBlue = startup_screen[i*3+2];
// Convert from 6-bit per component to 8-bit per component.
quad = (quad << 2) | ((quad >> 4) & 0x03030303);
StartupBitmap->bmiColors[i] = color;
}
// Fill in the bitmap data. Convert to chunky, because I can't figure out
// if Windows actually supports planar images or not, despite the presence
// of biPlanes in the BITMAPINFOHEADER.
ST_Util_PlanarToChunky4 (ST_Util_BitsForBitmap(StartupBitmap), startup_screen + 48, 640, 480);
ST_Util_SizeWindowForBitmap ();
LayoutMainWindow (Window, NULL);
InvalidateRect (StartupScreen, NULL, TRUE);
ST_Done = ST_Hexen_Done;
ST_Progress = ST_Hexen_Progress;
ST_NetInit = ST_Basic_NetInit;
ST_NetProgress = ST_Basic_NetProgress;
ST_NetMessage = ST_Basic_NetMessage;
ST_NetDone = ST_Basic_NetDone;
ST_NetLoop = ST_Basic_NetLoop;
// Not that this screen will be around long enough for anyone to
// really hear the music, but start it anyway.
S_ChangeMusic ("orb", true, true);
}
//==========================================================================
//
// ST_Hexen_Done
//
// Called just before entering graphics mode to deconstruct the startup
// screen.
//
//==========================================================================
static void ST_Hexen_Done()
{
if (StartupScreen != NULL)
{
DestroyWindow (StartupScreen);
StartupScreen = NULL;
}
if (StartupBitmap != NULL)
{
ST_Util_FreeBitmap (StartupBitmap);
StartupBitmap = NULL;
}
}
//==========================================================================
//
// ST_Hexen_Progress
//
// Bumps the progress meter one notch.
//
//==========================================================================
static void ST_Hexen_Progress()
{
int notch_pos, x, y;
if (CurPos < MaxPos)
{
CurPos++;
notch_pos = (CurPos * ST_MAX_NOTCHES) / MaxPos;
if (notch_pos != NotchPos)
{ // Time to draw another notch.
for (; NotchPos < notch_pos; NotchPos++)
{
x = ST_PROGRESS_X + ST_NOTCH_WIDTH * NotchPos;
y = ST_PROGRESS_Y;
ST_Util_DrawBlock (ST_Util_BitsForBitmap(StartupBitmap) + x / 2 + y * 320,
NotchBits, ST_NOTCH_WIDTH / 2, ST_NOTCH_HEIGHT, 320);
ST_Util_InvalidateRect (StartupScreen, StartupBitmap, x, y, x + ST_NOTCH_WIDTH, y + ST_NOTCH_HEIGHT);
}
S_Sound (CHAN_BODY, "StartupTick", 1, ATTN_NONE);
}
}
I_GetEvent ();
}
//==========================================================================
//
// ST_Heretic_Init
//
// Shows the Hexen startup screen. If the screen doesn't appear to be
// valid, it falls back to ST_Basic_Init.
//
// The loading screen is an 80x25 text screen with character data and
// attributes intermixed, which means it must be exactly 4000 bytes long.
//
//==========================================================================
static void ST_Heretic_Init ()
{
int loading_lump = Wads.CheckNumForName ("LOADING");
BYTE loading_screen[4000];
BYTE *font;
if (loading_lump < 0 || Wads.LumpLength (loading_lump) != 4000 || !ST_Util_CreateStartupWindow())
{
ST_Basic_Init ();
return;
}
font = ST_Util_LoadFont (TEXT_FONT_NAME, TEXT_FONT_HEIGHT);
if (font == NULL)
{
DestroyWindow (StartupScreen);
ST_Basic_Init ();
return;
}
Wads.ReadLump (loading_lump, loading_screen);
// Slap the Heretic minor version on the loading screen. Heretic
// did this inside the executable rather than coming with modified
// LOADING screens, so we need to do the same.
loading_screen[2*160 + 49*2] = HERETIC_MINOR_VERSION;
// Draw the loading screen to a bitmap.
StartupBitmap = ST_Util_AllocTextBitmap (font);
ST_Util_DrawTextScreen (StartupBitmap, loading_screen, font);
ST_Util_FreeFont (font);
ST_Util_SizeWindowForBitmap ();
LayoutMainWindow (Window, NULL);
InvalidateRect (StartupScreen, NULL, TRUE);
ThermX = THERM_X * 4;
ThermY = THERM_Y * TEXT_FONT_HEIGHT;
ThermWidth = THERM_LEN * 4;
ThermHeight = TEXT_FONT_HEIGHT;
ST_Done = ST_Hexen_Done;
ST_Progress = ST_Heretic_Progress;
ST_NetInit = ST_Basic_NetInit;
ST_NetProgress = ST_Basic_NetProgress;
ST_NetMessage = ST_Basic_NetMessage;
ST_NetDone = ST_Basic_NetDone;
ST_NetLoop = ST_Basic_NetLoop;
}
//==========================================================================
//
// ST_Heretic_Progress
//
// Bumps the progress meter one notch.
//
//==========================================================================
static void ST_Heretic_Progress()
{
int notch_pos;
if (CurPos < MaxPos)
{
CurPos++;
notch_pos = (CurPos * ThermWidth) / MaxPos;
if (notch_pos != NotchPos && !(notch_pos & 3))
{ // Time to draw another notch.
int left = NotchPos + ThermX;
int top = ThermY;
int right = notch_pos + ThermX;
int bottom = top + ThermHeight;
ST_Util_ClearBlock (ST_Util_BitsForBitmap(StartupBitmap) + left + top * 320, THERM_COLOR, right - left, bottom - top, 320);
ST_Util_InvalidateRect (StartupScreen, StartupBitmap, left*2, top, right*2, bottom);
NotchPos = notch_pos;
}
}
I_GetEvent ();
}
//==========================================================================
//
// ST_Endoom
//
// Shows an ENDOOM text screen
//
//==========================================================================
CVAR(Bool, showendoom, true, CVAR_ARCHIVE|CVAR_GLOBALCONFIG)
void ST_Endoom()
{
if (!showendoom) exit(0);
int endoom_lump = Wads.CheckNumForName (
gameinfo.gametype == GAME_Doom? "ENDOOM" :
gameinfo.gametype == GAME_Heretic? "ENDTEXT" :
gameinfo.gametype == GAME_Strife? "ENDSTRF" : NULL);
BYTE endoom_screen[4000];
BYTE *font;
MSG mess;
if (endoom_lump < 0 || Wads.LumpLength (endoom_lump) != 4000)
{
exit(0);
}
font = ST_Util_LoadFont (TEXT_FONT_NAME, TEXT_FONT_HEIGHT);
if (font == NULL)
{
exit(0);
}
if (!ST_Util_CreateStartupWindow())
{
ST_Util_FreeFont (font);
exit(0);
}
ST_Done = ST_Basic_Done;
I_ShutdownGraphics ();
RestoreConView ();
S_StopMusic(true);
Wads.ReadLump (endoom_lump, endoom_screen);
// Draw the loading screen to a bitmap.
StartupBitmap = ST_Util_AllocTextBitmap (font);
ST_Util_DrawTextScreen (StartupBitmap, endoom_screen, font);
ST_Util_FreeFont (font);
ST_Util_SizeWindowForBitmap ();
LayoutMainWindow (Window, NULL);
InvalidateRect (StartupScreen, NULL, TRUE);
// Wait until any key has been pressed or a quit message has been received
while (1)
{
if (PeekMessage (&mess, NULL, 0, 0, PM_REMOVE))
{
if (mess.message == WM_QUIT)
exit (int(mess.wParam));
if (mess.message == WM_KEYDOWN || mess.message == WM_SYSKEYDOWN || mess.message == WM_LBUTTONDOWN)
exit(0);
TranslateMessage (&mess);
DispatchMessage (&mess);
}
else WaitMessage();
}
}
//==========================================================================
//
// ST_Util_CreateStartupWindow
//
// Creates the static control that will draw the startup screen.
//
//==========================================================================
bool ST_Util_CreateStartupWindow ()
{
StartupScreen = CreateWindowEx (WS_EX_NOPARENTNOTIFY, "STATIC", NULL,
WS_CHILD | WS_VISIBLE | WS_CLIPSIBLINGS | SS_OWNERDRAW,
0, 0, 0, 0, Window, NULL, g_hInst, NULL);
if (StartupScreen == NULL)
{
return false;
}
SetWindowLong (StartupScreen, GWL_ID, IDC_STATIC_STARTUP);
return true;
}
//==========================================================================
//
// ST_Util_SizeWindowForBitmap
//
// Resizes the main window so that the startup bitmap will be draw 1:1.
//
//==========================================================================
void ST_Util_SizeWindowForBitmap ()
{
DEVMODE displaysettings;
int w, h, cx, cy, x, y;
RECT rect;
GetClientRect (GameTitleWindow, &rect);
w = StartupBitmap->bmiHeader.biWidth + GetSystemMetrics (SM_CXSIZEFRAME)*2;
h = StartupBitmap->bmiHeader.biHeight + rect.bottom
+ GetSystemMetrics (SM_CYSIZEFRAME) * 2 + GetSystemMetrics (SM_CYCAPTION);
// Resize the window, but keep its center point the same, unless that
// puts it partially offscreen.
memset (&displaysettings, 0, sizeof(displaysettings));
displaysettings.dmSize = sizeof(displaysettings);
EnumDisplaySettings (NULL, ENUM_CURRENT_SETTINGS, &displaysettings);
GetWindowRect (Window, &rect);
cx = (rect.left + rect.right) / 2;
cy = (rect.top + rect.bottom) / 2;
x = cx - w / 2;
y = cy - h / 2;
if (x + w > (int)displaysettings.dmPelsWidth)
{
x = displaysettings.dmPelsWidth - w;
}
if (x < 0)
{
x = 0;
}
if (y + h > (int)displaysettings.dmPelsHeight)
{
y = displaysettings.dmPelsHeight - h;
}
if (y < 0)
{
y = 0;
}
MoveWindow (Window, x, y, w, h, TRUE);
}
//==========================================================================
//
// ST_Util_PlanarToChunky4
//
// Convert a 4-bpp planar image to chunky pixels.
//
//==========================================================================
void ST_Util_PlanarToChunky4 (BYTE *dest, const BYTE *src, int width, int height)
{
int y, x;
const BYTE *src1, *src2, *src3, *src4;
size_t plane_size = width / 8 * height;
src1 = src;
src2 = src1 + plane_size;
src3 = src2 + plane_size;
src4 = src3 + plane_size;
for (y = height; y > 0; --y)
{
for (x = width; x > 0; x -= 8)
{
// Pixels 0 and 1
dest[0] = (*src4 & 0x80) | ((*src3 & 0x80) >> 1) | ((*src2 & 0x80) >> 2) | ((*src1 & 0x80) >> 3) |
((*src4 & 0x40) >> 3) | ((*src3 & 0x40) >> 4) | ((*src2 & 0x40) >> 5) | ((*src1 & 0x40) >> 6);
// Pixels 2 and 3
dest[1] = ((*src4 & 0x20) << 2) | ((*src3 & 0x20) << 1) | ((*src2 & 0x20)) | ((*src1 & 0x20) >> 1) |
((*src4 & 0x10) >> 1) | ((*src3 & 0x10) >> 2) | ((*src2 & 0x10) >> 3) | ((*src1 & 0x10) >> 4);
// Pixels 4 and 5
dest[2] = ((*src4 & 0x08) << 4) | ((*src3 & 0x08) << 3) | ((*src2 & 0x08) << 2) | ((*src1 & 0x08) << 1) |
((*src4 & 0x04) << 1) | ((*src3 & 0x04)) | ((*src2 & 0x04) >> 1) | ((*src1 & 0x04) >> 2);
// Pixels 6 and 7
dest[3] = ((*src4 & 0x02) << 6) | ((*src3 & 0x02) << 5) | ((*src2 & 0x02) << 4) | ((*src1 & 0x02) << 3) |
((*src4 & 0x01) << 3) | ((*src3 & 0x01) << 2) | ((*src2 & 0x01) << 1) | ((*src1 & 0x01));
dest += 4;
src1 += 1;
src2 += 1;
src3 += 1;
src4 += 1;
}
}
}
//==========================================================================
//
// ST_Util_DrawBlock
//
//==========================================================================
void ST_Util_DrawBlock (BYTE *dest, const BYTE *src, int bytewidth, int height, int destpitch)
{
if (bytewidth == 8)
{ // progress notches
for (; height > 0; --height)
{
((DWORD *)dest)[0] = ((const DWORD *)src)[0];
((DWORD *)dest)[1] = ((const DWORD *)src)[1];
dest += destpitch;
src += 8;
}
}
else if (bytewidth == 2)
{ // net progress notches
for (; height > 0; --height)
{
*((WORD *)dest) = *((const WORD *)src);
dest += destpitch;
src += 2;
}
}
}
//==========================================================================
//
// ST_Util_ClearBlock
//
//==========================================================================
void ST_Util_ClearBlock (BYTE *dest, BYTE fill, int bytewidth, int height, int destpitch)
{
while (height > 0)
{
memset (dest, fill, bytewidth);
dest += destpitch;
height--;
}
}
//==========================================================================
//
// ST_Util_CreateBitmap
//
// Creates a BITMAPINFOHEADER, RGBQUAD, and pixel data arranged
// consecutively in memory (in other words, a normal Windows BMP file).
// The BITMAPINFOHEADER will be filled in, and the caller must fill
// in the color and pixel data.
//
// You must pass 4 or 8 for color_bits.
//
//==========================================================================
BITMAPINFO *ST_Util_CreateBitmap (int width, int height, int color_bits)
{
DWORD size_image = (width * height) >> int(color_bits == 4);
BITMAPINFO *bitmap_info = (BITMAPINFO *)M_Malloc (sizeof(BITMAPINFOHEADER) +
(sizeof(RGBQUAD) << color_bits) + size_image);
// Initialize the header.
bitmap_info->bmiHeader.biSize = sizeof(BITMAPINFOHEADER);
bitmap_info->bmiHeader.biWidth = width;
bitmap_info->bmiHeader.biHeight = height;
bitmap_info->bmiHeader.biPlanes = 1;
bitmap_info->bmiHeader.biBitCount = color_bits;
bitmap_info->bmiHeader.biCompression = 0;
bitmap_info->bmiHeader.biSizeImage = size_image;
bitmap_info->bmiHeader.biXPelsPerMeter = 0;
bitmap_info->bmiHeader.biYPelsPerMeter = 0;
bitmap_info->bmiHeader.biClrUsed = 1 << color_bits;
bitmap_info->bmiHeader.biClrImportant = 0;
return bitmap_info;
}
//==========================================================================
//
// ST_Util_BitsForBitmap
//
// Given a bitmap created by ST_Util_CreateBitmap, returns the start
// address for the pixel data for the bitmap.
//
//==========================================================================
BYTE *ST_Util_BitsForBitmap (BITMAPINFO *bitmap_info)
{
return (BYTE *)bitmap_info + sizeof(BITMAPINFOHEADER) + (sizeof(RGBQUAD) << bitmap_info->bmiHeader.biBitCount);
}
//==========================================================================
//
// ST_Util_FreeBitmap
//
// Frees all the data for a bitmap created by ST_Util_CreateBitmap.
//
//==========================================================================
void ST_Util_FreeBitmap (BITMAPINFO *bitmap_info)
{
free (bitmap_info);
}
//==========================================================================
//
// ST_Util_InvalidateRect
//
// Invalidates the portion of the window that the specified rect of the
// bitmap appears in.
//
//==========================================================================
void ST_Util_InvalidateRect (HWND hwnd, BITMAPINFO *bitmap_info, int left, int top, int right, int bottom)
{
RECT rect;
GetClientRect (hwnd, &rect);
rect.left = left * rect.right / bitmap_info->bmiHeader.biWidth - 1;
rect.top = top * rect.bottom / bitmap_info->bmiHeader.biHeight - 1;
rect.right = right * rect.right / bitmap_info->bmiHeader.biWidth + 1;
rect.bottom = bottom * rect.bottom / bitmap_info->bmiHeader.biHeight + 1;
InvalidateRect (hwnd, &rect, FALSE);
}
//==========================================================================
//
// ST_Util_LoadFont
//
// Loads a monochrome fixed-width font. Every character is one byte
// (eight pixels) wide.
//
//==========================================================================
BYTE *ST_Util_LoadFont (const char *filename, int height)
{
int lumpnum, lumplen;
BYTE *font;
lumpnum = Wads.CheckNumForFullName (filename);
if (lumpnum < 0)
{ // font not found
return NULL;
}
lumplen = Wads.LumpLength (lumpnum);
if (lumplen != height << 8)
{ // font is a bad size
return NULL;
}
font = new BYTE[lumplen + 1];
font[0] = height; // Store font height in the first byte.
Wads.ReadLump (lumpnum, font + 1);
return font;
}
void ST_Util_FreeFont (BYTE *font)
{
delete[] font;
}
//==========================================================================
//
// ST_Util_AllocTextBitmap
//
// Returns a bitmap properly sized to hold an 80x25 display of characters
// using the specified font.
//
//==========================================================================
BITMAPINFO *ST_Util_AllocTextBitmap (const BYTE *font)
{
BITMAPINFO *bitmap = ST_Util_CreateBitmap (80 * 8, 25 * font[0], 4);
memcpy (bitmap->bmiColors, TextModePalette, sizeof(TextModePalette));
return bitmap;
}
//==========================================================================
//
// ST_Util_DrawTextScreen
//
// Draws the text screen to the bitmap. The bitmap must be the proper size
// for the font.
//
//==========================================================================
void ST_Util_DrawTextScreen (BITMAPINFO *bitmap_info, const BYTE *text_screen, const BYTE *font)
{
int x, y;
for (y = 0; y < 25; ++y)
{
for (x = 0; x < 80; ++x)
{
ST_Util_DrawChar (bitmap_info, font, x, y, text_screen[0], text_screen[1]);
text_screen += 2;
}
}
}
//==========================================================================
//
// ST_Util_DrawChar
//
// Draws a character on the bitmap. X and Y specify the character cell,
// and fg and bg are 4-bit colors.
//
//==========================================================================
void ST_Util_DrawChar (BITMAPINFO *screen, const BYTE *font, int x, int y, BYTE charnum, BYTE attrib)
{
const BYTE bg_left = attrib & 0xF0;
const BYTE fg = attrib & 0x0F;
const BYTE fg_left = fg << 4;
const BYTE bg = bg_left >> 4;
const BYTE color_array[4] = { bg_left | bg, attrib, fg_left | bg, fg_left | fg };
const BYTE *src = font + 1 + charnum * font[0];
int pitch = screen->bmiHeader.biWidth >> 1;
BYTE *dest = ST_Util_BitsForBitmap(screen) + x*4 + y * font[0] * pitch;
for (y = font[0]; y > 0; --y)
{
BYTE srcbyte = *src++;
// Pixels 0 and 1
dest[0] = color_array[(srcbyte >> 6) & 3];
// Pixels 2 and 3
dest[1] = color_array[(srcbyte >> 4) & 3];
// Pixels 4 and 5
dest[2] = color_array[(srcbyte >> 2) & 3];
// Pixels 6 and 7
dest[3] = color_array[(srcbyte) & 3];
dest += pitch;
}
}