qzdoom/src/i_net.cpp

1067 lines
24 KiB
C++
Raw Normal View History

// Emacs style mode select -*- C++ -*-
//-----------------------------------------------------------------------------
//
// $Id: i_net.c,v 1.2 1997/12/29 19:50:54 pekangas Exp $
//
// Copyright (C) 1993-1996 by id Software, Inc.
//
// This source is available for distribution and/or modification
// only under the terms of the DOOM Source Code License as
// published by id Software. All rights reserved.
//
// The source is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// FITNESS FOR A PARTICULAR PURPOSE. See the DOOM Source Code License
// for more details.
//
// DESCRIPTION:
// Low-level networking code. Uses BSD sockets for UDP networking.
//
//-----------------------------------------------------------------------------
/* [Petteri] Check if compiling for Win32: */
#if defined(__WINDOWS__) || defined(__NT__) || defined(_MSC_VER) || defined(_WIN32)
#ifndef __WIN32__
# define __WIN32__
#endif
#endif
/* Follow #ifdef __WIN32__ marks */
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
/* [Petteri] Use Winsock for Win32: */
#ifdef __WIN32__
# define WIN32_LEAN_AND_MEAN
# include <windows.h>
# include <winsock.h>
#define USE_WINDOWS_DWORD
#else
# include <sys/socket.h>
# include <netinet/in.h>
# include <arpa/inet.h>
# include <errno.h>
# include <unistd.h>
# include <netdb.h>
# include <sys/ioctl.h>
# ifdef __sun
# include <fcntl.h>
# endif
#endif
#include "doomtype.h"
#include "i_system.h"
#include "d_event.h"
#include "d_net.h"
#include "m_argv.h"
#include "m_swap.h"
#include "m_crc32.h"
#include "d_player.h"
#include "templates.h"
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
#include "c_console.h"
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
#include "st_start.h"
#include "m_misc.h"
#include "doomstat.h"
#include "i_net.h"
// As per http://support.microsoft.com/kb/q192599/ the standard
// size for network buffers is 8k.
#define TRANSMIT_SIZE 8000
/* [Petteri] Get more portable: */
#ifndef __WIN32__
typedef int SOCKET;
#define SOCKET_ERROR -1
#define INVALID_SOCKET -1
#define closesocket close
#define ioctlsocket ioctl
#define Sleep(x) usleep (x * 1000)
#define WSAEWOULDBLOCK EWOULDBLOCK
#define WSAECONNRESET ECONNRESET
#define WSAGetLastError() errno
#endif
#ifndef IPPORT_USERRESERVED
#define IPPORT_USERRESERVED 5000
#endif
#ifdef __WIN32__
typedef int socklen_t;
#endif
//
// NETWORKING
//
static u_short DOOMPORT = (IPPORT_USERRESERVED + 29);
static SOCKET mysocket = INVALID_SOCKET;
static sockaddr_in sendaddress[MAXNETNODES];
static BYTE sendplayer[MAXNETNODES];
#ifdef __WIN32__
const char *neterror (void);
#else
#define neterror() strerror(errno)
#endif
enum
{
PRE_CONNECT, // Sent from guest to host for initial connection
PRE_KEEPALIVE,
PRE_DISCONNECT, // Sent from guest that aborts the game
PRE_ALLHERE, // Sent from host to guest when everybody has connected
PRE_CONACK, // Sent from host to guest to acknowledge PRE_CONNECT receipt
PRE_ALLFULL, // Sent from host to an unwanted guest
PRE_ALLHEREACK, // Sent from guest to host to acknowledge PRE_ALLHEREACK receipt
PRE_GO // Sent from host to guest to continue game startup
};
// Set PreGamePacket.fake to this so that the game rejects any pregame packets
// after it starts. This translates to NCMD_SETUP|NCMD_MULTI.
#define PRE_FAKE 0x30
struct PreGamePacket
{
BYTE Fake;
BYTE Message;
BYTE NumNodes;
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
union
{
BYTE ConsoleNum;
BYTE NumPresent;
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
};
struct
{
DWORD address;
2014-10-21 10:44:14 +00:00
WORD port;
BYTE player;
BYTE pad;
} machines[MAXNETNODES];
};
BYTE TransmitBuffer[TRANSMIT_SIZE];
//
// UDPsocket
//
SOCKET UDPsocket (void)
{
SOCKET s;
// allocate a socket
s = socket (PF_INET, SOCK_DGRAM, IPPROTO_UDP);
if (s == INVALID_SOCKET)
I_FatalError ("can't create socket: %s", neterror ());
return s;
}
//
// BindToLocalPort
//
void BindToLocalPort (SOCKET s, u_short port)
{
int v;
sockaddr_in address;
memset (&address, 0, sizeof(address));
address.sin_family = AF_INET;
address.sin_addr.s_addr = INADDR_ANY;
address.sin_port = htons(port);
v = bind (s, (sockaddr *)&address, sizeof(address));
if (v == SOCKET_ERROR)
I_FatalError ("BindToPort: %s", neterror ());
}
int FindNode (const sockaddr_in *address)
{
int i;
// find remote node number
for (i = 0; i<doomcom.numnodes; i++)
if (address->sin_addr.s_addr == sendaddress[i].sin_addr.s_addr
&& address->sin_port == sendaddress[i].sin_port)
break;
if (i == doomcom.numnodes)
{
// packet is not from one of the players (new game broadcast?)
i = -1;
}
return i;
}
//
// PacketSend
//
void PacketSend (void)
{
int c;
// FIXME: Catch this before we've overflown the buffer. With long chat
// text and lots of backup tics, it could conceivably happen. (Though
// apparently it hasn't yet, which is good.)
if (doomcom.datalength > MAX_MSGLEN)
{
I_FatalError("Netbuffer overflow!");
}
assert(!(doomcom.data[0] & NCMD_COMPRESSED));
uLong size = TRANSMIT_SIZE - 1;
if (doomcom.datalength >= 10)
{
TransmitBuffer[0] = doomcom.data[0] | NCMD_COMPRESSED;
c = compress2(TransmitBuffer + 1, &size, doomcom.data + 1, doomcom.datalength - 1, 9);
size += 1;
}
else
{
c = -1; // Just some random error code to avoid sending the compressed buffer.
}
if (c == Z_OK && size < (uLong)doomcom.datalength)
{
// Printf("send %lu/%d\n", size, doomcom.datalength);
c = sendto(mysocket, (char *)TransmitBuffer, size,
0, (sockaddr *)&sendaddress[doomcom.remotenode],
sizeof(sendaddress[doomcom.remotenode]));
}
else
{
if (doomcom.datalength > TRANSMIT_SIZE)
{
I_Error("Net compression failed (zlib error %d)", c);
}
else
{
// Printf("send %d\n", doomcom.datalength);
c = sendto(mysocket, (char *)doomcom.data, doomcom.datalength,
0, (sockaddr *)&sendaddress[doomcom.remotenode],
sizeof(sendaddress[doomcom.remotenode]));
}
}
// if (c == -1)
// I_Error ("SendPacket error: %s",strerror(errno));
}
//
// PacketGet
//
void PacketGet (void)
{
int c;
socklen_t fromlen;
sockaddr_in fromaddress;
int node;
fromlen = sizeof(fromaddress);
c = recvfrom (mysocket, (char*)TransmitBuffer, TRANSMIT_SIZE, 0,
(sockaddr *)&fromaddress, &fromlen);
node = FindNode (&fromaddress);
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
if (node >= 0 && c == SOCKET_ERROR)
{
int err = WSAGetLastError();
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
if (err == WSAECONNRESET)
{ // The remote node aborted unexpectedly, so pretend it sent an exit packet
if (StartScreen != NULL)
{
StartScreen->NetMessage ("The connection from %s was dropped.\n",
players[sendplayer[node]].userinfo.GetName());
}
else
{
Printf("The connection from %s was dropped.\n",
players[sendplayer[node]].userinfo.GetName());
}
doomcom.data[0] = 0x80; // NCMD_EXIT
c = 1;
}
else if (err != WSAEWOULDBLOCK)
{
I_Error ("GetPacket: %s", neterror ());
}
else
{
doomcom.remotenode = -1; // no packet
return;
}
}
else if (node >= 0 && c > 0)
{
doomcom.data[0] = TransmitBuffer[0] & ~NCMD_COMPRESSED;
if (TransmitBuffer[0] & NCMD_COMPRESSED)
{
uLongf msgsize = MAX_MSGLEN - 1;
int err = uncompress(doomcom.data + 1, &msgsize, TransmitBuffer + 1, c - 1);
// Printf("recv %d/%lu\n", c, msgsize + 1);
if (err != Z_OK)
{
Printf("Net decompression failed (zlib error %s)\n", M_ZLibError(err).GetChars());
// Pretend no packet
doomcom.remotenode = -1;
return;
}
c = msgsize + 1;
}
else
{
// Printf("recv %d\n", c);
memcpy(doomcom.data + 1, TransmitBuffer + 1, c - 1);
}
}
else if (c > 0)
{ //The packet is not from any in-game node, so we might as well discard it.
// Don't show the message for disconnect notifications.
if (c != 2 || TransmitBuffer[0] != PRE_FAKE || TransmitBuffer[1] != PRE_DISCONNECT)
{
DPrintf(DMSG_WARNING, "Dropped packet: Unknown host (%s:%d)\n", inet_ntoa(fromaddress.sin_addr), fromaddress.sin_port);
}
doomcom.remotenode = -1;
return;
}
doomcom.remotenode = node;
doomcom.datalength = (short)c;
}
sockaddr_in *PreGet (void *buffer, int bufferlen, bool noabort)
{
static sockaddr_in fromaddress;
socklen_t fromlen;
int c;
fromlen = sizeof(fromaddress);
c = recvfrom (mysocket, (char *)buffer, bufferlen, 0,
(sockaddr *)&fromaddress, &fromlen);
if (c == SOCKET_ERROR)
{
int err = WSAGetLastError();
if (err == WSAEWOULDBLOCK || (noabort && err == WSAECONNRESET))
return NULL; // no packet
I_Error ("PreGet: %s", neterror ());
}
return &fromaddress;
}
void PreSend (const void *buffer, int bufferlen, const sockaddr_in *to)
{
sendto (mysocket, (const char *)buffer, bufferlen, 0, (const sockaddr *)to, sizeof(*to));
}
void BuildAddress (sockaddr_in *address, const char *name)
{
hostent *hostentry; // host information entry
u_short port;
const char *portpart;
bool isnamed = false;
int curchar;
char c;
FString target;
address->sin_family = AF_INET;
if ( (portpart = strchr (name, ':')) )
{
target = FString(name, portpart - name);
port = atoi (portpart + 1);
if (!port)
{
Printf ("Weird port: %s (using %d)\n", portpart + 1, DOOMPORT);
port = DOOMPORT;
}
}
else
{
target = name;
port = DOOMPORT;
}
address->sin_port = htons(port);
for (curchar = 0; (c = target[curchar]) ; curchar++)
{
if ((c < '0' || c > '9') && c != '.')
{
isnamed = true;
break;
}
}
if (!isnamed)
{
address->sin_addr.s_addr = inet_addr (target);
Printf ("Node number %d, address %s\n", doomcom.numnodes, target.GetChars());
}
else
{
hostentry = gethostbyname (target);
if (!hostentry)
I_FatalError ("gethostbyname: couldn't find %s\n%s", target.GetChars(), neterror());
address->sin_addr.s_addr = *(int *)hostentry->h_addr_list[0];
Printf ("Node number %d, hostname %s\n",
doomcom.numnodes, hostentry->h_name);
}
}
void CloseNetwork (void)
{
if (mysocket != INVALID_SOCKET)
{
closesocket (mysocket);
mysocket = INVALID_SOCKET;
}
#ifdef __WIN32__
WSACleanup ();
#endif
}
void StartNetwork (bool autoPort)
{
u_long trueval = 1;
#ifdef __WIN32__
WSADATA wsad;
if (WSAStartup (0x0101, &wsad))
{
I_FatalError ("Could not initialize Windows Sockets");
}
#endif
atterm (CloseNetwork);
netgame = true;
multiplayer = true;
// create communication socket
mysocket = UDPsocket ();
BindToLocalPort (mysocket, autoPort ? 0 : DOOMPORT);
#ifndef __sun
ioctlsocket (mysocket, FIONBIO, &trueval);
#else
fcntl(mysocket, F_SETFL, trueval | O_NONBLOCK);
#endif
}
void SendAbort (void)
{
BYTE dis[2] = { PRE_FAKE, PRE_DISCONNECT };
int i, j;
if (doomcom.numnodes > 1)
{
if (consoleplayer == 0)
{
// The host needs to let everyone know
for (i = 1; i < doomcom.numnodes; ++i)
{
for (j = 4; j > 0; --j)
{
PreSend (dis, 2, &sendaddress[i]);
}
}
}
else
{
// Guests only need to let the host know.
for (i = 4; i > 0; --i)
{
PreSend (dis, 2, &sendaddress[1]);
}
}
}
}
static void SendConAck (int num_connected, int num_needed)
{
PreGamePacket packet;
packet.Fake = PRE_FAKE;
packet.Message = PRE_CONACK;
packet.NumNodes = num_needed;
packet.NumPresent = num_connected;
for (int node = 1; node < doomcom.numnodes; ++node)
{
PreSend (&packet, 4, &sendaddress[node]);
}
StartScreen->NetProgress (doomcom.numnodes);
}
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 Host_CheckForConnects (void *userdata)
{
PreGamePacket packet;
int numplayers = (int)(intptr_t)userdata;
sockaddr_in *from;
int node;
while ( (from = PreGet (&packet, sizeof(packet), false)) )
{
if (packet.Fake != PRE_FAKE)
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
{
continue;
}
switch (packet.Message)
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
{
case PRE_CONNECT:
node = FindNode (from);
if (doomcom.numnodes == numplayers)
{
if (node == -1)
{
const BYTE *s_addr_bytes = (const BYTE *)&from->sin_addr;
StartScreen->NetMessage ("Got extra connect from %d.%d.%d.%d:%d",
s_addr_bytes[0], s_addr_bytes[1], s_addr_bytes[2], s_addr_bytes[3],
from->sin_port);
packet.Message = PRE_ALLFULL;
PreSend (&packet, 2, from);
}
}
else
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 (node == -1)
{
node = doomcom.numnodes++;
sendaddress[node] = *from;
StartScreen->NetMessage ("Got connect from node %d.", node);
}
// Let the new guest (and everyone else) know we got their message.
SendConAck (doomcom.numnodes, 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
}
break;
case PRE_DISCONNECT:
node = FindNode (from);
if (node >= 0)
{
StartScreen->NetMessage ("Got disconnect from node %d.", node);
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
doomcom.numnodes--;
while (node < doomcom.numnodes)
{
sendaddress[node] = sendaddress[node+1];
node++;
}
// Let remaining guests know that somebody left.
SendConAck (doomcom.numnodes, 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
}
break;
case PRE_KEEPALIVE:
break;
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 (doomcom.numnodes < numplayers)
{
// Send message to everyone as a keepalive
SendConAck(doomcom.numnodes, 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
return false;
}
// It's possible somebody bailed out after all players were found.
// Unfortunately, this isn't guaranteed to catch all of them.
// Oh well. Better than nothing.
while ( (from = PreGet (&packet, sizeof(packet), false)) )
{
if (packet.Fake == PRE_FAKE && packet.Message == PRE_DISCONNECT)
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
{
node = FindNode (from);
if (node >= 0)
{
doomcom.numnodes--;
while (node < doomcom.numnodes)
{
sendaddress[node] = sendaddress[node+1];
node++;
}
// Let remaining guests know that somebody left.
SendConAck (doomcom.numnodes, 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
}
break;
}
}
return doomcom.numnodes >= numplayers;
}
bool Host_SendAllHere (void *userdata)
{
int *gotack = (int *)userdata; // ackcount is at gotack[MAXNETNODES]
PreGamePacket packet;
int node;
sockaddr_in *from;
// Send out address information to all guests. Guests that have already
// acknowledged receipt effectively get just a heartbeat packet.
packet.Fake = PRE_FAKE;
packet.Message = PRE_ALLHERE;
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
for (node = 1; node < doomcom.numnodes; node++)
{
int machine, spot = 0;
packet.ConsoleNum = node;
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 (!gotack[node])
{
for (spot = 0, machine = 1; machine < doomcom.numnodes; machine++)
{
if (node != machine)
{
packet.machines[spot].address = sendaddress[machine].sin_addr.s_addr;
packet.machines[spot].port = sendaddress[machine].sin_port;
packet.machines[spot].player = node;
spot++; // fixes problem of new address replacing existing address in
// array; it's supposed to increment the index before getting
// and storing in the packet the next address.
}
}
packet.NumNodes = doomcom.numnodes - 2;
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
}
else
{
packet.NumNodes = 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
}
PreSend (&packet, 4 + spot*8, &sendaddress[node]);
}
// Check for replies.
while ( (from = PreGet (&packet, sizeof(packet), false)) )
{
if (packet.Fake == PRE_FAKE && packet.Message == PRE_ALLHEREACK)
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
{
node = FindNode (from);
if (node >= 0)
{
if (!gotack[node])
{
gotack[node] = true;
gotack[MAXNETNODES]++;
}
}
PreSend (&packet, 2, from);
}
}
// If everybody has replied, then this loop can end.
return gotack[MAXNETNODES] == doomcom.numnodes - 1;
}
void HostGame (int i)
{
PreGamePacket packet;
int numplayers;
int node;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
int gotack[MAXNETNODES+1];
if ((i == Args->NumArgs() - 1) || !(numplayers = atoi (Args->GetArg(i+1))))
{ // No player count specified, assume 2
numplayers = 2;
}
if (numplayers > MAXNETNODES)
{
I_FatalError("You cannot host a game with %d players. The limit is currently %d.", numplayers, MAXNETNODES);
return;
}
if (numplayers == 1)
{ // Special case: Only 1 player, so don't bother starting the network
netgame = false;
multiplayer = true;
doomcom.id = DOOMCOM_ID;
doomcom.numplayers = doomcom.numnodes = 1;
doomcom.consoleplayer = 0;
return;
}
StartNetwork (false);
// [JC] - this computer is starting the game, therefore it should
// be the Net Arbitrator.
doomcom.consoleplayer = 0;
Printf ("Console player number: %d\n", doomcom.consoleplayer);
doomcom.numnodes = 1;
atterm (SendAbort);
StartScreen->NetInit ("Waiting for players", numplayers);
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
// Wait for numplayers-1 different connections
if (!StartScreen->NetLoop (Host_CheckForConnects, (void *)(intptr_t)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
exit (0);
}
// Now inform everyone of all machines involved in the game
memset (gotack, 0, sizeof(gotack));
StartScreen->NetMessage ("Sending all here.");
StartScreen->NetInit ("Done waiting", 1);
if (!StartScreen->NetLoop (Host_SendAllHere, (void *)gotack))
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
{
exit (0);
}
popterm ();
// Now go
StartScreen->NetMessage ("Go");
packet.Fake = PRE_FAKE;
packet.Message = PRE_GO;
for (node = 1; node < doomcom.numnodes; node++)
{
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 we send the packets eight times to each guest,
// hopefully at least one of them will get through.
for (int i = 8; i != 0; --i)
{
PreSend (&packet, 2, &sendaddress[node]);
}
}
StartScreen->NetMessage ("Total players: %d", doomcom.numnodes);
doomcom.id = DOOMCOM_ID;
doomcom.numplayers = doomcom.numnodes;
// On the host, each player's number is the same as its node number
for (i = 0; i < doomcom.numnodes; ++i)
{
sendplayer[i] = i;
}
}
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
// This routine is used by a guest to notify the host of its presence.
// Once that host acknowledges receipt of the notification, this routine
// is never called again.
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 Guest_ContactHost (void *userdata)
{
sockaddr_in *from;
PreGamePacket packet;
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
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
// Let the host know we are here.
packet.Fake = PRE_FAKE;
packet.Message = PRE_CONNECT;
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
PreSend (&packet, 2, &sendaddress[1]);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
// Listen for a reply.
while ( (from = PreGet (&packet, sizeof(packet), true)) )
{
if (packet.Fake == PRE_FAKE && FindNode(from) == 1)
{
if (packet.Message == PRE_CONACK)
{
StartScreen->NetMessage ("Total players: %d", packet.NumNodes);
StartScreen->NetInit ("Waiting for other players", packet.NumNodes);
StartScreen->NetProgress (packet.NumPresent);
return true;
}
else if (packet.Message == PRE_DISCONNECT)
{
doomcom.numnodes = 0;
I_FatalError ("The host cancelled the game.");
}
else if (packet.Message == PRE_ALLFULL)
{
doomcom.numnodes = 0;
I_FatalError ("The game is full.");
}
}
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
}
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
// In case the progress bar could not be marqueed, bump it.
StartScreen->NetProgress (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
return false;
}
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 Guest_WaitForOthers (void *userdata)
{
sockaddr_in *from;
PreGamePacket packet;
while ( (from = PreGet (&packet, sizeof(packet), false)) )
{
if (packet.Fake != PRE_FAKE || FindNode(from) != 1)
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
{
continue;
}
switch (packet.Message)
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
{
case PRE_CONACK:
StartScreen->NetProgress (packet.NumPresent);
break;
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
case PRE_ALLHERE:
if (doomcom.numnodes == 2)
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
{
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
int node;
doomcom.numnodes = packet.NumNodes + 2;
sendplayer[0] = packet.ConsoleNum; // My player number
doomcom.consoleplayer = packet.ConsoleNum;
StartScreen->NetMessage ("Console player number: %d", doomcom.consoleplayer);
for (node = 0; node < packet.NumNodes; node++)
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
{
sendaddress[node+2].sin_addr.s_addr = packet.machines[node].address;
sendaddress[node+2].sin_port = packet.machines[node].port;
sendplayer[node+2] = packet.machines[node].player;
// [JC] - fixes problem of games not starting due to
// no address family being assigned to nodes stored in
// sendaddress[] from the All Here packet.
sendaddress[node+2].sin_family = AF_INET;
}
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
}
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
StartScreen->NetMessage ("Received All Here, sending ACK.");
packet.Fake = PRE_FAKE;
packet.Message = PRE_ALLHEREACK;
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
PreSend (&packet, 2, &sendaddress[1]);
break;
case PRE_GO:
StartScreen->NetMessage ("Received \"Go.\"");
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
return true;
case PRE_DISCONNECT:
I_FatalError ("The host cancelled the game.");
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
break;
- Added some hackery at the start of MouseRead_Win32() that prevents it from yanking the mouse around if they keys haven't been read yet to combat the same situation that causes the keyboard to return DIERR_NOTACQUIRED in KeyRead(): The window is sort of in focus and sort of not. User.dll considers it to be focused and it's drawn as such, but another focused window is on top of it, and DirectInput doesn't see it as focused. - Fixed: KeyRead() should handle DIERR_NOTACQUIRED errors the same way it handles DIERR_INPUTLOST errors. This can happen if our window had the focus stolen away from it before we tried to acquire the keyboard in DI_Init2(). Strangely, MouseRead_DI() already did this. - When a stack overflow occurs, report.txt now only includes the first and last 16KB of the stack to make it more manageable. - Limited StreamEditBinary() to the first 64KB of the file to keep it from taking too long on large dumps. - And now I know why gathering crash information in the same process that crashed can be bad: Stack overflows. You get one spare page to play with when the stack overflows. MiniDumpWriteDump() needs more than that and causes an access violation when it runs out of leftover stack, silently terminating the application. Windows XP x64 offers SetThreadStackGuarantee() to increase this, but that isn't available on anything older, including 32-bit XP. To get around this, a new thread is created to write the mini dump when the stack overflows. - Changed A_Burnination() to be closer to Strife's. - Fixed: When playing back demos, DoAddBot() can be called without an associated call to SpawnBot(). So if the bot can't spawn, botnum can go negative, which will cause problems later in DCajunMaster::Main() when it sees that wanted_botnum (0) is higher than botnum (-1). - Fixed: Stopping demo recording in multiplayer games should not abruptly drop the recorder out of the game without notifying the other players. In fact, there's no reason why it should drop them out of multiplayer at all. - Fixed: Earthquakes were unreliable in multiplayer games because P_PredictPlayer() did not preserve the player's xviewshift. - Fixed: PlayerIsGone() needs to stop any scripts that belong to the player who left, in addition to executing disconnect scripts. - Fixed: APlayerPawn::AddInventory() should also check for a NULL player->mo in case the player left but somebody still has a reference to their actor. - Fixed: DDrawFB::PaintToWindow() should simulate proper unlocking behavior and set Buffer to NULL. - Improved feedback for network game initialization with the console ticker. - Moved i_net.cpp and i_net.h out of sdl/ and win32/ and into the main source directory. They are identical, so keeping two copies of them is bad. - Fixed: (At least with Creative's driver's,) EAX settings are global and not per-application. So if you play a multiplayer ZDoom game on one computer (or even another EAX-using application), ZDoom needs to restore the environment when it regains focus. - Maybe fixed: (See http://forum.zdoom.org/potato.php?t=10689) Apparently, PacketGet can receive ECONNRESET from nodes that aren't in the game. It should be safe to just ignore these packets. - Fixed: PlayerIsGone() should set the gone player's camera to NULL in case the player who left was player 0. This is because if a remaining player receives a "recoverable" error, they will become player 0. Once that happens, they game will try to update sounds through their camera and crash in FMODSoundRenderer::UpdateListener() because the zones array is now NULL. G_NewInit() should also clear all the player structures. SVN r233 (trunk)
2006-06-30 02:13:26 +00:00
}
}
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
packet.Fake = PRE_FAKE;
packet.Message = PRE_KEEPALIVE;
PreSend(&packet, 2, &sendaddress[1]);
Note: I have not tried compiling these recent changes under Linux. I wouldn't be surprised if it doesn't work. - Reorganized the network startup loops so now they are event driven. There is a single function that gets called to drive it, and it uses callbacks to perform the different stages of the synchronization. This lets me have a nice, responsive abort button instead of the previous unannounced hit-escape-to- abort behavior, and I think the rearranged code is slightly easier to understand too. - Increased the number of bytes for version info during D_ArbitrateNetStart(), in preparation for the day when NETGAMEVERSION requires more than one byte. - I noticed an issue with Vista RC1 and the new fatal error setup. Even after releasing a DirectDraw or Direct3D interface, the DWM can still use the last image drawn using them when it composites the window. It doesn't always do it but it does often enough that it is a real problem. At this point, I don't know if it's a problem with the release version of Vista or not. After messing around, I discovered the problem was caused by ~Win32Video() hiding the window and then having it immediately shown soon after. The DWM kept an image of the window to do the transition effect with, and then when it didn't get a chance to do the transition, it didn't properly forget about its saved image and kept plastering it on top of everything else underneath. - Added a network synchronization panel to the window during netgame startup. - Fixed: PClass::CreateDerivedClass() must initialize StateList to NULL. Otherwise, classic DECORATE definitions generate a big, fat crash. - Resurrected the R_Init progress bar, now as a standard Windows control. - Removed the sound failure dialog. The FMOD setup already defaulted to no sound if initialization failed, so this only applies when snd_output is set to "alternate" which now also falls back to no sound. In addition, it wasn't working right, and I didn't feel like fixing it for the probably 0% of users it affected. - Fixed: The edit control used for logging output added text in reverse order on Win9x. - Went back to the roots and made graphics initialization one of the last things to happen during setup. Now the startup text is visible again. More importantly, the main window is no longer created invisible, which seems to cause trouble with it not always appearing in the taskbar. The fatal error dialog is now also embedded in the main window instead of being a separate modal dialog, so you can play with the log window to see any problems that might be reported there. Rather than completely restoring the original startup order, I tried to keep things as close to the way they were with early graphics startup. In particular, V_Init() now creates a dummy screen so that things that need screen dimensions can get them. It gets replaced by the real screen later in I_InitGraphics(). Will need to check this under Linux to make sure it didn't cause any problems there. - Removed the following stubs that just called functions in Video: - I_StartModeIterator() - I_NextMode() - I_DisplayType() I_FullscreenChanged() was also removed, and a new fullscreen parameter was added to IVideo::StartModeIterator(), since that's all it controlled. - Renamed I_InitHardware() back to I_InitGraphics(), since that's all it's initialized post-1.22. SVN r416 (trunk)
2006-12-19 04:09:10 +00:00
return false;
}
void JoinGame (int i)
{
if ((i == Args->NumArgs() - 1) ||
(Args->GetArg(i+1)[0] == '-') ||
(Args->GetArg(i+1)[0] == '+'))
I_FatalError ("You need to specify the host machine's address");
StartNetwork (true);
// Host is always node 1
BuildAddress (&sendaddress[1], Args->GetArg(i+1));
sendplayer[1] = 0;
doomcom.numnodes = 2;
atterm (SendAbort);
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
// Let host know we are here
StartScreen->NetInit ("Contacting host", 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
if (!StartScreen->NetLoop (Guest_ContactHost, NULL))
{
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
exit (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
// Wait for everyone else to connect
if (!StartScreen->NetLoop (Guest_WaitForOthers, 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
{
exit (0);
}
popterm ();
StartScreen->NetMessage ("Total players: %d", doomcom.numnodes);
doomcom.id = DOOMCOM_ID;
doomcom.numplayers = doomcom.numnodes;
}
static int PrivateNetOf(in_addr in)
{
int addr = ntohl(in.s_addr);
if ((addr & 0xFFFF0000) == 0xC0A80000) // 192.168.0.0
{
return 0xC0A80000;
}
else if ((addr & 0xFFF00000) == 0xAC100000) // 172.16.0.0
{
return 0xAC100000;
}
else if ((addr & 0xFF000000) == 0x0A000000) // 10.0.0.0
{
return 0x0A000000;
}
else if ((addr & 0xFF000000) == 0x7F000000) // 127.0.0.0 (localhost)
{
return 0x7F000000;
}
// Not a private IP
return 0;
}
//
// NodesOnSameNetwork
//
// The best I can really do here is check if the others are on the same
// private network, since that means we (probably) are too.
//
static bool NodesOnSameNetwork()
{
int net1;
net1 = PrivateNetOf(sendaddress[1].sin_addr);
// Printf("net1 = %08x\n", net1);
if (net1 == 0)
{
return false;
}
for (int i = 2; i < doomcom.numnodes; ++i)
{
int net = PrivateNetOf(sendaddress[i].sin_addr);
// Printf("Net[%d] = %08x\n", i, net);
if (net != net1)
{
return false;
}
}
return true;
}
//
// I_InitNetwork
//
// Returns true if packet server mode might be a good idea.
//
bool I_InitNetwork (void)
{
int i;
const char *v;
memset (&doomcom, 0, sizeof(doomcom));
// set up for network
v = Args->CheckValue ("-dup");
if (v)
{
doomcom.ticdup = clamp (atoi (v), 1, MAXTICDUP);
}
else
{
doomcom.ticdup = 1;
}
v = Args->CheckValue ("-port");
if (v)
{
DOOMPORT = atoi (v);
Printf ("using alternate port %i\n", DOOMPORT);
}
// parse network game options,
// player 1: -host <numplayers>
// player x: -join <player 1's address>
if ( (i = Args->CheckParm ("-host")) )
{
HostGame (i);
}
else if ( (i = Args->CheckParm ("-join")) )
{
JoinGame (i);
}
else
{
// single player game
netgame = false;
multiplayer = false;
doomcom.id = DOOMCOM_ID;
doomcom.numplayers = doomcom.numnodes = 1;
doomcom.consoleplayer = 0;
return false;
}
if (doomcom.numnodes < 3)
{ // Packet server mode with only two players is effectively the same as
// peer-to-peer but with some slightly larger packets.
return false;
}
return doomcom.numnodes > 3 || !NodesOnSameNetwork();
}
void I_NetCmd (void)
{
if (doomcom.command == CMD_SEND)
{
PacketSend ();
}
else if (doomcom.command == CMD_GET)
{
PacketGet ();
}
else
I_Error ("Bad net cmd: %i\n",doomcom.command);
}
#ifdef __WIN32__
const char *neterror (void)
{
static char neterr[16];
int code;
switch (code = WSAGetLastError ()) {
case WSAEACCES: return "EACCES";
case WSAEADDRINUSE: return "EADDRINUSE";
case WSAEADDRNOTAVAIL: return "EADDRNOTAVAIL";
case WSAEAFNOSUPPORT: return "EAFNOSUPPORT";
case WSAEALREADY: return "EALREADY";
case WSAECONNABORTED: return "ECONNABORTED";
case WSAECONNREFUSED: return "ECONNREFUSED";
case WSAECONNRESET: return "ECONNRESET";
case WSAEDESTADDRREQ: return "EDESTADDRREQ";
case WSAEFAULT: return "EFAULT";
case WSAEHOSTDOWN: return "EHOSTDOWN";
case WSAEHOSTUNREACH: return "EHOSTUNREACH";
case WSAEINPROGRESS: return "EINPROGRESS";
case WSAEINTR: return "EINTR";
case WSAEINVAL: return "EINVAL";
case WSAEISCONN: return "EISCONN";
case WSAEMFILE: return "EMFILE";
case WSAEMSGSIZE: return "EMSGSIZE";
case WSAENETDOWN: return "ENETDOWN";
case WSAENETRESET: return "ENETRESET";
case WSAENETUNREACH: return "ENETUNREACH";
case WSAENOBUFS: return "ENOBUFS";
case WSAENOPROTOOPT: return "ENOPROTOOPT";
case WSAENOTCONN: return "ENOTCONN";
case WSAENOTSOCK: return "ENOTSOCK";
case WSAEOPNOTSUPP: return "EOPNOTSUPP";
case WSAEPFNOSUPPORT: return "EPFNOSUPPORT";
case WSAEPROCLIM: return "EPROCLIM";
case WSAEPROTONOSUPPORT: return "EPROTONOSUPPORT";
case WSAEPROTOTYPE: return "EPROTOTYPE";
case WSAESHUTDOWN: return "ESHUTDOWN";
case WSAESOCKTNOSUPPORT: return "ESOCKTNOSUPPORT";
case WSAETIMEDOUT: return "ETIMEDOUT";
case WSAEWOULDBLOCK: return "EWOULDBLOCK";
case WSAHOST_NOT_FOUND: return "HOST_NOT_FOUND";
case WSANOTINITIALISED: return "NOTINITIALISED";
case WSANO_DATA: return "NO_DATA";
case WSANO_RECOVERY: return "NO_RECOVERY";
case WSASYSNOTREADY: return "SYSNOTREADY";
case WSATRY_AGAIN: return "TRY_AGAIN";
case WSAVERNOTSUPPORTED: return "VERNOTSUPPORTED";
case WSAEDISCON: return "EDISCON";
default:
About a week's worth of changes here. As a heads-up, I wouldn't be surprised if this doesn't build in Linux right now. The CMakeLists.txt were checked with MinGW and NMake, but how they fair under Linux is an unknown to me at this time. - Converted most sprintf (and all wsprintf) calls to either mysnprintf or FStrings, depending on the situation. - Changed the strings in the wbstartstruct to be FStrings. - Changed myvsnprintf() to output nothing if count is greater than INT_MAX. This is so that I can use a series of mysnprintf() calls and advance the pointer for each one. Once the pointer goes beyond the end of the buffer, the count will go negative, but since it's an unsigned type it will be seen as excessively huge instead. This should not be a problem, as there's no reason for ZDoom to be using text buffers larger than 2 GB anywhere. - Ripped out the disabled bit from FGameConfigFile::MigrateOldConfig(). - Changed CalcMapName() to return an FString instead of a pointer to a static buffer. - Changed startmap in d_main.cpp into an FString. - Changed CheckWarpTransMap() to take an FString& as the first argument. - Changed d_mapname in g_level.cpp into an FString. - Changed DoSubstitution() in ct_chat.cpp to place the substitutions in an FString. - Fixed: The MAPINFO parser wrote into the string buffer to construct a map name when given a Hexen map number. This was fine with the old scanner code, but only a happy coincidence prevents it from crashing with the new code - Added the 'B' conversion specifier to StringFormat::VWorker() for printing binary numbers. - Added CMake support for building with MinGW, MSYS, and NMake. Linux support is probably broken until I get around to booting into Linux again. Niceties provided over the existing Makefiles they're replacing: * All command-line builds can use the same build system, rather than having a separate one for MinGW and another for Linux. * Microsoft's NMake tool is supported as a target. * Progress meters. * Parallel makes work from a fresh checkout without needing to be primed first with a single-threaded make. * Porting to other architectures should be simplified, whenever that day comes. - Replaced the makewad tool with zipdir. This handles the dependency tracking itself instead of generating an external makefile to do it, since I couldn't figure out how to generate a makefile with an external tool and include it with a CMake-generated makefile. Where makewad used a master list of files to generate the package file, zipdir just zips the entire contents of one or more directories. - Added the gdtoa package from netlib's fp library so that ZDoom's printf-style formatting can be entirely independant of the CRT. SVN r1082 (trunk)
2008-07-23 04:57:26 +00:00
mysnprintf (neterr, countof(neterr), "%d", code);
return neterr;
}
}
#endif