2007-05-30 21:56:52 +00:00
|
|
|
/* ScummVM - Graphic Adventure Engine
|
|
|
|
*
|
|
|
|
* ScummVM is the legal property of its developers, whose names
|
|
|
|
* are too numerous to list here. Please refer to the COPYRIGHT
|
|
|
|
* file distributed with this source distribution.
|
2003-07-28 01:44:38 +00:00
|
|
|
*
|
2007-05-31 20:28:29 +00:00
|
|
|
* Additional copyright for this file:
|
|
|
|
* Copyright (C) 1994-1998 Revolution Software Ltd.
|
|
|
|
*
|
2003-07-28 01:44:38 +00:00
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version 2
|
|
|
|
* of the License, or (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software
|
2005-10-18 01:30:26 +00:00
|
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
2003-07-28 01:44:38 +00:00
|
|
|
*
|
2006-02-09 15:12:44 +00:00
|
|
|
* $URL$
|
|
|
|
* $Id$
|
2003-07-28 01:44:38 +00:00
|
|
|
*/
|
|
|
|
|
2007-09-19 08:40:12 +00:00
|
|
|
|
2006-02-17 15:07:36 +00:00
|
|
|
|
2003-10-28 19:51:30 +00:00
|
|
|
#include "sword2/sword2.h"
|
2006-02-17 15:07:36 +00:00
|
|
|
#include "sword2/header.h"
|
2004-02-05 14:19:07 +00:00
|
|
|
#include "sword2/resman.h"
|
2003-07-28 01:44:38 +00:00
|
|
|
|
2003-10-04 00:52:27 +00:00
|
|
|
namespace Sword2 {
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the first palette entry, given the pointer to the start
|
|
|
|
* of the screen file.
|
|
|
|
*/
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2004-04-23 07:02:11 +00:00
|
|
|
byte *Sword2Engine::fetchPalette(byte *screenFile) {
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
|
|
|
|
|
|
|
byte *palette = screenFile + ResHeader::size() + mscreenHeader.palette;
|
2003-09-19 13:55:19 +00:00
|
|
|
|
|
|
|
// Always set colour 0 to black, because while most background screen
|
|
|
|
// palettes have a bright colour 0 it should come out as black in the
|
|
|
|
// game.
|
2003-07-28 01:44:38 +00:00
|
|
|
|
2003-09-19 13:55:19 +00:00
|
|
|
palette[0] = 0;
|
|
|
|
palette[1] = 0;
|
|
|
|
palette[2] = 0;
|
2003-07-28 01:44:38 +00:00
|
|
|
palette[3] = 0;
|
2005-07-30 21:11:48 +00:00
|
|
|
|
2003-07-28 01:44:38 +00:00
|
|
|
return palette;
|
|
|
|
}
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the start of the palette match table, given the pointer
|
|
|
|
* to the start of the screen file.
|
|
|
|
*/
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2004-04-23 07:02:11 +00:00
|
|
|
byte *Sword2Engine::fetchPaletteMatchTable(byte *screenFile) {
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
MultiScreenHeader mscreenHeader;
|
|
|
|
|
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.paletteTable;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the screen header, given the pointer to the start of
|
|
|
|
* the screen file.
|
|
|
|
*/
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchScreenHeader(byte *screenFile) {
|
|
|
|
MultiScreenHeader mscreenHeader;
|
|
|
|
|
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.screen;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the requested layer header, given the pointer to the
|
|
|
|
* start of the screen file. Drops out if the requested layer number exceeds
|
|
|
|
* the number of layers on this screen.
|
|
|
|
*/
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchLayerHeader(byte *screenFile, uint16 layerNo) {
|
2004-11-16 09:15:25 +00:00
|
|
|
#ifdef SWORD2_DEBUG
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
ScreenHeader screenHead;
|
2003-09-07 03:18:27 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
screenHead.read(fetchScreenHeader(screenFile));
|
|
|
|
assert(layerNo < screenHead.noLayers);
|
2003-07-28 01:44:38 +00:00
|
|
|
#endif
|
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
2003-09-07 03:18:27 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.layers + layerNo * LayerHeader::size();
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the start of the shading mask, given the pointer to the
|
|
|
|
* start of the screen file.
|
|
|
|
*/
|
2003-07-28 01:44:38 +00:00
|
|
|
|
2004-04-23 07:02:11 +00:00
|
|
|
byte *Sword2Engine::fetchShadingMask(byte *screenFile) {
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
|
|
|
|
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.maskOffset;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the anim header, given the pointer to the start of the
|
|
|
|
* anim file.
|
|
|
|
*/
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchAnimHeader(byte *animFile) {
|
|
|
|
return animFile + ResHeader::size();
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the requested frame number's cdtEntry, given the
|
|
|
|
* pointer to the start of the anim file. Drops out if the requested frame
|
|
|
|
* number exceeds the number of frames in this anim.
|
|
|
|
*/
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchCdtEntry(byte *animFile, uint16 frameNo) {
|
2004-11-16 09:15:25 +00:00
|
|
|
#ifdef SWORD2_DEBUG
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
AnimHeader animHead;
|
|
|
|
|
|
|
|
animHead.read(fetchAnimHeader(animFile));
|
|
|
|
|
2003-09-19 13:55:19 +00:00
|
|
|
if (frameNo > animHead->noAnimFrames - 1)
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
error("fetchCdtEntry(animFile,%d) - anim only %d frames", frameNo, animHead.noAnimFrames);
|
2003-07-28 01:44:38 +00:00
|
|
|
#endif
|
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return fetchAnimHeader(animFile) + AnimHeader::size() + frameNo * CdtEntry::size();
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-07 03:18:27 +00:00
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the requested frame number's header, given the pointer
|
|
|
|
* to the start of the anim file. Drops out if the requested frame number
|
|
|
|
* exceeds the number of frames in this anim
|
|
|
|
*/
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchFrameHeader(byte *animFile, uint16 frameNo) {
|
2003-07-28 01:44:38 +00:00
|
|
|
// required address = (address of the start of the anim header) + frameOffset
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
CdtEntry cdt;
|
|
|
|
|
|
|
|
cdt.read(fetchCdtEntry(animFile, frameNo));
|
|
|
|
|
|
|
|
return animFile + ResHeader::size() + cdt.frameOffset;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2003-11-03 07:47:42 +00:00
|
|
|
/**
|
|
|
|
* Returns a pointer to the requested parallax layer data.
|
|
|
|
*/
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchBackgroundParallaxLayer(byte *screenFile, int layer) {
|
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
|
|
|
assert(mscreenHeader.bg_parallax[layer]);
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.bg_parallax[layer];
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchBackgroundLayer(byte *screenFile) {
|
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
|
|
|
assert(mscreenHeader.screen);
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.screen + ScreenHeader::size();
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
byte *Sword2Engine::fetchForegroundParallaxLayer(byte *screenFile, int layer) {
|
|
|
|
MultiScreenHeader mscreenHeader;
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
mscreenHeader.read(screenFile + ResHeader::size());
|
|
|
|
assert(mscreenHeader.fg_parallax[layer]);
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return screenFile + ResHeader::size() + mscreenHeader.fg_parallax[layer];
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2004-04-23 07:02:11 +00:00
|
|
|
byte *Sword2Engine::fetchTextLine(byte *file, uint32 text_line) {
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
TextHeader text_header;
|
|
|
|
static byte errorLine[128];
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
text_header.read(file + ResHeader::size());
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
if (text_line >= text_header.noOfLines) {
|
|
|
|
sprintf((char *)errorLine, "xxMissing line %d of %s (only 0..%d)", text_line, _resman->fetchName(file), text_header.noOfLines - 1);
|
2003-09-19 13:55:19 +00:00
|
|
|
|
|
|
|
// first 2 chars are NULL so that actor-number comes out as '0'
|
|
|
|
errorLine[0] = 0;
|
|
|
|
errorLine[1] = 0;
|
|
|
|
return errorLine;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
// The "number of lines" field is followed by a lookup table
|
2003-07-28 01:44:38 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return file + READ_LE_UINT32(file + ResHeader::size() + 4 + 4 * text_line);
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2003-10-18 08:11:50 +00:00
|
|
|
// Used for testing text & speech (see fnISpeak in speech.cpp)
|
2003-09-19 13:55:19 +00:00
|
|
|
|
2004-04-23 07:02:11 +00:00
|
|
|
bool Sword2Engine::checkTextLine(byte *file, uint32 text_line) {
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
TextHeader text_header;
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
text_header.read(file + ResHeader::size());
|
2003-09-19 13:55:19 +00:00
|
|
|
|
Applied my own patch #1341495, in an attempt to fix alignment issues
reported by Crilith.
To elaborate a bit, the engine no longer accesses resource data through
packed structs. Instead it uses memory streams and the READ/WRITE
functions.
If data is mainly read, not written, I have replaced the old struct with a
new one with a read() function to read the whole thing from memory into the
struct's variables, and a write() function to dump the struct's variables
to memory. In fact, most of these write() functions remain unused.
If data is both read and written, I have replaced the struct with a class
with individual get/set functions to replace the old variables. This
manipulates memory directly.
Since I'm fairly sure that these structs are frequently stored as local
variables for a script, all script variables (both local and global) are
stored as little-endian and accessed through the READ/WRITE functions,
rather than being treated as arrays of 32-bit integers.
On a positive note, the functions for doing endian conversion of resources
and save games have been removed, and some general cleanups have been made
to assist in the rewrite.
Initial reports indicate that this patch indeed fixes alignment issues, and
that I have not - surprisingly - broken the game on big-endian platforms.
At least not in any immediately obvious way. And there's still plenty of
time to fix regressions before 0.9.0, too.
svn-id: r19366
2005-10-29 21:24:54 +00:00
|
|
|
return text_line < text_header.noOfLines;
|
2003-07-28 01:44:38 +00:00
|
|
|
}
|
2003-10-04 00:52:27 +00:00
|
|
|
|
|
|
|
} // End of namespace Sword2
|