Commit graph

12782 commits

Author SHA1 Message Date
Christoph Oelckers
9cdaaff42b - migrated EGS and deletesprite to new actor interface.
Inline wrappers for old code exist.
The backupplayer inline was moved into player.cpp because that's the only place it gets used.
2020-10-21 19:14:41 +02:00
Christoph Oelckers
b8b79b6fa9 - renamed a few things. 2020-10-21 18:42:47 +02:00
Christoph Oelckers
b1f2475230 - serialize null pointers correctly. 2020-10-21 16:31:18 +02:00
Christoph Oelckers
6f4a0c94e8 - something on player is now a pointer as well 2020-10-21 16:31:18 +02:00
Christoph Oelckers
72329eb01b - wackedbyactor, too. 2020-10-21 16:31:18 +02:00
Christoph Oelckers
e94274a543 - holoduke_on is now a pointer as well. 2020-10-21 16:31:18 +02:00
Christoph Oelckers
d7604a3160 - made on_crane a pointer as well. 2020-10-21 16:31:17 +02:00
Christoph Oelckers
7c88de0b42 - made player.actorsqu a pointer and added the needed infrastructure for that. 2020-10-21 16:31:17 +02:00
Christoph Oelckers
39c154dc15 - fixed compile errors 2020-10-21 16:31:17 +02:00
Christoph Oelckers
8aca0774dc - adding more helper wrappers. 2020-10-21 16:31:17 +02:00
Christoph Oelckers
0f2c19522e - defined the iterators. 2020-10-21 16:31:17 +02:00
Christoph Oelckers
78afb67c7f - setup for Duke actor data transition to something more scripting friendly.
The main problem here is that there's two data arrays representing an actor - sprite and hittype and the engine only uses indices for reference.
By setting up hittype to contain a sprite reference, the function and iterator interface can be rewritten to use a single pointer instead to represent an actor.
The main objective is to reduce the number of accesses to the global arrays which constitute the biggest refactoring blocker.
2020-10-21 16:31:17 +02:00
Christoph Oelckers
05c095e448 - fixed parameter ordering for WW2GI's change weapon event call.
This happens when variables are ambiguously named... :?
2020-10-21 07:31:21 +02:00
Christoph Oelckers
99e3eeb19e - split up shoot_r as well and consolidated both blood splat functions. 2020-10-21 01:03:38 +02:00
Christoph Oelckers
4d1e229734 - split shoot_d into smaller, easier to manage chunks.
1100 lines for a single function is too much.
2020-10-21 00:17:09 +02:00
Christoph Oelckers
699c32ff4a - fixed quicksave setup
Fixes #133
2020-10-20 23:34:30 +02:00
Christoph Oelckers
a17f529c49 - fixed freezer projectile not moving.
These endless fallthrough switch/case blocks are a genuine menace. :(
Fixes #136
2020-10-20 23:21:38 +02:00
Christoph Oelckers
3004185646 - fixed bossguy. 2020-10-19 23:58:30 +02:00
Mitchell Richters
a00da23654 - searchpaths.cpp: When filtering duplicate games to delete, if CRC matches and CRC == 0, test whether name matches as well.
* Cryptic Passage for Blood has no CRC and therefore was failing this catch.
* Fixes #130
2020-10-19 20:46:22 +11:00
Christoph Oelckers
66c097b3b8 - fixed bad Holoduke check in alterang. 2020-10-18 21:54:30 +02:00
Mitchell Richters
ff9c7bf0bb - binaryangle.h: Create osum() for PlayerHorizon and PlayerAngle structs and use within each struct's interpolatedsum() method.
* Already doubled up by calculating `sum()` again within `interpolatedsum()` and no point changing that out without an `osum()`.
2020-10-18 22:43:42 +11:00
Mitchell Richters
cd683e6511 - binaryangle.h: Add missed line from 4e07942cff to calculate interpolation ratio using multiply instead of divide. Hopefully in the future we can just use a true double between 0 and 1 like GZDoom... 2020-10-18 22:05:21 +11:00
Mitchell Richters
4e07942cff - binaryangle.h: Minor clean up of PlayerHorizon/PlayerAngle structs.
* Sync PlayerHorizon's `settarget()` with changes to PlayerAngle's from 19d9e1a947.
* For each struct's `addadjustment()`, multiply value by unit there rather than always within `processhelpers()`.
* Properly use `xs_CRoundToUInt()` within PlayerAngle's `processhelpers()` method instead of the signed version.
* Use `abs()` within PlayerAngle's `processhelpers()` like PlayerHorizon's.
2020-10-18 22:00:39 +11:00
Mitchell Richters
19d9e1a947 - binaryangle.h: Fix issues with PlayerAngle.settarget() following changes in bf5bd01164.
* Just was not working whatsoever, really...
2020-10-18 21:31:36 +11:00
Mitchell Richters
666fe71201 - SW: Repair issues with sector object sounds following a506b45369.
* Fixes #132
2020-10-18 21:23:41 +11:00
Christoph Oelckers
a8eaf3575f - warning fixed and constants used. 2020-10-18 12:18:19 +02:00
Christoph Oelckers
c2828fe2e3 fixed undefined end of list behavior of iterators
Also fixed a few warnings
2020-10-18 12:14:05 +02:00
Christoph Oelckers
f4c79161ad - fixed crash when opening the first gate in RRRA's E1L1.
There was some poor coding at play here that fell victim to changed execution order.
Changed everything so that problem cases always get dealt with properly.
2020-10-18 12:10:46 +02:00
Christoph Oelckers
4d3135f4b8 - cleanup and consolidation of moveweapons_r. 2020-10-18 10:42:26 +02:00
Christoph Oelckers
08e29396e5 - split moveweapons_d further up into more smaller parts. 2020-10-18 09:45:41 +02:00
Christoph Oelckers
2b79b29fef - further splitting of moveweapons_d.
Progress with this but still not enough yet to break it down into manageable parts.
2020-10-18 08:38:29 +02:00
Christoph Oelckers
94e8213caf - split fireball handling out of moveweapons_d to reduce the function's size. 2020-10-17 19:53:04 +02:00
Christoph Oelckers
bc0a55de41 - backend update. 2020-10-17 16:01:31 +02:00
Mitchell Richters
a9ad043c36 - Partially resolve issues loading autoexec.cfg files.
* This works for Raze's running directory, but is not pulling in autoexec.cfg files from subdirectories from Raze's running directory (Duke, Duke.Duke) or paths where game content resides (Steam, for instance).
* Partially addresses #110
2020-10-17 12:58:00 +11:00
Mitchell Richters
3582717c43 - Duke: Fix issues in recordoldspritepos() following changes in 7043092fd0.
* Fixes #129.
2020-10-16 14:17:19 +11:00
Mitchell Richters
6385539144 - Port over i_timescale CVAR from GZDoom for use with debugging.
* Had situations where this would have been useful instead of manipulating GameTicRate and rebuilding.
2020-10-16 10:31:11 +11:00
Christoph Oelckers
197a5f642a - properly delete backing images in the texture manager.
Although the images themselves are automatically taken down, any memory they allocate is not - and some of the special tile classes do need dynamic storage.
2020-10-15 20:22:41 +02:00
Christoph Oelckers
4f7c0b696c - and finally the iterators in the backend. 2020-10-15 20:22:40 +02:00
Christoph Oelckers
db4f0d041e - all iterators in Exhumed handled. 2020-10-15 20:22:40 +02:00
Christoph Oelckers
a8e7d1ef74 - the rest for SW. 2020-10-15 20:22:40 +02:00
Christoph Oelckers
6eb6938bbe - another batch in SW. 2020-10-15 20:22:40 +02:00
Christoph Oelckers
727e2ceb29 - iterator variable cleanup. 2020-10-15 20:22:39 +02:00
Christoph Oelckers
679c394b29 - the second third of SW. 2020-10-15 20:22:39 +02:00
Christoph Oelckers
a506b45369 - one third of SW's iterators. 2020-10-15 20:22:39 +02:00
Christoph Oelckers
eb78b88598 - all Blood stat iterators ported. 2020-10-15 20:22:39 +02:00
Christoph Oelckers
33164694b9 - use iterator class for all stat loops in Blood's actor.cpp. 2020-10-15 20:22:38 +02:00
Christoph Oelckers
0a0a995cbd - some cleanup of build.h 2020-10-15 20:22:38 +02:00
Christoph Oelckers
be3292d29b - removed the mostly unused macros for sprite iteration. 2020-10-15 20:22:38 +02:00
Christoph Oelckers
aa5c7c3e34 - replaced all sector sprite iterators in Blood. 2020-10-15 20:22:38 +02:00
Mitchell Richters
b1536d8a0b - Duke (RRRA): Fix potential issue originating from 1bc51a7367 where horizon might have needed adjustment to 0 but wouldn't have done so.
* Previously used to determine whether to apply a horizon adjustment on whether local variable `horiz` was 0 or not (since baseline horizon was 100).
* For no particular reason, initalise `horiz` as FRACUNIT and process if it is not equal to one.
2020-10-15 20:33:28 +11:00