zdoom/gzdoom-gles
0
0
Fork 0
mirror of https://github.com/ZDoom/gzdoom-gles.git synced 2024-11-24 21:21:04 +00:00
Code Issues Projects Releases Packages Wiki Activity
7306 commits 6 branches 6 tags 107 MiB
Commit graph

7306 commits

This branch
This branch
All branches
Author SHA1 Message Date
Leonard2
eeff17c550 Account for this on the renderer's side instead 
This fixes the weapon's layer behaving differently when called from
A_Overlay (the flag would be set in this case breaking the offsets)
 2016-06-03 00:50:12 +02:00
Leonard2
543414d31f Added 2 new layer flags: PSPF_CVARFAST and PSPF_POWDOUBLE 
These flags allowed to easily restore a lost part of the targeter layers
behavior
 2016-06-03 00:50:11 +02:00
Leonard2
bc334ccc9f Avoid creating the targeter layers if the inventory item isn't attached to 
its owner yet
 2016-06-03 00:50:11 +02:00
Leonard2
e1f139ddcd The targeter layers now use a proper Caller 
This will avoid having to check for certain stuff and also allow them to not be destroyed when the ReadyWeapon is null like before
 2016-06-03 00:50:10 +02:00
Leonard2
8244d2c844 Make sure never to call GetPSprite when the ReadyWeapon is null 2016-06-03 00:37:55 +02:00
Leonard2
9ee5e57340 Call GetPSprite here to update the ReadyWeapon but avoid doing so if it 
is null

TickPSprites will take care of destroying the weapon/flash layers if the
ReadyWeapon is null
 2016-06-03 00:37:54 +02:00
Leonard2
f14a840777 The ReadyWeapon is now assumed to not be null 2016-06-03 00:37:54 +02:00
Leonard2
8668719bbc Revert "- re-added P_SetPSrite." 
This reverts commit e6d89b9f71.
 2016-06-02 20:02:03 +02:00
Leonard2
d933f77bf5 Fixed: the game could crash if some inventory items were destroyed in a 
certain way
 2016-06-02 20:00:22 +02:00
Christoph Oelckers
9cf1d96698 - added identification for delaweare.wad to the list of supported IWADs. 2016-06-02 12:04:35 +02:00
Christoph Oelckers
2afadb0108 - fixed: ADehackedPickup was missing NULL pointer checks in nearly all methods that used RealPickup. 
Some DECORATE hacks make it possible that this does not contain a valid pointer when these methods are called.
 2016-06-01 23:41:34 +02:00
Christoph Oelckers
26a15d0ccc - delete thinkers before starting to restore the savegame data to avoid interference of the old thinkers' Destroy() method with the newly loaded data. 
Note that even with this change it is still not possible to unarchive any thinker pointers before the thinker list has been loaded as it would create broken lists.
 2016-06-01 11:43:56 +02:00
Christoph Oelckers
978e522fb4 - fixed: portal-related sound calculations offset the sound in the wrong direction. 2016-06-01 11:14:25 +02:00
Christoph Oelckers
14e4a64c82 - fixed: Destroying a SkyViewpoint disabled all sector portals. 2016-06-01 10:10:18 +02:00
alexey.lysiuk
5348a15b75 Added option to control dynamic loading of OpenAL 
Dynamic loading is enabled by default, set DYN_OPENAL to OFF to link with static or dynamic library

# Conflicts:
#	src/sound/oalsound.cpp
#	src/sound/oalsound.h
 2016-05-29 14:25:04 +02:00
Edoardo Prezioso
243030046a - Fixed issues with invalid GetMaxInventory calls. 2016-05-29 12:50:37 +02:00
Christoph Oelckers
44f09edf4c - fixed: monsters cannot crouch, so their crouch factor has to be 1, as 0 would yield incorrect results if the value is used in subsequent calculations (e.g. adjusting a trajectory to a target which can be or may not be a player.) 2016-05-29 12:48:27 +02:00
MajorCooke
bb91723174 - Added GetCrouchFactor(ptr). 
Gets the crouch factor of a player. Can be set to target/master/tracer, as long as it's a player. Defaults to the first player.
 2016-05-29 12:43:46 +02:00
MajorCooke
2719905ade - Added source and inflictor parameters to all A_Damage/Kill functions. 
- Source is the actor to blame for the cause of damage (monster infighting for example). For missiles, modders should consider setting to AAPTR_TARGET.
- Inflictor is the actor doing the damage itself. Note that by changing this, it will take into account the flags on the pointed actor.
 2016-05-29 12:40:17 +02:00
MajorCooke
33ddbe8ab3 Added SKYEXPLODE support for P_LineAttack on horizons. 2016-05-29 12:38:37 +02:00
Benjamin Moir
6633e41cca added CheckProximity to ACS 2016-05-29 12:38:37 +02:00
Roadcrosser
334962da2c Corrected death message not being gender neutral. 2016-05-29 12:38:36 +02:00
alexey.lysiuk
89016020a8 Used ISO Latin 1 encoding for title text in startup window on OS X 
Windows version uses ANSI_CHARSET to handle the corresponding text
This solves the problem like in https://github.com/alexey-lysiuk/gzdoom/issues/63:
*** Assertion failure in -[NSTextFieldCell _objectValue:forString:errorDescription:], /Library/Caches/com.apple.xbs/Sources/AppKit/AppKit-1404.47/AppKit.subproj/NSCell.m:1684
Invalid parameter not satisfying: aString != nil
 2016-05-29 10:58:12 +03:00
Christoph Oelckers
fc25a74a03 - fixed: In case of a crash the buffered console output was not dumped to the rich edit control so it never got into the crash log. 2016-05-28 20:11:19 +02:00
Leonard2
360ad7a844 Replaced every remaining instances of the old layer names 2016-05-28 01:19:42 +02:00
Leonard2
2f5ae3b51e Changed the default layer indices and renamed them 
Note that this doesn't compile yet
 2016-05-28 01:19:41 +02:00
Leonard2
c82620129c Added A_OverlayFlags 
Allows psprites to follow the weapon and/or the player's bobbing
 2016-05-28 01:19:40 +02:00
Leonard2
8f360f3bea Small refactor 2016-05-27 23:37:35 +02:00
Leonard2
b220db4ebe Fixed: a layer's movements weren't interpolated in some cases 
The old position members would be incorrectly updated if a layer sets another layer's offsets and this layer ticks before the other
 2016-05-27 23:37:34 +02:00
Leonard2
bca9829950 Fixed: a few members weren't properly initialized 
This could go wrong if a layer had only one state of -1 tics
 2016-05-27 22:58:27 +02:00
Leonard2
599fa7cfd1 Fixed: it was no longer possible to set the flash state in the very first 
weapon's state

This was due to an oversight in my first commit
 2016-05-27 22:31:18 +02:00
Christoph Oelckers
d43ea33031 - fixed: A_CustomBulletAttack may not rely on the parameter default setter to get a valid puff. This must be checked explicitly in the functions body to ensure it is never NULL. 2016-05-27 11:39:44 +02:00
Christoph Oelckers
b926cf35d5 - fixed: The portal init code should only delete REJECT if there's actually some portals to consider. 2016-05-27 09:45:17 +02:00
Christoph Oelckers
19003a7973 - fixed: The node builder's FindMapBounds function included all vertices generated by the node builder. 
This could cause problems on maps with bogus nodes so it's better to only check vertices that are referenced by a linedef.
 2016-05-27 09:38:33 +02:00
Christoph Oelckers
34c67b9eae - fixed: The NoiseList array was not cleared before use. 2016-05-27 00:00:27 +02:00
Christoph Oelckers
cc685e982b - fixed: interactive line portals were always deactivated, regardless of alignment type. 2016-05-25 11:36:23 +02:00
Christoph Oelckers
f241d9773a - made P_NoiseAlert non-recursive to avoid stack overflow problems in large open-area maps with a high sector count. 2016-05-25 11:30:11 +02:00
Christoph Oelckers
72a3257400 - fixed: DIntermissionScreen::Init did not handle non-existent strings in the string table and crashed on a NULL pointer. 2016-05-25 01:45:55 +02:00
Christoph Oelckers
49234b1fc3 - fixed: UpdateRenderSectorList did some bad calculations for checking portal plane order. 2016-05-24 13:53:29 +02:00
Christoph Oelckers
e6d89b9f71 - re-added P_SetPSprite. 
For debugging purposes it's better to have this as a function.
 2016-05-24 13:05:43 +02:00
Christoph Oelckers
d058820c65 - fixed scroll speed calculation for Heretic. 2016-05-24 10:01:57 +02:00
Christoph Oelckers
d70601471a - fixed the spawn position of Strife's sentinel's attack's trail. 2016-05-24 09:37:01 +02:00
Christoph Oelckers
3bde8824a3 - fixed A_Burst's velocity calculation. 2016-05-24 09:32:30 +02:00
MajorCooke
1b1195df6a - Added limit parameter to A_RadiusGive. 
- The function ends operation if the number of successfully given actors reaches this count.
 2016-05-23 21:11:26 -05:00
alexey.lysiuk
aa6753383d Fixed fullscreen mode on non-default monitor in OS X backend 
Fullscreen window was incorrectly placed on the main screen instead of the current one
Honestly I have no idea what was the reason behind that [NSWindow setFrameOrigin:] call
Apparently it's redundant and moreover it's incorrect for multi-monitor configuration
 2016-05-22 13:22:07 +03:00
Christoph Oelckers
ae31e0ba72 - changed the default key name string because '#' is used as a comment in the config files. 2016-05-22 12:20:46 +02:00
Christoph Oelckers
db944df467 - fixed 3D floor rendering with nearby slopes. 2016-05-22 12:15:14 +02:00
Christoph Oelckers
eed32ee285 - fixed: The friction calculations for terrain based friction had some fixed/float mixups. 2016-05-22 12:11:39 +02:00
Christoph Oelckers
36dfa97802 - fixed: Polyobjects did not save their specialdata pointer. 
- fixed: DMovePoly did not save its m_Speedv member.

# Conflicts:
#	src/version.h
 2016-05-22 01:27:05 +02:00
Christoph Oelckers
f184438032 - fixed: Polyobjects did not save their specialdata pointer. 
- fixed: DMovePoly did not save its m_Speedv member.
 2016-05-22 01:26:13 +02:00