* GE_NIGHTSPULL (for it to chase you once paralooped) - Sun emblem
* GE_NIGHTSITEM (for hidden until paralooped) - Moon emblem
Combined means that you essentially have to loop it twice.
Also, revamped NIGHTSITEM to suck a little less (and have more compatibility with MT_EMBLEM).
<Alam_Squeeze> p_floor.c:2932: warning: 'widthfactor' might be used uninitialized in this function
<Alam_Squeeze> p_floor.c:2932: warning: 'heightfactor' might be used uninitialized in this function
<Alam_Squeeze> m_menu.c:1907: error: initializer element is not constant
<Alam_Squeeze> m_menu.c:1907: error: (near initialization for `OP_ScreenshotOptionsDef.drawroutine')
<Alam_Squeeze> m_menu.c:9353: warning: 'M_DrawScreenshotMenu' defined but not used
<Alam_Squeeze> Makefile:770: recipe for target '../objs/Mingw/Debug/m_menu.o' failed
<Alam_Squeeze> make: *** [../objs/Mingw/Debug/m_menu.o] Error 1
* Search case is now handled via cvar instead of assumed based on system.
* filemenusearch (previously strsystemstr) uses static char[] to prevent stack suffering.
* New cvar for searching from start of string instead of anywhere in it!
* Menu tweaked for the above.
* Reverted slash from pathsep.
* Clean up titlemapinaction so that it isn't true when you're playing
* Don't print "SPEEDING OFF TO [ZONE] [ACT 1]..." on screen if it's a titlemap - keep a nice and pretty black screen
Level specials setup stuff
Some changes to level setup, largely inconsequential for gameplay but maybe helpful for Lua scripting:
* gravity, weather, and some other internal variables are set to their defaults before loading things, so their values from a previous level don't affect mobj spawning potentially
* Tag lists are also initialised before loading things, so that tag-based search functions (such as P_FindSpecialLineFromTag) can be used internally for the maces and particle generator. This should probably extend to Lua as well
* the level header "forcecharacter" no longer recognises "255" as "None" anymore. This is because it no longer takes skin numbers (as of whenever that change was added... version 2.0?), and level headers are auto-cleared when edited anyway.
See merge request !88
* Improved layout of addons menu.
* Disabled input for 5 tics after a console-touching enter key command on the menu, so that weird stuff doesn't happen.
* Added Add-on options.
* cv_addons_option - chooses save location. A little smaller in scope than the weird Default/HOME/SRB2/Custom thing for screenshots - now it's SRB2 Folder and Custom.
* cv_addons_folder - goes with Custom for above.
* cv_addons_md5 - chooses whether to identify files on the addons menu with name comparison only, or with checksum comparison as well (more intensive hence not default).
* cv_addons_showall - chooses whether to show just supported file extensions, or all
* Some minor other refactors.
Bad:
* Screenshot options menu crashes on access for some reason (to do with itemOn not being valid?????) looking into
-DEFLATE-compressed lumps work properly now.
-All "big" lumps are supported now with the exception of WAD maps.
Notes:
-Compiler spits out a shitload of warnings still.
-Individual lump reading clashes with folders of the same name (see TEXTURES lump, and the Textures/ folder).
Signed-off-by: Nevur <apophycens@gmail.com>
New User Experiences - Menus, Closed Captioning, and other non-substantial tweaks
Hello! I'm going to words.
* Menu stuff!
* Scrolling menus when there's lots of options instead of doing lots of nestling!
* Video Options, Server Options, and Screenshot Options have been updated to take advantage of this.
* Headings are prettier in general to match the level platter's headers!
* Expose a bunch of stuff Mystic wanted exposed, without iundating the player!
* Little arrows!
* Better, clearer control menus!
* I know this is probably related to the above, but whatever.
* Very organised with lots of renaming and magical clarifying context headings!
* Killed the analog-only Camera L/R keys in favour of merging them with the regular turn keys in that context!
* Also little arrows!
* Statistics!
* One page instead of two, loses a lot less info than you'd expect!
* The benefit of having them on one page is greater than the loss of being able to individually discern which of your Score/Time/Rings is incomplete.
* ALSO also little arrows!
* Secrets checklist!
* Scrolls vertically!
* Allows for autogeneration of conditions based off conditionset, which allows for names of required levels to be hidden if you haven't played them yet!
* Still allows overriding condition string per traditional checklist.
* This allows us to show every secret's condition without letting it be spoiled or overflowing the everything.
* Also ALSO also little arrows!
* Closed captioning!
* Console command "closedcaptioning on/off".
* Shows stuff in the bottom right hand corner of the screen.
* sfxinfo[sfx_name].caption = "Butt.".
* S_StartMusicCaption(string, tics, optional player) allows for a caption to play without being anchored to a specific sound.
* Color Profile!
* Using only the best in advanced cube-based majyks.
* cv_gamma now goes from -15 to +5!
* cv_saturation - from 0 to 10!
* A new menu for modifying the hue, saturation and gamma of specific corners of the cube (RYGCBM) independently of the master options!
* Screenshots/Video Mode now supports turning on/off circumstantial color profile alterations (cv_screenshot_colorprofile).
* Spring cleaning!
* cv_allcaps is dead.
* Sudden death and old match scoring (25 points for shield damage as the only change??) is dead
* A few things are taken off the menu because they're not that relevant to have.
* cv_competitionboxes now applies to race as well.
* Failed teleports go nuh-uh.
* Question mark boxes don't have doomednums anymore, so they can't be placed in singleplayer and used to get free lives by playing with the tv_extralife etc stuff.
* Other tidbits!
* HUD offsets are tweaked to give one digit less space to score and one digit more space each to time and rings to match up the right edges.
* Emerald tokens show up on the level complete screen and are loaded into a coin slot at the end of the tally.
* Speaking of Emerald tokens, they now give score all the time, and continues instead of 50 rings when you have all Emeralds.
* Moved Race HUD down, and made it go bing bing bing BONG.
* V_DrawCroppedPatch now properly handles topdeltas.
* Minor tweak to GFZ3 eggmobile's pinch to be more lenient.
* Hudnames for skins now replace full stops with katakana dots. We could make it replace a different character instead if we really wanted to, but I figure this matches what is desired.
Check out ```<root>/!LatestSRB2Files/srb2win_branch_deeznux.exe``` to try it out.
See merge request !91
-Expanded folder recognition for PK3s. Some resources are still not loaded from them yet.
-Took a glimpse at how maps are loaded, since the flat recognition is rooted somewhere there; did nothing yet about it though.
-Working towards "generalizing" how new resources are handled. Some var and functionality redundancy is still present.
* Menu shuffling for addons. Shows up on main menu and MP pause menu if you're host/admin (not encouraged in SP).
* The menu yells at you in red text if you're playing, warning of issues.
https://gfycat.com/DizzyEnergeticCygnet
I will make it prettier.
(Also: stristr. Since the Windows (and others?) standard library doesn't have it for some reason, I modified code someone put on StackExchange as a stopgap; once I'm ready to get on IRC for the day, we'll discuss attribution/rewriting.)
* M_StartMessages when file loading goes wrong! (Determined by CONS_Alerts of warning level CONS_WARNING and CONS_ERROR happening after W_LoadWadFile has been called.)
* Now actively tries to keep your place on the menu if files are deleted between calls to preparefilemenu().
* More enums! DIR_ (for if you want to try embed more metadata in the dirmenu strings) and REFRESHDIR_ (for refreshing the menu and handling warnings).
* Handle changing size menu between calls to opendir() better.
* Don't crash on draw/enter attempt if one of the dirmenu's is null.
* Moved the addons menu to OP_MainMenu instead of MainMenu for now so it can be tested in MP without needing to mess with several menus.
* Display the amount of space used for serverinfo_pak's fileneeded on the addons menu, both visually and with a percentage.
* Instead of needing to press backspace, going back is now handled via an "(up arrow) UP..." entry in the table.
* Maximum depth is now #define'd instead of dotted around everywhere.
* Enums!
* Exec for .txt (with confirmation dialog) and .cfg (without)!
* Complaints about folders and etc now mention the folder names.
* A shrinking folder now prevents dir_on[] from going beyond its end.
Basic as fuck UI, and the controls for manipulating it suck, but the basic elements of the feature set I'm looking for have been implemented in some form or another. More at a later time.
MI, be glad I didn't do this in deeznux ;P
Skybox improvements
Some improvements to skyboxes and new features for them:
* *The Thing's Z position now determines the relative z position of a skybox viewpoint from the ground, the Thing's angle no longer affects the Z position at all.* This should also fix objectplace breaking skyboxes after you place a Thing, and allow Z heights to be more easily controlled by custom skybox mobjs set by Lua's `P_SetSkyboxMobj`.
* The skybox viewpoint Thing's "Parameter" value (or the multiple of 4096 added to the Thing type, if you're still using SRB2DB) determines the ID number for a viewpoint or centerpoint, a value between 0 and 15. By default the game uses the viewpoint and centerpoint using ID 0. Note: this ID system does not apply to custom skybox mobjs set by Lua, only to `MT_SKYBOX` mobjs spawned by Things for a map.
* **Linedef type 448** has been added as a linedef executor special for changing the skybox viewpoint and/or centerpoint used, making use of the new ID system mentioned above:
* By default, only the *viewpoint* will be changed. The front x offsets set the ID of the new viewpoint to use; if the number is out of range (numbers accepted are 0 to 15), the game will simply not use any viewpoint at all.
* If the "*Block Enemies*" flag is set, the centerpoint will be changed as well. In this case the front y offsets set the new centerpoint ID to use.
* If the "*Solid Midtexture*" flag is set, the viewpoint will **not** be changed at all. Combine this with Block Enemies to make the linedef type change only the centerpoint if needed (otherwise the linedef will do nothing lol).
* By default, only the player who triggered the linedef will see the skybox being changed. If a player did not trigger the linedef executor directly, no change will be seen by anyone. If the "*Not Climbable*" flag is set, the skybox change will be "global", i.e. it will be seen by all players in the map.
In my folder on the FTP I've made a subfolder called "skybox improvements", and in it are the following files:
* srb2win-skyboximprov.exe - the test exe for this branch I've compiled myself
* skyboxchange-test.wad - an edit of my test map for my skybox change Lua script release on the MB, to test out the new skybox change linedef type.
See merge request !89
-Moved the MD5 check for added files up so it avoids unnecessary work when you mess up and re-add a file.
-Using compression enum for compressed lumps now.
-Vastly improved central directory seeking algorithm, big files are read fine now. Thanks a lot JTE!
-Improved remaining central directory navigation algorithm, we know and expect what data is coming from now on, after all.
-TX_ textures and sounds are replaced, but textures crash the game on mapload, and sounds are simply mute when replaced. Might have to do something with caching, I don't know yet.
Admin ban fix
Okay, THIS time admin bans should work properly. Turns out the relevant code for banning did not properly consider the case of admins doing the banning, at least until now.
Unlike my last attempt at fixing admin bans (!165), this one would require both host and admin to have the fix exe assuming everyone else would be using 2.1.17, so I'm merging to next instead of master.
See merge request !167
Allowteamchange fix
Fixes "allowteamchange no" to actually work properly and NOT desync you from the game in most cases. You can now switch TO spectator freely (as intended it turns out), and switching FROM spectator is prohibited. The "Enter Game" menu option in Match/Tag now displays one of those blue window notices just like with "Change Teams" in CTF/Team Match if you cannot change teams right now. Everything else works as you'd expect.
The de-sync issue was originally reported here https://mb.srb2.org/showthread.php?p=789965#85
See merge request !168
Used instead of pLocalPalette when attempting to determine objective truths, such as "the colours of this gif without color profile modification" and "what indicies should this colormap remap to".
Also, made f_wipe.c's paldiv only get calculated once.
-Removed a redundant boolean related to texture loading in P_AddWadFile.
-Started working on handling PK3s differently, except that I'm not sure about what I'm doing.
I don't know what to do from now on for today, so I'll leave it here for now.
-PK3 files are now properly loaded as files.
-PK3 lumps are still being treated the same way as WAD files. It means they're dependant on markers, if that makes any sense in the context of a ZIP file (it doesn't). I haven't worked out this part yet but I obviously intend to do so.
-I don't know if the lumps' position address is correct, we'll figure out when I fix the thing from above.
-Tweaked compression conditional to a switch-case format, looking towards adding several comrpession algorithms; haven't removed the previous "compressed" boolean yet.
-Added dynamically allocated name strings for lumps; haven't removed the previous name field.
-Added rudimentary PK3 central directory recognition; nothing is actually loaded in the end.
* Serious refactor of - and internal exposure of - what has become S_StartCaption().
* Renaming of a few existing captions.
* The prevention of access to - or writing over - sfx_None's stuff in SOC or Lua.
* The new Lua wrapper function S_StartMusicCaption(string, tics, optional player), which essentially allows custom music captions to be created. (This is best used for stuff like final lap music in SRB2 Kart or bonus time in a thokker-like game, not a comprehensive Now Playing expy. That'd be a different kettle of fish.)
Also, updated all the lock-on stuff to use P_IsLocalPlayer instead of rolling my own.
A less memetic message: On permission of Mystic, rejiggered a few things such that Race uses cv_competitionboxes, allowing the menu to be nicer.
Also:
* disabled the doomednum for ? boxes
* made A_MixUp play sfx_lose if you're in a gametype that can't use it
* Removed the all-teleporters option from cv_competitionboxes. It could just be done by tweaking the random monitor toggles, anyways.
This means setting the skybox view angle is no longer stupid, objectplace no longer breaks skyboxes, and Lua-defined skybox mobjs can actually set a Z position more easily now.
Falling rocks fix
This fixes the rock spawners' rocks not being removed if they rolled down to the ground from a slope rather than bouncing (see https://mb.srb2.org/showthread.php?t=41963)
See merge request !162
This allows tag lists, gravity, weather, and the "CheckFor" vars to be initialised before running P_LoadThings or P_ResetDynamicSlopes, in case they could affect mobj spawning or cause a netgame desync somehow by carrying over the previous level's values
No doubling texture heights plz
If you didn't know, apparently SRB2 was hardcoded to double the height of 64x64 textures created using TX_START/TX_END. This has frustrated some a lot. This branch removes this hack at last (there doesn't appear to be any visual issues resulting from removing it, anyway).
srb2win-nodoubletexheight.exe can be found in my FTP folder.
See merge request !85
First person HUD drown numbers fix
This branch fixes the following issues with the first person HUD when drowning:
* "0" turning into Sonic's drown sprite, since both sprites happen to be named DRWNA0
* SF_MACHINE's special drowning numbers from 3rd person not being used
See merge request !80
Basically in preperation of supporting colourblindness modes I implemented the following link when loading palettes.
http://http.developer.nvidia.com/GPUGems2/gpugems2_chapter24.html
This basically means I can do whatever the hell I want to the colour profile of incoming paletties, and nobody can stop me. Muahahahaha etc.
Also, I added a saturation feature to show off its full potential, converted gamma from a table to a factor of the calculation, tweaked some menus and made the default value of cvars show up on sliders. Because that's how I roll.
MI's unimportant code cleanup
Just removing a bunch of unused variables/function prototypes from the source code and similarly minor stuff, nothing here should change gameplay much (if at all).
See merge request !71
MI minor refactoring
Just moving the locations of some consvar externs from d_netcmd.h to other header files in which I think they should REALLY belong. Ultimately this should not affect gameplay whatsoever, it's just code cleanup.
See merge request !81
* Put brackets around the sound names on the reccomendation of a friend who works with disability in gaming.
* Update a bunch of sound names to better match what they actually do in SRB2.
* Provides a helpful description for the hearing impaired, whether permanent, temporary or situational.
* Consvar "closedcaptioning", with on/off values available.
* Only works if sounds are on. This is frustrating. I will see what I can do to allow it to work with sounds off in future, but for now it's dependent in order to properly accept or decline sounds.
* Thanks to MI for making that wiki page a lot more comprehensive a few days ago.
* Weapon ring is now a seperate sfx_wepfir instead of doubling up on sfx_thok.
* Also, made V_StringWidth work with V_NOSCALESTART.
* Made it possible to switch statistics pages using left and right arrows. Their previous functionality has been given to Page Up and Page Down.
* Swap over header style to the new version for everything.
* Move Master Server into Data Options, since it needs to be exposed somewhere.
* Show resolution on main Video page.
* Optimise by using V_RightAlignedString instead of V_DrawString with V_StringWidth.
* Revamp sliders. (Requires a new patch.dta I'll upload in a little bit in order to be perfect, but does pretty well w/o replacement graphics too.)
* Revamped base video menu and moved all the relevant stuff from the Game Options Menu onto it. Still mulling on where to put the Master Server reference.
doesn't draw off of the sides, and doesn't ignore snapping or widths for reasons that don't make sense
(for instance: the green bar in MI's test script showed *above* the blue one in non-green resolutions in 2.1.17)
Recreated my previous code using floats in case the issue was overflowing that could be mitigated using floating point precision - but still doesn't work. Pulling my hair out.
Command-line skin selecting fix
Fixed http://mb.srb2.org/showthread.php?t=42312.
Also, fixed being able to change your skin colour during credits/cutscenes in singleplayer.
See merge request !158
Revamped Level Select (platter view)
Seriously revamped every level select instance in the game to use a more attractive and easier to navigate platter view, including:
* Game CLEAR! save (Title/Pause menus)
* Secrets Menu (including custom ones)
* Record Attack/NiGHTS Mode
* Server creation (Online/2P)
* MP Pause level select (Online/2P)
As a result, the layout of the last three above menus has been changed to varying degrees of difference.
Also, bonus feature: using level select (or MAP MAPxx without -force) in Co-op multiplayer won't reset your score, will keep any lives you have above the startinglives variable, and will not take away all your emeralds. The -force thing prevents both warping directly to special stages to rack up the emeralds AND ensures there IS a way to start a new game.
Check out <root>/!LatestSRB2Files/srb2win_branch_levelselect.exe with the latest patch.dta to see more.
Also, LF2_WIDEICON lets you do this. https://gfycat.com/MenacingClearAngora
See merge request !68
Suggested default settings
Making a merge request for suggested new default settings, because saying "hey, I think this would be a better default, you guys should try it" in IRC won't actually get people to give you opinions on them ;p
Camera distance is 160 now, height is 25, and speed is 0.3 (previously 128, 20, 0.25, respectively). Mouse sensitivity on both axes is now 12, previously 35. The default move/look/jump/spin keys is Mystic's proposed scheme from the New Player Experiences topic (https://mb.srb2.org/showthread.php?t=42095).
Camera settings also now can save to your config.cfg, and they won't try resetting back to the default at every chance it gets. (~~As a side effect, analog mode won't have higher cam_dist anymore. Poor analog mode, no one cries for you.~~ Now increases cam_dist in analog mode by x1.2!) This makes it so everything that's different can be easily reverted back to the older setting.
Test exe in my folder, <root>/TehRealSalt/srb2win_changed-defaults.exe
See merge request !75
* Getting hit whilst on a swinging mace doesn't send you off in roll/jump state.
* P_GetJumpFlags replaces a bunch of copypasted code.
* Homing attacking a spring sends you directly on its path, rather than rocketing off in the direction you were preiously hurtling in.
Lets modders set how the game should be saved. Likely useless for
vanilla, but helpful for other mods. This was spawned out of selfishness
for SUGOI to make it only save at the hub, but mods like Boss Mayhem,
which has a hidden map and an act 3 (without any other acts) in the
normal SP campaign, and other mods where the default is detrimental
would benefit as well.
0 for default, 1 for always, 2 for never (no constants because bonustype
doesn't have them either). Won't save if the game is modified, if using
a no-save slot, if playing ultimate mode, or if in a special stage, even
if savemode is set to always.
Analog mode's camera is now x1.2 of your actual cam_dist setting,
instead of being strictly 192. "analog on" seems to reset this modifier
every once in a while, but "useranalog on" prevents this. The wiki
itself says that you should only change useranalog though, so it
shouldn't be too big of a deal.
Camera settings: distance from 128 to 160, height from 20 to 32, speed
from 0.25 to 0.3
Mouse sensitivity from 35 to 12
Default control setup is Mystic's proposed scheme from the New Player
Experiences topic (https://mb.srb2.org/showthread.php?t=42095)
Also, camera settings now save to config, and they no longer try to
reset back to the default at every chance it gets (you can die, start a
new game, or exit, all while still keeping your preferred setting)
* Fixed bug with the magnet shield homing attack and charability homing attack not properly updating the player's target/tracer.
* Made flashing of lockon/got flag icons slower.
* Lock on targets bob now, and are used for CA_HOMINGTHOK and SH_ATTRACT as well.
* Flag stuff is now animated like it was designed to be but which was kinda messed up.
* Cork sounds.
* P_SpawnLockOn(player_t player, mobj_t lockon, statenum_t state) for Lua.
* Added homing/firing CA2_GUNSLINGER stuff at egg guard shields.
* Fixed homing stuff wrt egg guard shields and sea egg balloons.
* Fixed attract orb goin' gold when doing a CA_HOMINGTHOK homing ability.
* Fixed positioning of player during homing attack.
* Cleaned up lockon spawn code.
*P_LookForEnemies is now side-effect-less and only provides a pointer to the found mobj
*player-jumping is dead, long live PF_STARTJUMP
*per Mystic's request, CA2_GUNSLINGER has a targeting icon. It also has a more restricted vertical aiming range.
*mobj for this is in the game as requested
*fast teetering animation flag
*general code cleanup
Simple port of something I made for a 2.1 exe mod that Mystic mentioned
needed doing on the 2.2 Priorities topic
(http://mb.srb2.org/showpost.php?p=790613&postcount=4). Adds emblem type
"map", which gives you an emblem upon beating the map it's for. Var sets
more specific conditions; 1 for all emeralds completion, 2 for Ultimate
mode completion, 3 for Perfect Bonus completion. (These can be easily
removed if requested; these were added simply because it was easy to
implement for modders.)
Criticism on the way it's coded and/or how it is implemented is highly
encouraged. Test wad is <root>/TehRealSalt/levelemblems.wad,
pre-compiled exe is <root>/TehRealSalt/levelemblems.exe.
* Swapped usage of player->spinitem to player->revitem for bullet type selection, since spinitem is used coming out of zoom tubes (which are universal).
* Switched the optional boolean for "bullet" in the Lua wrapper for P_LookForEnemies to be false by default.
* Allowed CA2_GUNSLINGER users to target Detons.
* Cleaned up some code here and there.
* CA2_MELEE/CA_TWINSPIN combination abilities (ground hit, somersalt) disabled.
* Enabled movement controls whilst using CA2_MELEE in the air.
* CA2_GUNSLINGER no longer requires charging. In exchange, it involves a cooldown in which you can't move.
* CA2_GUNSLINGER lock-on distance improved, and aiming allowance reduced to compensate.
* Lots of cleanup/gravflip allowances for homing/targeting code.
* Unrelated: Flickies now properly face the direction they're supposed to when getting knocked out of the egg capsule.
* Unrelated: Thok trail fades like spindash trail now.
* Unrelated: Using CA_HOMINGATTACK when nowhere near an enemy uncurls you.
* Mildly hacky proof of concept! Will de-hack later.
* Charge up your ability!
* You can only damage others if it's at full power.
* Fires directly at enemies/bosses when at full power. Does NOT aim at players.
* Fixed a major hack from when I SPR2ised NiGHTS by spawning in a null object.
* Made P_LookForEnemies better in preperation for the future.
* Changed some defaulting characteristics of P_GetMobjSprite2.
* Fixed CA_DOUBLEJUMP potentially counting as a float.
* Killed some extremely small boolean functions and replaced them with #defines.
Transparent patch support in TEXTURES
Adds transparent patch functionality for the TEXTURES format.
Duplicates source-to-cache column drawer functions and adds extra functionality which uses trans tables.
Changes how the column drawer is picked so it's done once per patch instead of per column.
See merge request !72
* Melee now has a proper weightiness to it with a landing frame.
* If you have twinspin too, do a somersalt when running and hitting the melee button.
* If you have melee, go into the melee landing frame when twinspinning into the ground.
* Made it so (player->availabilities & 1 << skinnum) is only true if skins[skinnum] DOES have an availability method, and always false otherwise.
* This means that setting your availabilities to INT32_MAX is no longer the equivalent of having gotten all skins.
* This means we can detect and kick cheat engine script kiddies who try to fool the server that they can use everything.
* This means availabilities of 0x00 is now valid, so make it all F's for an invalid not-in-game.
There's an outstanding issue - you can set forceskin to whatever you want to as the host. This needs to be fixed, but I'm commiting my successes first.
* PF_NIGHTSMODE is now CR_NIGHTSMODE as part of player->powers[pw_carry]. This is because it's mutually exclusive to every other "carry" type.
* PF_SUPERREADY is dead because it literally just checked for all 7 emeralds and 50 rings. That's it. You couldn't even appreciably alter its presence with Lua. That logic has been placed back in P_SuperReady.
Crumble fof
Basic (but kind of WIPish still) support for remote falling of FOFs by linedef execs or Lua. Originally started for Nev3r to use his dark magic on.
Making this MR so people actually remember I was doing something regarding crumbling FOFs, it's been months since I did stuff for this now mind...
* linedef type 446: basic setup is same as the remote shatter one (436), except the FOF of course falls down rather than shatters. By default the FOF respawns
* *No Climb*: the FOF *doesn't* respawn
* *Block Enemies*: respawning ability is determined by the FOF's flags (if it has FF_NORETURN it doesn't return, if it doesn't it does return)
* *Block Enemies + No Climb*: inverted version of above
* EV_StartCrumble for lua, format: `EV_StartCrumble(controlsec, rover, [floating?, [player, [origalpha, [crumblereturn?]]]])`
* *controlsec* is the FOF's control sector
* *rover* is the FOF itself
* (optional) *floating?* does the FOF float on water after crumbling? Defaults to false (NOTE: probably should be set to `rover.flags & FF_FLOATBOB` for best results currently, kind of weird otherwise)
* (optional) *player* is the player that caused the FOF to fall; needed for some effects such as who to award points to if you killed someone =3 Defaults to nil
* (optional) *origalpha* is the FOF's original alpha before crumbling (the thinker for respawning + floating crumbling FOFs tinkers with the alpha for some reason). Defaults to rover.alpha
* (optional) *crumblereturn?* will the FOF respawn afterwards? Defaults to false (NOTE: probably should be set to `not (rover.flags & FF_NORETURN)` for best results currently, kind of weird otherwise)
* the return value of EV_StartCrumble means something I forget offhand now ...it's either true or false though, mind
See MonsterIestyn/crumble-fof on the FTP for a test exe (srb2win-crumblefof.exe), a test map for the linedef, and a test lua script for EV_StartCrumble (which can also be tested in the map).
See merge request !55
Lua more stuff
More new Lua features and fixes:
* Most Lua functions that deal with stuff that exists only in levels now should spout Lua errors instead of crashing the game if you try to use them outside of levels. Likewise, accessing any of the tables that contain level-only stuff (players, sectors, lines, etc etc) spouts Lua errors too outside of levels.
* `userdataType(variable)` now exists: this function simply returns the type of the userdata variable given as a string (e.g. `userdataType(players[0])` returns "player_t", `userdataType(sectors[0])` returns "sector_t"). This also includes "minor" userdata types for array members of other userdata types, such as .powers of player_t variables or .lines of sector_t variables (which would give the strings "player_t.powers" and "sector_t.lines" respectively).
* The Lua hook "MobjMoveBlocked" now exists: functions for this hook are called whenever a mobj attempts to move horizontally but is blocked by a wall or solid mobj (or whatever else can cause P_TryMove to return false, assuming it doesn't remove the mobj). In theory this hook could be very useful for behaviour such as sliding or bouncing off walls without the need of flags like MF_SLIDEME or MF_BOUNCE etc. Format for use is just like most generic mobj hooks: `addHook("MobjMoveBlocked", functionname, MT_OBJECTTYPE)`, where `functionname` is a function that takes a single mobj_t argument.
See MonsterIestyn/lua-more-stuff on the FTP for a test exe (srb2win-lua-more-stuff.exe) and some test scripts for the above changes.
See merge request !67
* Super sprites are now deliniated via an additional S_SUPER lump between S_START and S_END. Above are normal sprites, below are super sprites. Handled internally via FF_SPR2SUPER.
* Sprite2 numbers are now appropriately limited for the data type that stores them.
* SPR2_SPIN is now SPR2_ROLL, SPR2_DASH is now SPR2_SPIN and SPR2_PEEL is now SPR2_DASH. Makes more sense, right?
Add a vertical flip variant for the translucent column drawer.
Translucency is now properly distributed: 0.00 leads to no render at all, 0.1 to TRANS10, ..., 0.9 to TRANS90, 1 to regular column drawer.
(various large invisible blocks used in the level cause rain to make splashes high above the main level, high enough to make ghostly rain splash sprite artifacts appear sometimes in nearby areas)
This is used by Command_Addfile in the MD5 calculation code, so that it can search subfolders properly and allow addfile in netgames to treat them the same way as in SP
* Changed the order of Record/Nights Attack and its level select menu option so that you go from the main menu to the map page to level select, rather than main menu to level select to map page.
* Cleaned up a lot of code.
http://i.imgur.com/jJ3YZUd.png
Will be discussing what to do with the space freed under "server options" the heading on IRC tomorrow.
(MI, if you're looking - the mistake was that I set up an IT_CALL menu line as IT_CVAR by mistake, which of course caused a crash when trying to get ((consvar_t *)M_GameTypeChange)->string. ;P )
Updated the mid-game gametype/level select menu! Just need to combine it with the server creation menu, and then I'll be ready to make a merge request...
This fixes the grid floors in TD's Lava Mountain freezing the game if they go off the bottom of the screen far enough (they have ACWRFL1A as the wall texture, which is a single patch texture with holes)
Still need to figure out the way to sort pressing esc on reselect level sending you back to the title screen. This will need further whittling before I can call this a solved section and move onto MP stuff.
* More comprehensive "pick up where you left off" system.
* Made individual map name selection actually take into account the map name AND the act number if the heading isn't the zone name.
* Named all the new functions/etc "Platter" instead of "NewList", to make maintenence easier and also because it sounds delicious.
* The code, as a whole, is much cleaner.
* Allows for multiple items under the same heading now, with highlights and spacing consistently handled.
* Allows for picking up where you left off between menus now (assuming the same map is available on multiple screens).
* Defined M_CanShowLevelInList in terms of the two compartmentalised functions cleaved from it, in order to ensure consistency and prevent code duplication.
* Fixing an inconsistency with being able to change skin colours when you shouldn't be able to, much like the previous skin change issue that was fixed.
Better papercollision
Diagonal bounding boxes are no longer root 2 larger than they should be. Testwad is <root>/toaster/sawb.wad. Nice and simple.
See merge request !49
Slopewall transfer
As @Nev3r wants. Go along a slope, hit a wall? LAUNCH! Well, I mean. He wants it slightly less weak. I have already made it less weak than P_SlopeLaunch.
Also also I made the trail that goes behind the player when they're rolling take into account vertical momentum as well as horizontal, so that being launched directly up didn't leave you trailless.
Test with <root>/!LatestSRB2Files/srb2win_branch_transfer.exe and <root>/Nev3r/ACZMaster.wad (roll down slope directly in front of you).
See merge request !66
Revamps of several dated linedef type effects
Felt like being productive so made some things I hate suck less.
Linedef type 7 (flat alignment) now works as follows.
* Linedef angle (from v1 to v2) is flat angle.
* Origin of flat is manipulated to match v1's coordinates - unless ML_NOKNUX is flagged, in which case uses frontside x and y offsets as flat offsets instead.
* ML_NOSONIC prevents changing floor.
* ML_NOTAILS prevents changing ceiling.
* Both of the above flags at once prints a warning.
* (THZ and probably a few other maps need updating.)
Linedef type 540 (friction) now works as follows.
* Controlled by x offset instead of length - offset of -100 is maximum iciness, offset of +483(!!!) is the maximum sludginess BUT things are scaled such that +100 is about the maximum sludginess any reasonable human being would want in a level, 0 is ORIG_FRICTION)
* Not reliant on a sector special to function (can be applied solely by tag to in-map sectors or solid FOF control sectors)
* Uses less memory, perform less calculations, and not bug out objects which just happen to use movefactor for something
* PHYSICS CHANGE: Low friction surfaces actively impede your acceleration (and make your animation speeds faster to give off that Looney Tunes out-of-control effect)
* (ACZ and Frozen Hillside needs updating.)
Sector types 1 and 3 in section 3 no longer do anything.
Linedef type 3 (zoom tube) now works as follows.
* Abs of frontside x offset is speed in units of FRACUNIT/8.
* Abs of frontside y offset is waypoint sequence number.
* Effect 4 flag rotates the player to match the waypoint direction (same as before).
* (ERZ needs updating.)
Linedef type 11 (rope hang) now works as follows.
* Abs of frontside x offset is speed in units of FRACUNIT/8.
* Abs of frontside y offset is waypoint sequence number.
* Effect 1 flag makes rope transfer one-way. (same as before)
* No climb flag makes rope static. (same as before)
* (ACZ and maybe ERZ need updating.)
Linedef type 258 (laser) is almost the same as before, but has altered functionality.
* Its flashing is less obnoxious. (56.25% to 68.75% instead of 100% to 0%, at half the speed)
Linedef type 14 (Bustable block parameters) is new, and works as follows.
* Tagged to the control sector of a block that is likely to be busted (via executor or collision, doesn't matter)
* Concatenation of frontside textures is MT_ object type to spawn, defaults to MT_ROCKCRUMBLE1 if not present
* Sound played when being busted is object type's activesound
* Frontside x offset is spacing (in fracunits) of spawned particles, defaults to 32<<FRACBITS
* Frontside y offset is the fuse of spawned particles in tics, defaults to 3*TICRATE, if set to -1 fuse is not set
* Effect 1 flag makes particles "fly out" horizontally and vertically from the center of the FOF
Sector types 7 through 15 in section 3 no longer do anything.
Linedef type 250 (Mario Block) is almost the same as before, but has new functionality and is slightly different in another way.
* No Climb flag turns it into a brick block (set on Custom FOF via presence of FF_SHATTERBOTTOM - busts when hit from the bottom, player hits their head/fist/whatever, no more upwards momentum)
* Effect 1 flag makes it an invisible block (set on Custom FOF via absence of (FF_SOLID|FF_RENDERALL|FF_CUTLEVEL) - intangible from every side except the bottom. Becomes visible and tangible when it's hit once)
* Only updates side textures when moving down or stationary, not whilst moving up - more fidelity to the source material, less processing on the CPU, everyone wins.
* If FOF master linedef has a backside, then: [If there's items in it the FOF's flats are set to that of the backside sector's ceiling, otherwise the floor.] Otherwise, no flat change.
Linedef type 4 (Speed Pad parameters) now works as follows.
* Frontside x offset is speed in fracunits.
* Effect 4 flag doesn't center the player. (same as before)
* Effect 5 flag sends them off in rolling frames.
* Frontside upper texture is sound to play on launch, defaults to sfx_spdpad when not given
Sector type 6 in section 3 no longer does anything.
Object type 757 is changed as follows, and no longer works on its own - see Linedef type 15 for more details.
* Tag via its angle field
* Number of objects to spawn per tic around it via its z field, if zero then just spawn at center
* Is flipped if given MTF_OBJECTFLIP.
* (ERZ and Seraphic Skylands - if we use that map - needs updating)
Linedef type 15 (Fan particle spawner parameters) is new, and works as follows.
* Tag is tag of object(s!)
* Object type set via concatenation of frontside textures, MT_PARTICLE is default
* The length of the linedef is the radius the particle is spawned out (zeroed if z field is 0)
* Frontside x offset is speed upwards
* Frontside y offset is number of degrees to turn each tic (zeroed if z field is 0)
* Frontside floor and ceiling heights are the heights in which the particle is bound through some fun mathematics
A_ParticleSpawn is changed, too.
* Object type is now changed through actor->threshold instead of var1 to support Linedef type 15.
Test executable uploaded at /toaster/srb2win_branch_flatalignment.exe on the ftp.
Test file uploaded at /toaster/flatalignment.wad on the ftp.
Breaks compatibility with 2.1 so put in internal.
See merge request !29
Plane displacement
Basic support for plane displacement, no netplay support as of writing but this is what Nev3r wanted moreorless at least.
Linedef types 66-68 are the plane displacement specials; 66 = move floor only, 67 = move ceiling only, 68 = both
Front sector = control sector
Tag = tag of target sectors to be moved whenever the control sector floor moves
Linedef length = movement factor relative to control sector movement, 256 = 1:1 with control sector. 128 is half as much, 512 is twice as much, etc.
Whenever the control sector floor moves (not ceiling, never ceiling), the selected planes of the tagged sectors all move (mind, they have a tic delay due to how it all works, it's not perfect really). No support for reverse speeds yet sorry.
EDIT: oh, btw, test exe and test map are included in my folder on the FTP (srb2win-plane-displacement.exe and plane-disp-test.wad, respectively)
See merge request !61
* Made the name of P_PrepareSlopeToWallTransfer (now P_GetWallTransferMomz) make sense.
* Bugfixed the case where you couldn't reliably guess the angle of the wall you just hit.
* Seriously reworked a lot of stuff behind the scenes, thanks to Red's comments.
* More consistent behaviour.
* Launching power is now scaled to the slope's angle relative to the wall you hit, so no massive launches when you hit the side of a steep slope.
* The code is a lot messier and the function name doesn't make any sense any more, so I need to clean that up in the morning.
* Need to figure out how to prevent tiny launches that uncurl you when spindashing up against walls.
* Spindashing trails now takes into account vertical momentum as well as horizontal - it was stupid not seeing your spintrail just because you weren't moving very fast horizontally!
No exe uploaded because it's almost 2am.
GFZ3 Unfuckening
Some tweaks to the GFZ3 Eggmobile's behaviour to make it less frustrating for old and new players alike.
* GFZ3 Eggmobile's laser won't change its angle when you move left and right anymore.
* Changed because its ability to move quickly was extremely punishing for new players (see ProJared's youtube video on the matter).
* http://gfycat.com/PassionateUnknownAgama (the twitching on the third laser has been fixed since this was recorded)
* GFZ3 Eggmobile will, when too far away from the ground and moving upwards, slow itself down vertically.
* This was punishing and annoying for both old and new players alike.
* http://gfycat.com/CornyLeftDungbeetle (old values - the typical settling height is slightly lower now)
See merge request !42
No more double-decker Detons
Oversight of a tiny hack when working on the flickies branch meant detons released twin flickies when hitting stuff. As cute as that is, it's also pretty weird.
See merge request !62
This new function returns a string representing the type of userdata the variable given (e.g. mobj_t, player_t), it even works on "minor" userdata types (player_t.powers, skin_t.soundsid, sector_t.lines, etc)
* Value was getting populated even when it was irrelevant. That's fine, just move it after the valueless parsemode switch.
* Oh, it's struppr'd, of course it's going to compare badly with a lowercase string.
* ...oh, strncmp returns zero when it's equal, not true. Right.
<root>/toaster/smilespatch.wad and !LSF test exe updated on the FTP, btw.
* Create a normal skin, but replace the S_SKIN lump with an almost empty one named P_SKIN.
* Provide "name = existing skin name as one of the first line (comments allowed)
* Reset individual spritenames back to zero sprite2s (or one for SPR2_STND to prevent complete removal) via "reset = SPR2_****" statements.
* No support for changing skin properties that are non-visual at the moment, but I don't really feel like it's wise to have those changable with patch alone. (Hello, tons of "modern Sonic" wadfiles which just change vanilla Sonic's ability to CA_HOMINGTHOK!)
* Sprites patch over.
To see an example in action, look at <root>/toaster/smilespatch.wad.
-If the server tries to kick a joiner who is downloading the game state, they will get a timeout instead, because a regular kick would only happen once the game state has been downloaded
-Added a timeout for player ticcmd packets, again to prevent freezes to happen in some cases
-File/game state downloading is now faster, the speed is controlled by the "downloadspeed" cvar, in packets per tic
-The reason is now properly shown when the server refuses connection
-Changed the default values of "nettimeout" to 10 seconds (previously 15) and "maxsend" to 4 MB (previously 1)
-Added a "noticedownload" cvar that displays a message in the server console when someone is downloading a file
CMake: Fix nasm Linux builds
When compiling under Linux with CMake and nasm/yasm enabled, the operation would fail during linking with undefined symbols related to the tmap files.
This commit adds support for passing flags to the assembler and passes ```-DLINUX``` in order to compile.
After this change, binaries are successfully compiled with either nasm or yasm on Linux systems.
Tested on Ubuntu.
See merge request !153
Other notes:
* on second thought I'll keep the hw_clip functions' gld prefixes rather than HWR, not like it matters either way
* despite the extra lag it does fix the issues with translucent walls and such when displayed at different vertical angles, such as with the GFZ1 waterfall
Other notes:
* Renamed all new functions to have HWR_ prefix instead of gld_, for consistency
* HWR_FrustrumSetup and HWR_SphereInFrustum are disabled and require HAVE_SPHEREFRUSTRUM. This is because 1) SRB2CB did not need the code, so presumably neither will we, and 2) there are some OpenGL API functions used there that due to our way of using OpenGL we don't use outside of r_opengl.c, which makes dealing with HWR_FrustrumSetup complicated in theory
* The new clipping functions are not added to OpenGL's "main" rendering code itself just yet, they're just available to use now once hw_clip.h is included
Hardcoded VAda Flickies
Many thanks to MI for his help, even if he has sinful opinions on what the collection of creatures should be called. ;P
* Flickies are now handled via A_FlickySpawn instead of hardcoded in P_KillMobj, so there can be mobjtypes with MF_ENEMY which don't create flickies, or other mechanisms which can much easier.
* Added map header "FlickyList" (aka "AnimalList") parameter, which can either be set to:
* A species (eg: "Rabbit" or "Bluebird", amongst 17 currently supported types in dehacked.c table FLICKYTYPES - including the seed from Sonic CD, which isn't limited to 'soniccd on' in the console now)
* Any valid mobjtype that isn't MT_NULL (eg: "MT_FLICKY_GHOST")
* A comma-seperated list of either of the above, up to 64 entries long (eg: "Cow,MT_FLICKY_SPIDER,Chicken")
* "All" - sets behind-the-scenes stuff to use every 'normal' type of flicky in FLICKYTYPES (a distinction which can be utilised to hide secret level flickies where they wouldn't be appropriate for the main game)
* "Demo" - sets behind-the-scenes stuff to use the five flickies closest to the species used in the game's long history.
* "None" - prevents any flickies from spawning.
"Demo" is functionally the default value if you don't include a FlickyList parameter in the header at all.
Of note, a bunch of functions are now created:
* A_FlickySpawn - spawns flicky.
* A_FlickyAim - aims for area near target, but not directly on them - turns around when hitting wall
* A_FlickyFly - flies/swims around target (calls A_FlickyAim)
* A_FlickySoar - hacky alternate fly (calls A_FlickyAim)
* A_FlickyCoast - slowing down before going off again
* A_FlickyHop - fracunit-scale precision for A_BunnyHop
* A_FlickyFlounder - A_FlickyHop with randomisation
* A_FlickyCheck - State-setter for falling, or being on-ground
* A_FlickyHeightCheck - State-setter for falling, or being below a certain height relative to target
* A_FlickyFlutter - A_FlickyCheck, but with a slow fall/movement (calls A_FlickyCheck and A_FlickyAim)
I don't need to enumerate the object types and states that have been added, do I?
Oh yeah, I also made it so get_mobjtype's failure value was MT_NULL and prohibited SOC from editing the properties of it to compensate.
IN ADDITION: Killed "soniccd" console command, since it made things more complicated and honestly being able to specify Sonic CD seeds in the level header is a better option.
See merge request !60
Savegame hotfixes
Yes, this another netgame related fix! Named this "savegame hotfixes" rather than "ERCZ hotfix" or the like in advance of the fact we probably need a general branch for other $$$.sav fixes?
Anyway, fixes included in by this branch:
* ERCZ's lava falls now act as expected rather than being stuck in place for the map's duration, fixing all other known desynchronisation issues known in the map for the time being (yay)
* Changing sector tags in a map no longer causes joiners to netgames to have their SRB2 application stop responding etc
* Player names should no longer turn red in the console in gametypes outside of CTF/Team Match if a resynch occurs (not $$$.sav related, but it's related to netplay so I lumped it in anyway)
See merge request !150
One Eggscalibur fix
Use mobj->watertop instead of mobj->spawnpoint->z. CEZ3 is far from fixed in multiplayer, but this is still one less thing to worry about.
See merge request !148
Software plane rendering fixes/cleanup
This is mostly cleanup, with the exception that polyobject planes should now behave somewhat better than they do in 2.1 currently (translucent polyobj planes should now always be actually translucent regardless of where you view them from, for instance). I can't claim I've fixed their rendering once and for all though, there may still be bugs for all I know...
See merge request !136
Some texture-related fixes
Bugs fixed in this branch:
* upper/lower/middle textures with non-existent texture ids being capable of crashing the game. For instance, RVZ1 has colormap codes on non-colormap linedefs, which causes them to wind up with invalid texture ids because of how the game tries to interpret lower/upper textures with "#" followed by characters on normal linedefs. Fortunately these "textures" are normally not visible anyway (since they're all in control sectors) unless they are swapped with in-level textures by some crazy Lua script of some sort...
* animated single-patch textures with holes displaying garbage on first viewing (see this thread: https://mb.srb2.org/showthread.php?t=42195)
* the heights of the lighting (shadows or colormapping) from water/translucent/shadowcasting/etc FOFs become messed up when displayed on repeated midtextures.
See merge request !144
Flat changing hotfix
This should fix how Lua scripts that change flats in levels, particularly to flats that don't already exist in the level, cause SRB2 to crash for anyone who joins a netgame afterwards. (I'm not sure if it's a consistent thing or not though)
Probably needs a good test to verify the issue is completely fixed before being merged in. Maybe there are other minor things I didn't account for in this fix? I don't know!
See merge request !141
Dunno who thought it was a good idea to lump tag, nexttag and firsttag together in $$$.sav, but that meant changing sector tags caused SRB2 to stop responding since it lead to the tag lists being broken
This fixes ERCZ's lava falls not moving after joining a netgame at that stage. Possibly explains other various floor/ceiling related netgame quirks that haven't been noticed until now, maybe even in multiplayer levels.
Remove secplanes
This is just removing SSNTails's old port of ZDoom's secplane code and math, from back when he attempted slopes himself. The slopes we've had since 2.1.15 however do not need these, so we can pretty much toss the code for them out now (nothing uses them anyway).
See merge request !149
-Updated packet name list so the debug file no longer shows garbage packet names
-Replaced byte values with actual net command names in the debug file. Only the first net command in a packet will be shown though
-Added a MOBJCONSISTANCY define that makes the game takes all revelant mobjs to be counted in the synch seed
-Added a PACKETDROP define that adds two console commands "drop" and "droprate" to simulate bad internet by dropping packets
-Added/changed comments here in there in the netcode
-Fixed a minor error that would ignore one of the urgent ack slots
-Added a space between the map name and "zone" for the messages shown in a joiner's console
This fixes Brak's electric barrier disappearing for joiners to ERZC. There seems to be some issues with the lava falls there too I've found, but the electric barrier actually stays around now at least
* Set the default flicky group to be equivalent to Flickylist = Demo.
* Refactored a little. (This also removed a crash-causing typo in a Z_Free in P_ClearSingleMapHeaderInfo.)
* MT_EXPLODE's anims partially FF_ANIMATE'd.
* A_FlickySpawn turned into a wrapper for internal function that does what A_FlickySpawn did but returns a pointer to the spawned flicky.
* Added the ability to read in mobjtypes as well as from a short list of flicky constants.
* Added a new state, S_XPLD_FLICKY, which is the new deathstate of all badniks.
* Put the disabled animal-spawn-on-P_KillMobj behind a #define.
* Renamed the branch and associated properties to something not rife with sin.
No idea if this compiles yet (let alone if what is there works at all), just getting this out the way so I can work on certain other things *cough plane displacement cough*
this fixes a crash in (old) GFZ2 at the ramp as a result of creating pv1/pv2. This probably means before pv1/pv2 there could have been some silly typecasting from vertex_t to polyvertex_t to get fixed vertex coords and such...
Spindash dust
Charging a spindash kicks up dust, we all know this feature was dying to get in at some point. Bubble and flame forms of spin dust are included from FSonic, for underwater and elemental respectively.
Oh, and as a bonus I reorganised the spindash/spinning/other ability2 stuff code to look a bit neater and more organised.
New resources:
* MT_SPINDUST - the object
* S_SPINDUST1 to 4 - the normal form's states
* S_SPINDUST_BUBBLE1 to 4 - the bubble form's states
* S_SPINDUST_FIRE1 to 4 - the bubble form's states
* SPR_DUST - the normal form's sprite set (uses frames A to D, just pinch FSonic's sprites really)
* SPR_FPRT - the flame form's sprite set (frame A only)
SF_NOSPINDASHDUST disables spindash dust for a character
See merge request !52
Multi-patch texture support for transparency AND translucency
Fixes the transparent bits of the multi-patch glass texture in THZ1 turning cyan when linedef types 900-908 are applied for translucency
See merge request !56
Zoom tube camera fix
Fixed a mistake where I accidentially allowed people to change the player angle whilst in a zoom tube when previously improving them.
See merge request !57
Smoother ropes and zoom tubes
Makes rope hangs and zoom tubes suck less. Specifically, they handle corners (and vertical height changes, ala sloped rope hangs) a LOT nicer. Ported from internal.
See merge request !139
As LJSonic has pointed out, there's no need for a for loop in either case; just use sector->floorpic/ceilingpic as a levelflats index directly
(Besides, if that was to stop any out-of-bounds indexes being used, that's hardly the way to do it anyway)
*Added FLIPX/FLIPY support for multi-patch textures and single-patch textures without holes
*Added FLIPY support for single-patch textures with holes; I'll sort FLIPX support out later
To re-enable support for the above, uncomment the define HAVE_LUA_SEGS line in lua_script.h. Plain bbox userdata stuff is not disabled (though currently it's not used anyway)
* Cleaner A_ParticleSpawn code.
* Removed several hacks with Mario blocks, and corrected a mistaken assumption with when the FOF's thinker is being applied.
* Made the flashing of T_LaserFlash even less obnoxious.
Turns out sdl12's version of this function only did stuff for DC/GP2X ports; support for them have been cut out for SDL2, so for now let's just not use the function at all
I'm convinced there's going to be some stupid side effects from doing this, but it's the quickest way I can fix the polyobj planes not all appearing anyway
Sprite2 changes
Some stuff!
* Lua access to sprite2.
* Introducing new Lua-exclusive function, P_IsValidSprite2(mo, spr2). Basically just a wrapper for (((skin_t *)mobj->skin)->sprites[spr2].numframes > 0), useful for creating custom sprite2 defaulting functions since hooking into P_GetMobjSprite2 wouldn't be worth it.
* FF_ANIMATE support for sprite2s. The var2 of the state works identically to normal FF_ANIMATE, but var1 is completely disregarded - it just runs all of the frames available to that one sprite2 animation set.
* As a result, a bunch of states which were either not previously animatable or had animated at constant speed now get animation without state changes.
* P_SetMobjState now supports sprite2 defaulting like P_SetPlayerMobjState does.
See merge request !51
Damage control + Match rebalancing
This is two branches in one since while I was working on damage-control's changes months back, I felt it was best Match rebalancing was merged in here too (thanks JTE for helping me do so).
Changes from damage-control:
* `player->health` (formerly the "HUD copy" of `player->mo->health`) is now `player->rings`, and is also now the player's actual ring count.
* `player->mo->health` (formerly rings + 1) is now always 1 when alive, regardless of ring count; if player with rings is damaged, this is untouched.
* P_RingDamage now includes ring spilling code.
* P_ShieldDamage now has a damagetype argument, allowing me to remove the last MT_NULL hack left in from the pre-damagetype days that I forgot about.
* The old "switch-to-seestate" enemy damaging behavior in P_DamageMobj has been removed. This was a Doom left-over and doesn't really affect SRB2's enemies anyway - see, Doom enemies had a random chance of using seestate or painstate, SRB2 enemies always use painstate.
* Other minor efforts to reorganise damaging code and have it make more sense, but otherwise nothing that should affect gameplay in general.
Changes from match-rebalancing:
* New weapon/ammo dropping behavior: if you have the weapon panel + ammo, you drop the panel (but not the ammo); if you don't, you just drop the ammo.
* New Match ammo consumption: Weapon rings can now be fired with no rings at double the ammo cost.
* New emerald behaviour: collecting all 7 emeralds no longer turns you super (read: Match super is dead now) but instead steal points from enemies and gives you and teammates invincibility + sneakers
* Tails ringslinger buff: Any character with CA_FLY will now throw rings 1.5x as fast.
See merge request !28
To that end, I've made MF_MONITOR *without* MF_SHOOTABLE act like MF_MONITOR does, and MF_SHOOTABLE allow for both missiles AND players to destroy. This has had the side effect of allowing MF_ENEMY and MF_BOSS to use MF_MONITOR as a substitute for MF_SHOOTABLE that doesn't allow missiles to cause them damage, which might be useful for Jetty-syns.
* All Lua-originated sprite2 settings are now forced through P_GetMobjSprite2. Makes sense because of SPR2_JUMP, which none of the main characters have sprites for yet all use.
* Cleaned up P_GetMobjSprite2 to not set irrelevant, otherwise-unused variable.
Console improvements
* Unused console backgrounds no longer take up memory
* Input cursor can now move left and right
* Support for selections (Shift-arrow keys, CTRL-A, etc)
* Clipboard support (CTRL-C to copy, CTRL-V to paste, CTRL-X to cut)
* Fixed handling of simultaneous modifier key presses
https://dl.dropboxusercontent.com/u/3518218/21/console.gif
None of this affects chat yet, that's not part of the console. I'll look at it later.
See merge request !133
I added similar checks for the other num* but it seems some MD2s break the other limits without knowing anyway ...so I've commented these checks out for now, unless we have further discussion regarding them later on
* Space sector support complete, bounds of drowning now altered slightly.
* Knuckles climb now has symmetrical slope support for both normal and reverse gravity.
* All slope-determining topheight and bottomheight code is now identical in form.
* Camera postimages now support slopes properly.
Moved the handling of P_PlayerInSpecialSector to P_PlayerAfterThink from P_PlayerThink.
* This allows the player to get hurt on sloped lava surfaces that are moving downwards.
* Also prevents the player from standing on death pits for 1 tic.
* Prevents the player moving 1 extra tic's worth of movement of pain when hit by sector.
* Thankfully, no consequences re conveyors.
* Like, the only consequences I've found have been positive. However, this DOES need to be thoroughly investigated before it can be allowed anywhere near Next itself.
(neither does P_CanRunOnWater, but I don't think it's necessary to make that check more complicated as you probably shouldn't be able to waterrun up steep surfaces anyways)
Also, the "teleport" devmode command can now gracefully handle coordinates specified outside maps with no/few thok barriers, which previously prevented teleport via the thok barrier bleed's sector floor and ceiling being equal.
Fixing various problems with MF_SOLID collision
(I, toast, wrote the original merge request, but Alam made this merge request because I made a mistake. Twice. In a row.)
* Solid objects are now no longer intangible when your z is less than the other object's.
* I originally thought the bug was limited to just solid objects on slopes (hence the branch name), and was wondering about undesired side effects - but looking at this gfy, this was a major bug that slopes made easier to expose.
* Unfixed = http://gfycat.com/BareLimitedCavy
* Fixed = http://gfycat.com/JubilantOffensiveGar
* You now properly lose your momentum when jumping at a solid object.
* The way I fixed this was a hack, but it was originally a much worse hack which resulted in very bizzare physics issues, so...
* Unfixed = http://gfycat.com/ShockingAbsoluteArthropods (also demonstrates the first bug)
* Fixed = http://gfycat.com/EmbellishedCourageousGordonsetter
See merge request !119
Fixed players disappearing when spindashing whilst wearing multi-layer shields
* Before: http://i.imgur.com/gxUFElo.png
* After: http://i.imgur.com/NxsOg94.png
The problem was that MT_OVERLAY's default radius and height were never getting changed from 1*FRACUNIT, and that meant that when you spindashed, the game considered it completely below the surface of the flat you were standing on. Since you're not usually clipped on flats that don't belong to FOFs, we didn't notice this aspect of the issue sooner.
(SORRY FOR THE MULTIPLE MR'S AGAIN I TOTALLY SUCK AT REMEMBERING TO MAKE SURE IT GOES TO NEXT INSTEAD OF MASTER)
See merge request !124
my P_SearchBlockmap_* functions are now a single searchBlockmap function, you can choose between "objects" and "lines" with the first arg to decide what to iterate through. I also rearranged the argument order a bit for easy stack cleanup etc
I'll remove the old stuff later, don't worry, it's disabled for now
Development improvements
An improvement to Objectplace I've wanted to make for a loooong time.
* Object number up/down is now assigned to ringslinger weapon next/prev. SO much nicer to use, especially since most people have that stuff bound to the scroll wheel.
Also, some changes to the devmode overlay:
* Plays nicely with showfps on - see http://imgur.com/a/zSzvm for the various cases (showfps on, devmode 65535, and both).
* Add PF_THOKKED to the flags in devmode DBG_DETAILED (as "TH").
* "Flags:" replaced with "PF:", "Heap used" replaced with "Heap". The latter is a debatable change, but the former isn't - otherwise the line is way too long compared to the rest of the stuff.
See merge request !45
Hardcoded music name switches
Like I said earlier, it's better if all the hardcoded music switches in 2.2 start with an underscore so that it's harder to accidentally overwrite them.
Should be accepted when music.dta is updated.
See merge request !17
Same no non-music lump requirement as music.dta, and is ignored with no error if music_new.dta doesn't exist
(should hopefully make playing around with music easier)
Also, only one shield ability is selectively blocked by being Super now, and that's because Invincibility does it too and therefore I assume that's a match balance thing instead of a keyboard clash.
v.drawScaled negative scales fix
Lua now errors if you try to use negative scales with v.drawScaled, rather than crashing as it does in 2.1.16 currently. Not much else to say here.
See merge request !126
* Bubblewrap shield bounce now no longer allows thokking post-bounce, but still allows bouncing
* plus a bunch of tiny changes to clean up code around the place.
Fixing FF_REVERSEPLATFORM clipping fixes
Forgot the case where == 0.
Tested by @wolfy852 in Top Down, where the bugs this caused were first discovered.
See merge request !121
Even more slope-related fixes
What it says on the tin pretty much:
Currently included fixes:
* Pusher specials applied to FOFs did not properly account for slopes applied to the bottom plane; this probably was just a minor slipup that appeared on adding the support in the first place, I take it? (some waterfall in the last section of SUGOI's Fudge Canyon is the only place I can think of offhand where this bug is noticable)
See merge request !114
* Some shield sounds swapped (can be reverted later).
* Partial implementation of S3K shield abilities.
* The ability to elemental-groundpound onto gold monitors without going straight through them.
* Force shield ability removed because nobody could agree on it, we'll keep it blank until another idea can get through the disagreement juggernaut.
* shields give you 1000 points
* redundant shields don't make you puase
* checkpoints give you 2000 points
* falling down a deathpit is just falling, not bouncing
Treating " as whitespace in TEXTURES (and animdefs too i guess)
SLADE adds this character in its TEXTURES editor and makes SRB2 cry unless the lump is manually edited; this just treats it as whitespace so we don't have to think too hard about it.
Also, two I_Errors now refer to the correct lump name.
See merge request !122
Proper barrel sky distortion
Skies now use proper barrel distortion in both axes, rather than just the x-axis (i.e. just horizontally) as in all previous SRB2 versions. Credit goes to Nev3r for the original change and idea.
Was going to make an option for a non-barrel distorted sky, assuming barrel distortion is the default, but I just can't quite get it to work in a way that is pleasing to the eyes. So I've not bothered with it here for now, unless someone objects to barrel distorted skies for any reason?
See merge request !35
FOF wall slope skewing
FOF walls now can optionally skew with respect to slopes (software mode only currently):
* Upper Unpegged on the CONTROL linedef enables wall skewing
* Lower Unpegged on the IN-LEVEL linedefs ^1 determine which slope to skew with respect to (off = top slope, on = bottom slope)
* If Transfer Line is used however, Lower Unpegged on the control sector's linedefs does the above's job instead
^1 (this is because they already control pegging of individual FOF walls as it is, so this is for convenience and my sanity that they also deal with skewing)
See merge request !39
Fixes to horizontal spring collision
Fixed various issues arising from collision with exclusively horizontal springs. Thamks to @Inuyasha for the heads up! Of note:
* If you hold down your jump button whilst jumping into it, you no longer immediately use your ability.
* Characters with (CA_DOUBLEJUMP && CA2_MULTIABILITY) or CA_FLOAT or CA_HOVER no longer lose track of their jump count.
Also:
* Upped the strength of info.c's red and yellow horizontal springs.
See merge request !41
Horizon lines
Horizon lines for software mode! Place a Linedef type 41 somewhere on a one-sided linedef and it'll do fancy rendering hacks to draw the adjacent floor/ceiling to the "horizon". One small thing to sort out when I remember to do so, but this is good enough for merge already.
See merge request !44
FF_ANIMATE additions: globally synced animations
FF_GLOBALANIM = makes the animation synced to the level's timer, so all objects will display the same frame at the same time
I mean, doesn't that look so much prettier?
There's also some changes to FF_MIDDLESTARTCHANCE (FF_ANIMATE behavior was split due to not acting consistent between that and SPR2 behavior).
See merge request !46
* Object number up/down is now assigned to ringslinger weapon next/prev. SO much nicer to use, especially since most people have that stuff bound to the scroll wheel.
* introducing the new friend, SH_FORCEHP (which is used as a bitmask to get the extra hitpoints of a force-shield user)
* P_DamageMobj now considers the unimplemented shield constants as well as the implemented ones.
* When not pressing any direction, you now go backwards by default - to emphasise that this is primarily for defensive, not offensive, purposes.
* The camera can now handle the player going backwards without them going completely off-screen.
* Fixed some overzealous checks.
* The spinfire ring is now capable of damaging enemies. (god, what a terrible hack this is)
* When ground pounding, you now bounce off the floor a little bit to make the ability less spammable.
* The Dodge Dash
* Allows you to dash - no control, no falling, no key response - for 2 + (number of extra shield hitpoints) tics.
* If you're holding movement keys down, you dash in the direction you're holding - otherwise, you dash directly forward.
* You're spinning (spindash spin, not jump spin) until your dash is over, then your momentum is cut down significantly and you end up in falling frames.
* It may not necessarily be super useful for Sonic, but it helps the other characters.
* http://gfycat.com/BogusFailingFritillarybutterfly
* http://gfycat.com/PoliticalIdealisticBallpython (outdated speed, shows any direction)
Animated sky support
What it says on the tin: skies can be animated textures now. Just set them up as normal animated textures (keeping in mind the starting texture still has to comply with the SKYn/SKYnn/SKYnnn naming format) and hey presto, your sky animates.
See merge request !34
* GFZ3 Eggmobile's laser won't change its angle when you move left and right anymore.
* Changed because its ability to move quickly was extremely punishing for new players (see ProJared's youtube video on the matter).
* http://gfycat.com/PassionateUnknownAgama (the twitching on the third laser has been fixed since this was recorded)
* GFZ3 Eggmobile will, when too far away from the ground and moving upwards, slow itself down vertically.
* This was punishing and annoying for both old and new players alike.
* http://gfycat.com/ShabbyAncientEarthworm (old values - the typical settling height is slightly lower now)
* If you hold down your jump button whilst jumping into it, you no longer immediately use your ability.
* Characters with CA_DOUBLEJUMP and CA2_MULTIABILITY no longer lose track of their jump count.
(the problem was that MT_OVERLAY's default radius and height were never getting changed from 1*FRACUNIT, and that meant that when you spindashed, the game considered it completely below the surface of the flat you were standing on. Since you're not usually clipped on flats that don't belong to FOFs, we didn't notice this issue sooner.)
* Camerascale, shieldscale, height and spinheight are now player attributes which are set to the skin attribute on skin change, not read directly from the skin.
* P_GetPlayerHeight and P_GetPlayerSpinHeight are now macros instead of functions.
* Extra protection against switching to a locked skin.
* flips the sprite ala MFE_VERTICALFLIP except you don't need to flip the direction of gravity for the object just to draw upside down
* stacks properly with reverse gravity
HOWEVER, since these changes to PIT_CheckThing do raise questions about whether there may be unintended side effects here. As a result, I may remake this for internal only if necessary.
* Sets the sortscale of the mobj to that of its tracer.
* Basically, Smiles' tails won't clip through shields thanks to this.
* http://gfycat.com/GraveGlassEwe
* Also has support for chains of MF2_LINKDRAW!
More slope fixes (aka sorry guys I made another quick SUGOI fix)
Another slopes fix branch!
This branch currently includes a fix for:
* Knuckles gliding into a slope while in 2D mode causes him to try to "climb" on air above them, if the original non-sloped height is higher. Unfortunately he still tries to grab onto places that he can't really climb on, but at least now it's on the slope itself and not in mid-air lol. This issue is encountered in SUGOI's Retro Hill Zone, if you want to check for yourselves.
* Bustable FOF-busting code for both players AND pushables not accounting for slopes
* (Unrelated to slopes): Fix FF_SHATTERBOTTOM FOFs acting like THZ goop when stood on; I added this fix as a bonus because I encountered it in a test map of mine for the bustables fix
(Other fixes may or may not appear here in the near future, I haven't decided yet. Don't wait on me to get in any further fixes before merging, that said)
See merge request !110
Last-minute NiGHTS exiting bugfix
One single-line fix concerning the end of NiGHTS maps. Extremely small, literally zero side effects here, the only reason I'm not committing directly to master (aside from ettiquette) is that I don't have the ability to do so.
* The NiGHTS drone had a single tic of visibility when you hit the goal, which is evident stepping frame by frame through http://gfycat.com/ComplicatedComposedAoudad (the contents of which may or may not make it into 2.2). This is no longer the case.
See merge request !100
Also, the particles made via spindashing in shallow water are now located behind the player. This does move the running particles too, but that's okay.
* Fixed bug where being pushed off a platform whilst charging a spindash would leave you in your charging frames instead of your rolling ones when you hit the ground (http://gfycat.com/MassiveThreadbareItalianbrownbear for how it works now, http://gfycat.com/MarvelousEnlightenedAuk is how it used to work)
* Fixed bug where spindashing on top of a bubble spawnpoint led to you being able to move around in spindash frames (no gif since obvious desired behaviour is obvious)
* Spindash animation speeds up the faster you'll shoot off.
* The spin charging mechanism is now scale-independent, and only multiplies by scale when shooting off - less FixedMul calls, and potentially deals with weird quirks of changing scale whilst spindashing that nobody's discovered because there's no place to find that in the main game!
Also:
* Climbing animation defaults to rolling instead of walking, because what.
Skewing direction is decided per in-level wall by the Lower Unpegged flag on in-level linedefs themselves, since they already decide the stuff for FOF wall pegging as it is. That is unless Transfer Line is involved which moves everything to the control sector linedefs instead...
Cutscene switch fix
Yet another fix for SUGOI, lol: apparently going from a final level's post-cutscene straight to a custom credits cutscene gets the latter stuck forever in the first scene (because the game accidentally stops the cutscene responder from realising the custom credits cutscene is running). This branch of course makes a fix for that.
See merge request !108
Wall collision fixes
Fixes included in this branch:
* A fix for a specific crash encountered in a SUGOI map (sound familiar?) caused by bouncing off a wall adjacent to a slope
* A fix for solid midtextures not accounting for Effect 3/Peg Midtexture properly (https://mb.srb2.org/showthread.php?t=41462)
* A fix for solid midtextures not accounting for texture y offsets properly for non-Lower Unpegged/Peg Midtextured midtextures
* A fix for solid midtextures not accounting for "infinite" repeats (Repeat Midtexture + no repeat count set)
* ~~A fix for Effect 4 on Polyobject First Line making that particular linedef's midtexture solid in addition to making planes visible - this is not wanted if you want a polyobject with both visible planes and full intangibility.~~ Apparently they never did this anyway, don't mind me \o/
See merge request !104
Fix non-player objects having busted step-up/down on slopes
Why the fuck did I make this a player conditional in the first place holy shit
See merge request !103
RA menu can be opened if the game is modified, but Start can't be selected. (Obvious reason: so scripts to display info can be used easier)
Fixed the stupid background for messages in the Record Attack menu, it was bugging me
Also, I added more sortscale handling in the places where I forgot it.
I probably need some help with the maths here to get this to work nicely. http://gfycat.com/LimpAgedDowitcher
http://i.imgur.com/UyOKX5u.png <-- this common glitch with crawlas given MF_PAPER (THEY'RE NOT GOOD AT TURNING NEAR EDGES) used to show the behind-crawlas in front of the front-crawlas.
Unfortunately, I've just discovered this issue (which happens with the old version of the sorting code too): http://i.imgur.com/QNjbATB.png but to be fair these crawlas have gotten stuck inside the edges of this platform, so I'm not sure I can do anything about this without cutting off Sonic's feet when he stands on the ground? shrug
* MF_AMBUSH is now MF2_AMBUSH, because it's something you turn on in a map editor, not with a SOC definition.
* Where MF_AMBUSH was is now MF_PAPER.
* MF_PAPER accesses all the stuff I did previously in this branch...
* ...as well as turn on paper-thin collision detection between mobjs, which I've gotten working but isn't perfect but it's still good enough for non-solid objects!!
* recognising that the offsets weren't going to be accurate if you just SWAPPED yscale and yscale2 over 180 degrees
* taking scale into account consistently
also, some optimisations
also, i've sussed out WHAT'S going wrong here - the topleft pixel of the sprite will always be rendered at the height on the screen it would be rendered otherwise, which is causing the waving. now to figure out what to change to get that to move appropriately...
some complicated mathemagic leads to something which... seems CLOSE, but not perfectly accurate, so i think i need to tweak it more
http://gfycat.com/JovialSpitefulAmericancrayfish for current behaviour
I DID make some steps towards re-implementing PIT_CheckThing for solids only in order to replace the hack long-term and hopefully use less CPU, but is currently disabled via #if 0 since I'm not comfortable changing the function signature of P_CheckPosition right now.
If there's any reason to bring the old system back we could make it togglable by one of the linedef flags I suppose. Not that many people would actually use it though, most likely
remove all the places where i've set vis->scalestep to anything other than 0 to see something that LOOKS okay, but doesn't fulfil exactly what i want (that being a sprite that looks exactly like a midtexture)
* Effect 3 (Peg Midtexture) is now accounted for properly, flipping the collision box position to match the actual rendered position of the midtexture
* Fixed incorrect application of y-offsets for non-lower unpegged midtextures collision boxes; +ve always goes up, -ve always goes down!
* Effect 4 now doesn't make midtextures solid for polyobjects at all - this "conflicted" with First Line having both Effect 4 (visible planes) and Effect 3 (intangible) simultaneously, where we kind of expect the first line's wall to not be made solid. This may be less of a problem in future SRB2 versions, but for now solid midtextures for polyobjects are disabled.
. tmthing can be NULL if called from PTR_SlideTraverse, so we should use slidemo instead
This fixes a crash that occurs in yet ANOTHER SUGOI map, involving bouncy walls next to sloped floors/ceilings
To understand: look at AjustSegs(void) in hw_bsp.c. It reallocates the vetex_t pointers for lines as POLYVERTEX_T pointers, and of COURSE things are gonna get wacky when you're casting pointers.
I dunno how resilient the FLOAT_TO_FIXED solution is or whether it'll be netgame compatible (yayyy float precision loss) but it's not like our builds are netgame compatible with themselves
* The NiGHTS drone had a single tic of visibility when you hit the goal, which is evident stepping frame by frame through http://gfycat.com/ComplicatedComposedAoudad (the contents of which may or may not make it into 2.2).
* When completing a NiGHTS stage with a non-zero link, the link could flash up in the final few tics before the fade to black. This just checks for player->exiting to make sure it shouldn't be shown.
* That hacky anti-NiGHTS-deaxisment code I commented out because I thought it was visual only? Reimplemented in a way that is both more and less hacky. It's identical in result to the original code, but takes a roundabout method to get there.
* Sprite references for SUPE, SUPZ and NDRL are removed because they are now unused.
* Helper's flashing conditional is restructured to do less flag swapping.
* The check for super setting FF_FULLBRIGHT is limited to MAXTRANSLATIONS now, and also correctly takes into account MAXSKINCOLORS == SKINCOLOR_SUPERSILVER1.
* NiGHTS collision bounds aren't hardcoded anymore.
* NiGHTS link will never display when leaving stage.
* Slightly tweaked rules for the supercolor setting when doing a NiGHTS transformation, but only meaningful for setting FF_FULLBRIGHT.
* if you can turn SF_SUPER, flash your skin's supercolor, otherwise be your normal color
* if your skin doesn't have a SPR2_NGT0 (horizontal fly), use Sonic's (this will hopefully be replaced by 2.2 with sprites of NiGHTS themselves)
* MT_NIGHTSCHAR made irrelevant, everything follows actor->target instead of actor->target->tracer now
* emerald is now player->mo->tracer instead of player->mo->tracer->target
* nightopian helpers flash for the 35 tics before they disappear
* nights capsule makes boss explosions/noises now (i can change it back i just like it better)
* drill off into the sky instead of fly up in floating pose (but no noise yet)
ALSO:
* default maxdash is now 70
* forgot to add supercolor to lua, it is there now
* SPR2_SMSL renamed to SPR2_SSTN (stun)
* any player with a skincolor that's in the super range is set to FF_FULLBRIGHT at state-set time, so no need to keep super players non-fullbright just because they use spin stuff
OpenGL planes fix
This branch removes a specific hack in the OpenGL code for detecting if a plane is for the floor or ceiling of a sector. This it turns out fixes ceiling slopes in Boinciel's SUGOI map going missing. Though, It doesn't fix that other glitch with one FOF (must be unrelated).
If you want to test these fixes out properly, make sure working in-level sectors, FOFs AND also polyobjects all still work fine, sloped or not.
See merge request !97
Fof slope crash fix
This fixes yet another software renderer crash, this time relating to sloped FOFs: sometimes the renderer thinks parts of FOFs where the top and bottom heights are the same height are actually the bottom going through the top (in other words a negative height). This causes problems when drawing normal walls around such FOFs since the game obviously doesn't expect negative heights anywhere along the FOF - before slopes, the game could just flip around the top and bottom heights automatically with no problems. This branch also should fix crashes for all genuine cases of negative FOF heights when slopes are involved, I suppose.
Hopefully this stops FuriousFox's SUGOI map crashing for the time being, all seemed fine when I tested it out myself.
See merge request !96
* Don't allow arbitrary numbers to be used for a skin's supercolor, only strings.
* Devmode is no fun allowed mode, so turn off the god hyper flash there.
* Several new supercolours.
- SKINCOLOR_SUPERSILVER1-5 (for fun) - "Silver"
- SKINCOLOR_SUPERPERIDOT1-5 (nyeheheh) - "Peridot"
- SKINCOLOR_SUPERCYAN1-5 (for fun) - "Cyan"
- SKINCOLOR_SUPERPURPLE1-5 (for fun) - "Purple"
- SKINCOLOR_SUPERRUST1-5 (mecha/metal sonic) - "Rust"
- SKINCOLOR_SUPERTAN1-5 (shadow/silver the hedgehog) - "Tan"
* SKINCOLOR_SUPER1-5 renamed to SKINCOLOR_SUPERGOLD1-5, one index for darkest is changed - "Gold"
* SKINCOLOR_TSUPER1-5 renamed to SKINCOLOR_SUPERORANGE1-5, ported properly to the new palette - "Orange"
* SKINCOLOR_KSUPER1-5 renamed to SKINCOLOR_SUPERRED1-5, ported properly to the new palette - "Red"
* new S_SKIN attribute - supercolor - uses an entirely different function to get the names (R_GetSuperColorByName instead of R_GetColorByName)
* a fun little secret - typing "god on" in the console whilst super makes the player hyper (visual only, no sparkles - just rainbow flash) - can be removed if no fun is allowed
-warp checking for invalid map names
I've noticed a bunch of new people getting the "Cannot warp to map 0 (out of range)" error when testing their custom maps. On asking what map name they used, it was clear they didn't use a MAPxx name at all. Sadly it was not obvious to them that other kind of map names are not accepted by SRB2, so I thought it best we make that more clear in-game.
SRB2 now gives you the error "Cannot warp to map \[your input\] (invalid map name)" if you used -warp with a param that is neither a MAPxx name or a plain number.
See merge request !99
Fix R_AddSingleSpriteDef's I_Error messages
Whoops, seems I forgot about this little branch. Basically this fixes how for a character's sprites, a full sprite lump name is displayed instead of just its sprite prefix in the "R_AddSingleSpriteDef: No patches found for [sprite prefix] frame [frame character]" message (unlike when it occurs for non-character sprites), resulting in something like "R_AddSingleSpriteDef: No patches found for PLAYC2C8 frame \^" which does NOT help custom character authors at all as it just confuses them instead. (for the record, the problem would actually with the ^ frame and not the ones displayed after "PLAY")
Oh, and the same problem is also fixed for this similar message: "R_AddSingleSpriteDef: Sprite [sprite prefix] frame [frame character] is missing rotations"
See merge request !95
Spr2 freeslots
Needed for Miles, generally a good idea to have around anyway. Considering Miles uses it without problems, seems to work fine.
See merge request !24
child can be 0 or 1 (or "right" and "left", alternatively)
bbox coord can be 0,1,2 or 3 (or "top", "bottom", "left" and "right", alternatively)
Also added some support for bbox userdata taking "top" "bottom" "left" "right" fields. Not that there's any use for non-node bbox userdata just yet...
for some apparent reason the compiler didn't like the while loop condition edit on its own (it complained about inline failures for P_MobjReadyToTrigger for some reason), so I had to add that extra bit above the while loop... and it was happy again, huh
If you want more specifics, sloped FOFs are to blame it turns out: sometimes the bottom of an FOF wall blocking a segment of an in-level wall column can be considered ABOVE the top part of the FOF there (yikes), and then the dc_y* values go offscreen, and then BOOM
* Press spin in midair to make the shield flash solid repeatedly and make a number of ding noises.
* When the player with a flashing, dinging shield hits the ground, they are sent off in spinning form at the maximum of 2*abs(momz) VS the 3D hypotenuse of momx, momy, and momz.
Software crashes fix... fix
Fixes a typo introduced by merge request !75 that caused upper textures to set the wrong ceiling clipping value when not visible, allowing all sorts behind the walls to be visible. This is most noticable in GFZ2 when the inside of the tunnel section is visible
...probably a good idea to make sure this one doesn't introduce any MORE visual glitches by mistake (again, compare with 2.1.15 if possible)
See merge request !93
Fix for flats with transparent pixels on slopes
This fixes how R_DrawTiltedSplat_8 unintentionally allows the cyan pixels to NOT be considered "transparent" if after being remapped, depending on sector brightness and/or linedef type 606 colormaps, the result of remapping is not palette index 247 (the cyan we typically turn transparent). That is, the original colors from the source flat graphic are not checked, but instead the __result__ of coloring the flat under the respective colormap is checked for "transparent" pixels. This is only a problem for the tilted splat drawing function, not the regular one for non-sloped planes with cyan-pixel-using flats.
I found out about this bug from the issues Ritz was having with sloped 255-alpha translucent FOFs using transparent flats and his custom COLORMAP lump (and later when he applied a linedef type 606 colormap to the FOF) for his custom map. Thankfully he has some workarounds, but this should fix the code-side issues that caused his problems in the first place.
I also fixed stuff with another splat drawing function that's not currently used atm (maybe it will be in the future, if splats themselves are ever enabled again? *shrugs*).
See merge request !92
Skybox rendering offset fix for third person/alt view camera
Fixes the issue reported in this thread: https://mb.srb2.org/showthread.php?t=41729
I dunno if this will negatively affect any existing skyboxes in SRB2's own levels, that said. I tried out THZ2 and CEZ1 with this fix at least but I forgot to compare them with how they are in 2.1.15 so _*shrugs_*
See merge request !94
OS X Makefile build setup
This merge request:
* Cleans up the OS X bundle resource location code and fixes a SIGSEGV and memory leak
* Simplifies and fixes the OS X desktop alert code, closing more leaks
* Adds the MACOSX build flag to the Makefiles, to allow building a binary (but not Mac app yet) of SRB2.
This is intended to make it easier for developers to build on Mac OS X, without having to pull in all of XCode. You can keep using CMake if you prefer.
To test, use `make -C src MACOSX=1 NONX86=1 SDL=1 NOASM=1` for a release build.
Left to do:
* Add a content bundling script to be run after building, and a flag to trigger doing that.
`MACOSX_BUNDLE` maybe?
* Somehow get access to a Mac running PowerPC and figure out how to build a multi-platform binary.
* Add the proper magic to compile using gcc if requested. (Right now, compilation is done via LLVM/Clang)
See merge request !72
Gas jets, fan objects, springs and slopes
Whoop whoop whoop minor bugs that only get noticed due to weird experiments whoop whoop whoop
* For gas jets only: the object's standingslope is ALWAYS null. No drifting down the slope for you.
* For gas jets and fan objects: Now checks whether the player's top is below the bottom of the launcher, instead of its bottom. zdist calculation not affected, since it's signed and okay with being negative by about the height of the player.
* For all 3: the player's standingslope is NULL'd so the player isn't launched off at a wacky angle if they're standing on a slope then walk into the thing.
See merge request !91
Remove i_net.c
The code in i_net.c doesn't actually seem to be used in SRB2. I was able to compile a build without it, and hosting and joining netgames worked just fine (well, as fine as they can with the current state of the netcode...).
The vast majority of code in the file seems to be contained in HAVE_SDLNET ifdefs, and I'm pretty sure SRB2 has never used SDLNET in a public build. The only bit not contained in that block is I_InitNetwork(), which just prints an error and returns false.
Do we really need to keep it around? If not, I say get rid of it. It seems like useless clutter that is just going to confuse people who are trying to understand the source code.
See merge request !73
Re-factoring AA tree code from m_misc.c/m_misc.h into its own files
What the title says. The AA tree-related code now lives in the files m_aatree.c and m_aatree.h. Part of why I did this was to solve this m_misc.h/w_wad.h cyclic dependency problem (involving MAX_WADPATH and AA trees themselves) mentioned in the now-removed comments, another reason was ...only OpenGL uses AA trees at all, why include the relevent structs/functions/otherwise anywhere except where is necessary (which is very few files as it turns out)?
Otherwise, it just looked better on its own rather than mixed with all the other stuff already in the m_misc files. Not really important or anything affecting gameplay at all I guess.
See merge request !82
* When moving slowly, P_InstaThrust at S_SKIN's maxdash forward, and set momz to S_SKIN's mindash upwards. Plays a tok noise (not thok). Hurts enemies/bosses, busts spikes/monitors/all types of bustable blocks.
* When moving quickly... doesn't do anything yet, but WILL do a slide.
Also, P_DoSpinDash is now renamed to P_DoSpinAbility, and CA_TWINSPIN users can bust all bustable blocks on collision too.
* CA_TWINSPIN - you can do it multiple times
* CA_DOUBLEJUMP - your jump height is reduced each time, like in Kirby games. The number of extra jumps once you've left the ground that are available is determined by the character's actionspd.
* Electric sparks coming off entire body instead of bubbles coming out mouth
* Different sounds.
* Different icons.
These sprites are currently local only, but I'll be doing a lot of asset updating this evening since Rob asked me to so it won't be long until you can get them.
Sideways springs, horizontal hog-launchers, perpendicular plungers...
Call them what you like, they're in the code now. And have been for months!
Nev3r uses the hell out of these and I'm fed up of them being <!>'s all over the place, so please have 'em in master so we can update srb2.srb and make things better for all of us.
See merge request !30
* FF_MIDDLESTARTCHANCE - has a 50% chance of starting the spr2 or FF_ANIMATE animation halfway in
* FF_SPR2ENDSTATE - if var1 == S_NULL, don't loop, just stop incrementing the frames. Otherwise, go to the state represented by var1.
The former is just something I did for fun, the latter is something that'll come in handy when porting in new-character-moves.
Setup:
* S_SKIN needs an "availability" line. Set to 0 to have always available, otherwise is assumed to be UINT8 unlockable number. If it's a valid unlockable, its name will be copied over with the character's realname.
* Define an unlockable via SOC or hardcode. If you don't want it visible on the Secrets menu, just set its type to None. Everything else is as normal - you can set the conditionset to whatever, objective, height, nochecklist, nocecho...
* radius - sets the player's radius for that skin.
* height - sets the player's normal height for that skin.
* spinheight - sets the player's spinheight for that skin.
* shieldscale - see http://i.imgur.com/BQ5DhKC.png for justification
R_SkinUnlock defines the circumstances under which a skin is available. For simplicty's sake, I've currently bound it to an S_SKIN variable so I can toggle it easily, but it WILL be replaced with a hook into the savegame system at some point.
* Currently has three tiers of unlock - freebie (forceskin or modeattacking via a loaded replay), Ringslinger Only, and SP/Coop and Ringslinger.
* I don't know anything about netcode so I basically decided to make R_SkinUnlock relevant only under local circumstances, try as hard as possible to stop bad skin info from getting sent to the server, and then admit defeat once the server has the information. If this is a bad choice, please discipline me and show me how to fix it.
* Character Select now checks for whether the character is hidden or not on menu load and does/undoes it based on that info, but will never touch one disabled via SOC. I also used this opportunity to optimise, checking for/filling out charsel pictures instead of doing it later. (It now also includes special casing for a select screen with zero characters!)
* Mode Attack now hides hidden characters in its character select based on SP rules.
Things that still need to be done:
* ForceSkin_OnChange. Is there a graceful way to handle this?
* No obvious skin name conflicts. Add a salt to the names of hidden skins, and then remove it when they're unhidden?
* The gap between Knuckles' skin number and the first custom character anybody adds will be way too obvious. A seperate hidden skin numbering system? Start at 32 and count up from there? There's a few ways...
* If a character select character image is not set, don't iterate every tic - iterate on first image get and then save to the struct.
* A character select screen with only two characters now has special case handling.
* A memory leak in the making has been plugged. (specifically, picname not being Z_Free'd if the loop fails to do so)
* Logic/operation simplification.
Also, some typo corrections and clarity case movements of stuff in other files I've been looking at.
* SF_NOJUMPSPIN - Player's height is full whilst jumping, SPR2_JUMP defaults to SPR2_SPNG instead of SPR2_SPIN, and the player goes into fall frames if they start moving downwards or use their ability.
* PA_JUMP - for jumping (upwards in the case of SF_NOJUMPSPIN.
* SF_NOJUMPDAMAGE - Ala rosy.wad, don't damage enemies, etc when jumping into them.
* SF_STOMPDAMAGE - Just for fun. Ala in Mario, always damage enemies when you land on top of them (your gravity reference, not theirs).
* SF_MARIODAMAGE - SF_NOJUMPDAMAGE|SF_STOMPDAMAGE is reasonably accurate to the Mario games, and might as well be surfaced as such.
Also, a minor change:
* Instead of not spawning the revitem if your SPR2_ is SPR2_DASH, don't spawn it if it's set to 0. This requires the player.dta I uploaded a couple days ago to behave as it was previously.
* Don't get stuck in spindash frames if your maxdash is 0, and don't flash rolling frames if you're on goop.
* Characters disabled through SOC are outright removed. No awkward gap in scrolling - no hint they were ever there in the first place.
* Vertical loop - the character select images are visually continuous. No matter where you are in the chain, you'll always see a hint of the character above or below your current selection in the chain.
* Smooth scrolling - Moto and Prime showed me a gfy from back during 2.1 development where it was super smooth. I didn't make it as slow as that one, but the smoothness was easy to add and the reason it was removed previously - gaps in the character select leading to varying speeds - is no longer relevant.
* SPR2_PEEL, SPR2_SPEE.
* S_PLAY_PEEL, S_PLAY_SUPER_PEEL.
* PA_PEEL.
* Dashmode actually starts charging from runspeed instead of the arbitrarily calculated (normalspeed - 5*FRACUNIT). This just made things easier, honestly, and it's 1 FU of difference compared to the current test case.
- Press spin in midair to stomp directly downwards (losing horizontal momentum), creating a flickering fireball around you.
- No bouncing on enemies, item boxes, etc - just go straight through.
- Hurts other players on touch whilst you're stomping.
- Spawns a bunch of flames around you when you hit the ground.
Also:
- Electric shield's ability now uses different sounds, because I'm picky.
