For some reason, the variable `k` is on entityclass and gets mutated in
createentity() and createblock(). Then updateentities() uses it without
checking if it's valid, because either `k` or the size of `entities`
could have changed in the meantime. To fix any potential undefined
behavior, these bounds checks should be added.
The entity getters I'm referring to are entityclass::getscm(),
entityclass::getlineat(), entityclass::getcrewman(), and
entityclass::getcustomcrewman().
Even though the player should always exist, and the player should always
be indice 0, I wouldn't want to make that assumption. I've been wrong
before.
Also, these functions returning 0 lull you into a false sense of
security. If you assume that commands using these functions are fine,
you'll forget about the fact that `i` in those commands could be
potentially anything, given an invalid argument. In fact, it's possible
to index createactivityzone(), flipgravity(), and customposition()
out-of-bounds by setting `i` to anything! Well, WAS possible. I fixed it
so now they can't.
Furthermore, in the game.scmmoveme block in gamelogic(), obj.getplayer()
wasn't even checked, even though it's been checked in all other places.
I only caught it just now because I wanted to bounds-check all usages of
obj.getscm(), too, and that game.scmmove block also used obj.getscm()
without bounds-checking it as well.
When this is done, there is potential for a mistake to occur when
writing out the bounds check, which is eliminated when using the macro
instead. Luckily, this doesn't seem to have happened, but what's even
worse is I hardcoded 400 instead of using SDL_arraysize(ed.level), so if
the size of ed.level the bounds checks would all be wrong, which
wouldn't be good. But that's fixed now, too.
This is because if they are manually written out, they are more likely
to contain mistakes.
In fact, after further review, there are several functions with
incorrect manually-written bounds checks:
* entityclass::entitycollide()
* entityclass::removeentity()
* entityclass::removeblock()
* entityclass::copylinecross()
* entityclass::revertlinecross()
All of those functions forgot to do 'greater than or equal to' instead
of 'greater than' when comparing against the size of the vector. So they
were erroneous. But they are now fixed.
It's better to do INBOUNDS_VEC(i, obj.entities) instead of 'i > -1'.
'i > -1' is used in cases like obj.getplayer(), which COULD return a
sentinel value of -1 and so correct code will have to check that value.
However, I am now of the opinion that INBOUNDS_VEC() should be used and
isn't unnecessary.
Consider the case of the face() script command: it's not enough to check
i > -1, you should read the routine carefully. Because if you look
closely, you'll see that it's not guaranteed that 'i' will be initialized
at all in that command. Indeed, if you call face() with invalid
arguments, it won't be. And so, 'i' could be something like 215, and
that would index out-of-bounds, and that wouldn't be good. Therefore,
it's better to have the full bounds check instead of checking only one
bounds. Many commands are like this, after some searching I can also
name position(), changemood(), changetile(), changegravity(), etc.
It also makes the code more explicit. Now you don't have to wonder what
-1 means or why it's being checked, you can just read the 'INBOUNDS' and
go "oh, that checks if it's actually inbounds or not".
Since there's an INBOUNDS_ARR() macro, it's much better to specify the
macro for the vector is a macro for the vector, to avoid confusion.
All usages of this macro have been renamed accordingly.
Stuck prevention (pushing the player/supercrewmate out if they are
inside a wall) has been factored out into its own function, so it's no
longer copy-pasted but slightly tweaked just for the supercrewmate.
Instead of having two separate functions to move entities along vertical
moving platforms, one for the player and one for the supercrewmate, they
have been consolidated into one function.
So, I was staring at VVVVVV code one day, as I usually do, and I noticed
that warp lines had this curious code in entityclass::updateentities()
that set their statedelay to 2, and I thought, hm, maybe the pre-2.1
warp line bypass is caused by this statedelay. And, it doesn't seem like
this is the primary code used to detect if the player collides with warp
lines, the actual code is commented with "Rewritten system for mobile
update" and bolted-on in gamelogic() instead of properly being in
entityclass::entitycollisioncheck().
So, after getting tripped up on the misleading indentation of that
"Rewritten system" block, I removed the rewritten system, re-added
collision detection for rule 7 (horizontal warp lines), and after
checking the resulting behavior, it appears to be nearly identical to
that of warp lines in 2.0.
You see, if you use warp lines to flip up from the top of the screen
onto the bottom of the screen, close to the edge of the bottom of the
screen, Viridian's head will display on the top of the screen in 2.0. In
2.1 and later, this doesn't happen, confirming that my theory is
correct. I also performed warp line bypass multiple times in 2.0 and
with my restored code, and it is pretty much the exact same behavior.
So now, the pre-2.1 warp line bypass glitch has been re-enabled in
glitchrunner mode.
I noticed that if I have a large amount of entities in the room, the
game starts to freeze and one frame would take a very long time. I
identified an obvious cause of this, which is that the entity collision
checking in entityclass::entitycollisioncheck() is O(n²), n being the
number of entities in the room.
But it doesn't need to be O(n²). The only entities you need to check
against all other entities are the player and the supercrewmate. You
don't need to "test entity to entity" if 99% of the pairs of entities
you're checking don't involve the player or supercrewmate.
How do we make it O(n)? Well, just hoist the rule 0 and type 14 checks
out of the inner for-loop. That way, the inner for-loop won't be
unconditionally ran, meaning that in most cases it will always be O(n).
However, if you start having large amounts of duplicate player or
supercrewmate entities (I don't know why you would), it would start
approaching O(n²), but I feel like that's fair in that case. But most of
the time, it will be O(n).
So that's how collision checking is now O(n) instead.
This makes it so the fix for crewmates' drawframes being wrong for
1-frame is fixed for all crewmates regardless of when they get created.
Sure, crewmates created in mapclass::loadlevel() have their drawframes
fixed there, but for crewmates that get created from scripting (such as
Violet when gotorooming to the Ship teleporter room after Space Station
1), this fix doesn't apply to them. But now it does, and Violet will no
longer be facing the wrong way for 1 frame when teleporting to the Ship
teleporter room in the Space Station 1 Level Complete cutscene.
If the player is stuck in a wall (which shouldn't happen in the first
place), their sprite would always default to being flipped, even if they
were unflipped.
Being stuck in a wall is characterized by having both positive onfloor
and onground.
Some levels rely specifically on the fact that certain script boxes are
loaded using gamestates, instead of directly loading the script and
bypassing the gamestate system. Then weird things could happen. This
restores compatibility with those levels.
mapclass::twoframedelayfix() doesn't need to be updated because the
point of that function is to bypass the gamestate system entirely
anyways.
This moves the teleporter, activity prompt, and trophy text "don't draw"
conditionals to the part where the game checks collision with them,
instead of whenever the game draws them.
This makes it so that the game smoothly does the fade-in/fade-out
animation instead of suddenly stopping rendering them whenever their
"don't draw" conditions apply. Now, the "Press ENTER to activate
terminal" prompt will no longer suddenly disappear whenever you activate
one, and the "- Press ENTER to Teleport -" prompt will smoothly fade
back in after teleporting, instead of suddenly popping in on screen.
Including a header file inside another header file means a bunch of
files are going to be unnecessarily recompiled whenever that inner
header file is changed. So I minimized the amount of header files
included in a header file, and only included the ones that were
necessary (system includes don't count, I'm only talking about includes
from within this project). Then the includes are only in the .cpp files
themselves.
This also minimizes problems such as a NO_CUSTOM_LEVELS build failing
because some file depended on an include that got included in editor.h,
which is another benefit of removing unnecessary includes from header
files.
There were many different ways I could've fixed it, but one thing that
stood out to me was the fact that touching the teleporter wasn't
guaranteed to set its onentity to 0, even though it should be. So now,
every time Viridian touches the teleporter, the teleporter's onentity
will be set to 0, and thus there's no chance of the teleporter
interrupting its own teleport animation and softlocking the game.
We should still do what I suggested in #391, namely setting
game.hascontrol to true if the game is in gamestate 0 and script.running
is false, and also always allowing Esc/Enter to be pressed regardless of
game.hascontrol. But this softlock is fixed now.
Fixes#391.
Instead of using gamestates, just directly use the 'script' attribute of
a script box if it is non-empty.
This is accomplished by having to return the index of the block that the
player collides with, so callers can inspect the 'script' attribute of
the block themselves, and do their logic accordingly.
To avoid going through gamestates, we'll need to carry the name of the
script on the script box itself. And to do that, we'll need to set the
'script' attribute of script boxes when translating edentities into real
entities in custom levels.
I ran the game through cppcheck and it spat out a bunch of member
attributes that weren't being initialized. So I initialized them.
In the previous version of this commit, I added constructors to
GraphicsResources, otherlevelclass, labclass, warpclass, and finalclass,
but flibit says this changes the code flow enough that it's risky to
merge before 2.4, so I got rid of those constructors, too.
Since they're always fixed-size, they don't need to be dynamically-sized
vectors.
entityclass::customcrewmoods is now a proper bool instead of an int now,
and I replaced the hardcoded constant 6 with a static const int Game
attribute to make it easier to change.
Their drawframe needs to be incremented by 2 instead of 1, because
they're double-sized.
Animation type 3 is used by the cloud emitter in The Solution is
Dilution, animation type 6 is used by the radar dish in Comms Relay.
Animation type 4 is used by the maverick bus in B-B-B-Busted, but it's
not noticeable since it spawns offscreen. This bug would cause all of
those entities to appear incorrectly for the deltaframes between the
tick the room got loaded and the next tick after that.
This is noticeable in flibit's tweet showing off my over-30-FPS patch:
https://twitter.com/flibitijibibo/status/1273983014930993153
The only reason why gray Warp Zone entities were green originally was
because there is a giant concatenated list of tileset+tilecol
combinations, and by using tileset 3 tilecol 6 you're using the entry
of tileset 4 tilecol 0, which is the green Ship tileset.
So without interfering with the green Ship tileset's entry, I've decided
that the best thing to do is to just add special cases. The enemy color
was easy enough to fix. The platform color was also easy to fix.
However, there exist no existing textures for gray conveyors, so at that
point I decided to just tint the existing green one gray, and then I did
the same for platforms.
This patch is very kludge-y, but at least it fixes a semi-noticeable
visual issue in custom levels that use internal scripts to spawn
entities when loading a room.
Basically, the problem here is that when the game checks for script
boxes and sets newscript, newscript has already been processed for that
frame, and when the game does load a script, script.run() has already
been processed for that frame.
That issue can be fixed, but it turns out that due to my over-30-FPS
game loop changes, there's now ANOTHER visible frame of delay between
room load and entity creation, because the render function gets called
in between the script being loaded at the end of gamelogic() and the
script actually getting run.
So... I have to temporary move script.run() to the end of gamelogic()
(in map.twoframedelayfix()), and make sure it doesn't get run next
frame, because double-evaluations are bad. To do that, I have to
introduce the kludge variable script.dontrunnextframe, which does
exactly as it says.
And with all that work, the two-frame (now three-frame) delay is fixed.
What happens here is that the entity gets created and then gets
immediately updated on the next frame, but there's no time for their
walkingframe of 0 to be rendered, so it'll look like they have just
started with walkingframe 1. However in the delta-timestep rendering
it'll render with walkingframe 0. So we need to fix their drawframe and
increment it when creating them.
This is because their oldxp wasn't being updated when they move (or
rather, teleport) and wrap around the screen.
These enemies are ZZT centipedes, but they're referred to as ASCII
snakes in comments in the code.
Okay, so the problem here is that Graphics::setcol() is called right
before a sprite is drawn in a render function, but render functions are
done in deltatime, meaning that the color of a sprite keeps being
recalculated every time. This only affects sprites that use fRandom()
(the other thing that can dynamically determine a color is help.glow,
but that's only updated in the fixed-timestep loop), but is especially
noticeable for sprites that flash wildly, like the teleporter, trinket,
and elephant.
To fix this, we need to make the color be recalculated only in the
fixed-timestep loop. However, this means that we MUST store the color of
the sprite SOMEWHERE for the delta-timesteps to render it, otherwise the
color calculation will just be lost or something.
So each entity now has a new attribute, `realcol`, which is the actual
raw color used to render the sprite in render functions. This is not to
be confused with their `colour` attribute, which is more akin to a color
"ID" of sorts, but which isn't an actual color.
At the end of gamelogic(), as well as when an entity is first created,
the `colour` is given to Graphics::setcol() and then `realcol` gets set
to the actual color. Then when it comes time to render the entity,
`realcol` gets used instead.
Gravitron squares are a somewhat tricky case where there's technically
TWO colors for it - one is the actual sprite itself and the other is the
indicator. However, usually the indicator and the square aren't both
onscreen at the same time, so we can simply switch the realcol between
the two as needed.
However, we can't use this system for the sprite colors used on the
title and map screen, so we'll have to do something else for those.
Just to make sure it's extra smooth. Not that it will be noticeable, and
you can't access the Secret Lab in slowmode without modifying the game,
but it's nice to have this.
This has to be done in order to fix rendering when on a conveyor or
moving platform and actively moving with or against it. Pretty sure this
shouldn't break anything, oldxp/oldyp is mostly visual after all (and by
the time it's used for gravity line collision checking,
updateentitylogic() would've already gotten around to it anyway).
Incidentally, this also fixes a jitter that would occur if you were
moving at the time you died or collected a trinket or custom crewmate,
due to the game temporarily freezing and either doing deathsequence or
completestop.
This fixes entities being drawframe 0 for 1 frame when being first
created. Incidentally, this also fixes entities created during
completestop being the player sprite, too, which is something not many
people notice.
This prevents undefined behavior because we use oldxp/oldyp to do linear
interpolation.
It's also initialized in entclass::entclass(), just to be sure. And I've
deduplicated the regular xp/yp initialization in createentity(), too.
If a trinket or crewmate ID is out-of-bounds, it will not be created.
This is not only because the `collect`/`customcollect` check in
entityclass::createentity() would then be out-of-bounds, but also
touching it would also be out-of-bounds, too.
Display trinkets will always be created if the ID is out-of-bounds.
Apparently some people (@AllyTally) have been creating display trinkets
with IDs of -1 in order to get a display trinket that always shows up,
which is rather horrifying. But it makes sense, there's a lot more
nonzero int values than there are the amount of int values that are
zero, so it's fairly likely that the `collect` check will always come up
to be true (nonzero). Also, it's more useful to be able to have a
display trinket that always shows up without having to collect a trinket
beforehand, than it is to have it not be created (because technically by
default, you're already in the world where you don't have it created).
Display trinkets still have their `para` set to their ID, though, and if
they managed to gain an `onentity` of 1, bad things could happen... So
just to be sure, I added INBOUNDS checks to crewmates and trinkets in
entityclass::updateentities() so no UB will happen if you collect a
crewmate/trinket with an out-of-bounds ID. Also, I de-duplicated the
`collect`/`customcollect` setting, too.
I tracked down all the functions that took in an entity's drawframe and
made sure that no matter what value an entity's drawframe was, the game
would never segfault.
To deal with using a different image file for Flip Mode, it looks like
copy-paste was used. This isn't exactly maintainable code. So I'm
replacing it with a reference that changes depending on if the game is
in Flip Mode or not, instead.
Removing the player entity has all sorts of nasty effects, such as
softlocking the game because many inputs require there to be a player
present, such as opening the quit menu.
The most infamous glitch to remove the player entity is the Gravitron
Fling, where the game doesn't see a gravity line at a specific
y-position in the current room, and when it moves the bottom gravity
line it moves the player instead. When the gravity line gets outside the
room, it gets destroyed, so if the player gets dragged outside the room,
they get destroyed, too. (Don't misinterpret this as saying anytime the
player gets dragged outside the room, they get destroyed - it's only the
Gravitron logic that destroys them.)
Also, there are many places in the code that use entity-getting
functions that have a fallback value of 0. If it was possible to remove
the player, then it's possible for this fallback value of 0 to index
obj.entities out-of-bounds, which is not good.
To fix this, entityclass::removeentity() is now a bool that signifies if
the entity was successfully removed or not. If the entity given is the
player (meaning it first checks if it's rule 0, just so in 99% of cases
it'll short-circuit and won't do the next check, which is if
entityclass::getplayer() says the indice to be removed is the player),
then it'll refuse to remove the entity, and return false.
This is a change in behavior where callers might expect
entityclass::removeentity() to always succeed, so I changed the
removeentity_iter() macro to only decrement if removing the entity
succeeded. I also changed entityclass::updateentities() from
'removeentity(i); return true;' to 'return removeentity(i);'.
obj.getplayer() can return -1, which can cause out-of-bounds indexing of
obj.entities, which is really bad. This was by far the most changes, as
obj.getplayer() is the most used entity-getting function that returns
-1, as well as the most-used function whose sentinel value goes
unchecked.
To deal with the usage of obj.getplayer() in mapclass::warpto(), I just
added general bounds checks inside that function instead of changing all
the callers.
Instead of using somewhat-obtuse for-loops to initialize or reset these
vectors, it takes up less lines of code and is clearer if we use
std::vector::resize() and std::vector::clear() instead.
We need to replace an "or" with an "and".
My best guess for this oversight happening was because of the weird
ordering. The code originally did "temp < 30" first and "temp > -30"
second instead of the other way around. With the weird ordering, it
becomes more natural to insert an "or" instead of an "and". So I swapped
around the ordering just for good measure.
This is also fixed in the mobile version.
Due to the previous commit, these will no longer be regularly taking in
out-of-bounds entity indices. Or at least they shouldn't, so I'm putting
in these print statements here on the off-chance that they do.
Otherwise, this would result in the game updating an entity twice, which
isn't good. This is most noticeable in the Gravitron, where many
Gravitron squares are created and destroyed at a time, and it's
especially noticeable during the part near the end of the Gravitron
where the pattern is two Gravitron squares, one at the top and bottom,
and then two Gravitron squares in the middle afterwards. The timing is
just right such that the top one of the two middle ones would be
misaligned with the bottom one of the two when a Gravitron square gets
outside the screen.
To do this, I changed entityclass::updateentities() into a bool, and
made every single caller check its return value. I only needed to do
this for the ones preceding updateentitylogic() and
entitymapcollision(), but I wanted to play it safe and be defensive, so
I did it for the disappearing platform kludge, as well as the
updateentities() within the updateentities() function.
The main ones to beware of here are entityclass::updateentities(),
entityclass::updateentitylogic(), and entityclass::entitymapcollision().
They would index out-of-bounds and thus commit Undefined Behavior if the
entity was removed in entityclass::updateentities(). And it would've
been fine enough if I only added bounds checks to those functions.
However, I decided to be a bit more defensive and play it safe, and
added bounds checks to ALL functions taking in not only an entity
indice, but also blocks and linecrosskludge indices.
It's treated like a bool anyway, so might as well make it one.
This also necessitates updating every single instance where it or an
element inside it is used, too.
game.trinkets is supposed to be correlated with obj.collect, however why
not just count obj.collect directly?
This turns game.trinkets into a function, game.trinkets(), which will
directly count the number of collected trinkets and return it. This will
fix a few corner cases where the number of trinkets can desync with the
actual collection statuses of trinkets.
In order to keep save compatibility with previous versions of VVVVVV,
the game will still write the <trinkets> variable. However, it will not
read the <trinkets> variable from a save file.
Whenever you collect a trinket, game.stat_trinkets gets updated (if
applicable) to signify the greatest amount of trinkets you have ever
collected in the game. However, custom levels shouldn't be able to
affect this, as their trinkets are not the same trinkets as the main
game.
Just like earlier, these are of the form
if (cond1) { if (cond2) { if (cond3) { thing; } } }
and are really annoying to read.
Also this removes the remnants of the 'active' system that have been
replaced with 'if (true)' conditionals in order to not add noise to the
diff.
This removes the variables obj.nblocks, as well as removing the 'active'
attribute from the block object. Now every block is guaranteed to be
real without having to check the 'active' variable.
Removing a block while iterating now uses the removeblock_iter() macro.
I guess these were here earlier when there were 'active' conditionals,
but then I removed those, so now they look weird next to the 'i != j'
conditionals, so I'm removing them.
These would be of the form
if (cond1) { if (cond2) { if (cond3) { thing; } } }
which is really annoying to read and could've been written as
if (cond1 && cond2 && cond3) { thing; }
so that's what I'm fixing here.
There will be another commit later that fixes this but in places related
to blocks.
Not sure why this function is here. It makes sense if you know that the
game will only do certain moving platform things if you already have a
moving platform in the room, however apparently this function has
absolutely nothing to do with it.
This function's sole purpose was to make sure obj.nentity was in sync,
and that obj.nentity-1 pointed to the last 'active' entity in
obj.entities. But now that obj.nentity is removed and we use
obj.entities.size() instead, it is no longer necessary.
In the previous commit, if an if-statement consisted solely of checking
the active attribute of an entity, I temporarily changed it to 'true'
and put a comment to remove it later, because it would add too much
noise to unindent everything in the same commit.
This removes the variables obj.nentity and obj.nlinecrosskludge, as well
as removing the 'active' attribute from the entity class object. Now
every entity you access is guaranteed to be real and you don't have to
check the 'active' variable.
The biggest part of this is changing createentity() to modify a
newly-created entity object and push it back instead of already
modifying an indice in obj.entities.
As well, removing an entity now uses the new obj.removeentity() function
and removeentity_iter() macro.
Ok, once we switch everything over to not use the 'active' system, it's
easier to read removeentity(t) than it is to read
entities.erase(entities.begin() + t).
This moves the setenemyroom() function onto the entity object itself, so
I can more easily change all 'entities[k].' to 'entity.' in
entityclass::createentity() later.
Additionally, I've had to move the rn() macro from Entity.h to Ent.h, or
else entclass::setenemyroom() won't know what it is.
This moves the setenemy() function onto the entity object itself,
instead of having to give the indice of the entity in obj.entities. This
makes the code more object-oriented so later I can simply change all
'entities[k]' to 'entity.' in entityclass::createentity().
This changes something like UtilityClass::String to help.String,
basically. It takes less typing this way, and is a neat effect of having
global args actually be global variables.
This commit removes all global args from functions on the entityclass
object, and updates the callers of those functions in other files
accordingly (most significantly, the game level files Finalclass.cpp,
Labclass.cpp, Otherlevel.cpp, Spacestation2.cpp, WarpClass.cpp, due to
them using createentity()), as well as renaming all instances of 'dwgfx'
in Entity.cpp to 'graphics'.
Previously, in tower mode, being inside walls would just kill you, unlike
being inside walls outside tower mode, which was somewhat confusing.
Also, spikes behaved differently with regards to invincibility, being
unsolid in towers but solid outside them.
This does not change the behaviour of the "edge" spikes in towers.
This fixes a bug where warp lines created through createentity wouldn't
work without there already being a warp line present in the room as an
edentity.
Yikes. Somebody brought this to my attention, I didn't even remember that I'd written it. "Spa" or "Spastic" is kind of a south park esque slang term that used to be pretty common in Ireland, which I used without even thinking about it. It's definitely not something I would say anymore, 10 years on, and it's something I shouldn't have said at the time either. I'm sorry :(
(somebody on twitter was asking me about how much cleaning up of the source code I did before launching this. I think this commit kinda answers that)
* Add entity type attribute checks to getcrewman()
This means that the game is no longer able to target other non-crewmate
entities when it looks for a crewmate.
This has actually happened in practice. A command like
position(cyan,above) could position the text box above a horizontal warp
line instead of the intended crewmate.
This is because getcrewman() previously only checked the rule attributes
of entities, and if their rule happened to be 6 or 7. Usually this
corresponds with crewmates being unflipped or flipped, respectively.
But warp lines' rules overlap with rules 6 and 7. If a warp line is
vertical, its rule is 5, and if it is horizontal, its rule is 7.
Now, usually, this wouldn't be an issue, but getcrewman() does a color
check as well. And for cyan, that is color 0. However, if an entity
doesn't use a color, it just defaults to color 0. So, getcrewman() found
an entity with a rule of 7 and a color of 0. This must mean it's a cyan
crewmate! But, well, it's actually just a horizontal warp line.
This commit prevents the above from happening.
Resizing the vector does the same thing that the loops did, it changes
the size for the vector and initializes it with default-constructed
elements (or 0 or its equivalent for POD types).
Where a specific value is needed, it is set with the second
parameter of resize().
