I have a model called door
Inside I have a BoolValue named Open
I have a model called Top that has all of the door blocks named Work Mabey Comeon and Proboblynot
And I have Block that when touched is supposed to make Top move up
Directly inside door I have this script
door = script.Parent
open = door.Open
Top = door.Top
opener = 18
speed = 100
steps = speed
startl = Top.CFrame
function MoveDoorToCFrame(cfrm,dr)
dr.Work.CFrame = cfrm
dr.Mabey.CFrame = dr.Work.CFrame * CFrame.new(0,-7.2,0)
dr.Comeon.CFrame = dr.Work.CFrame * CFrame.new(0,10.8,0)
dr.Problynot.CFrame = dr.Work.CFrame * CFrame.new(0,10.8,0)
end
function Update()
if speed/steps < 0.5 then
calc = 1-math.cos(math.rad((-90/speed)*steps*2))
else
calc = 1+math.sin(math.rad((90/speed)*((speed/2)-steps)*2))
end
MoveDoorToCFrame(startl * CFrame.new(0,(calc/2)*opener,0),Top)
end
Update()
while true do
wait()
if not open.Value and steps < speed then
steps = steps + 1
Update()
elseif open.Value and steps > 0 then
steps = steps - 1
Update()
end
end
Inside the button that is supposed to activate on touch I have
script.Parent.Touched:connect(function()
script.Parent.Parent.Open.Value = not script.Parent.Parent.Open.Value
end)
script.Parent.Parent.Open.Changed:connect(Update)
Update()
If you know how to fix this it would be gladly appreciated.
Update November 2015:
Using PrimaryPart
Since writing this post, ROBLOX has changed a lot in regards to the API. To move a model like requested, you should set the PrimaryPart property of the model to a central part inside the model. This will act as the origin for the model's movements.
You can then use model:SetPrimaryPartCFrame(cframe) to set the CFrame of the model. You can also retrieve this property by using model:GetPrimaryPartCFrame(), although I believe that is just a shortcut method for model.PrimaryPart.CFrame.
In code, it would look like this:
-- Set PrimaryPart:
MODEL.PrimaryPart = MODEL.SomeCentralPart
...
-- CFrame movement:
local movement = CFrame.new(0, 10, 0)
-- Move the model:
MODEL:SetPrimaryPartCFrame(MODEL:GetPrimaryPartCFrame() * movement)
Option A: Use Model's methods
I think you are making this much more difficult than it needs to be. Whenever you run into an issue like this, be sure to check the current APIs provided. The ROBLOX Model object contains a nifty method called 'TranslateBy' which takes a Vector3 argument to translate the model.
Using MODEL:TranslateBy(Vector3) is similar to moving a model via CFrame, since it ignores collisions.
Another alternative is MODEL:MoveTo(Vector3) which moves a whole model to the given Vector3 world position. The downside to this is that it does collide.
One way to get the same MoveTo effect but without collisions can be done with the TranslateBy method:
MODEL:TranslateBy(Vector3Position - MODEL:GetModelCFrame().p)
Option B: Write a custom function to manipulate the model's CFrame
Another alternative would be to manipulate the whole model's CFrame entirely. To do this, you can write a clever function that will move a whole model relative to an 'origin' point. This is similar to moving shapes on a grid given their points and an origin, except in three dimensions. Using ROBLOX's built-in functions, this is much easier though.
A good way to do this would be to write a function that lets you actually assign a CFrame value to a whole model. Another way would be to allow a translation via CFrame too.
Here's an example:
function ModelCFrameAPI(model)
local parts = {} -- Hold all BasePart objects
local cf = {} -- API for CFrame manipulation
do
-- Recurse to get all parts:
local function Scan(parent)
for k,v in pairs(parent:GetChildren()) do
if (v:IsA("BasePart")) then
table.insert(parts, v)
end
Scan(v)
end
end
Scan(model)
end
-- Set the model's CFrame
-- NOTE: 'GetModelCFrame()' will return the model's CFrame
-- based on the given PrimaryPart. If no PrimaryPart is provided
-- (which by default is true), ROBLOX will try to determine
-- the center CFrame of the model and return that.
function cf:SetCFrame(cf)
local originInverse = model:GetModelCFrame():inverse()
for _,v in pairs(parts) do
v.CFrame = (cf * (originInverse * v.CFrame))
end
end
-- Translate the model's CFrame
function cf:TranslateCFrame(deltaCf)
local cf = (model:GetModelCFrame() * deltaCf)
self:SetCFrame(cf)
end
return cf
end
-- Usage:
local myModel = game.Workspace.SOME_MODEL
local myModelCF = ModelCFrameAPI(myModel)
-- Move to 10,10,10 and rotate Y-axis by 180 degrees:
myModelCF:SetCFrame(CFrame.new(10, 10, 10) * CFrame.Angles(0, math.pi, 0))
-- Translate by 30,0,-10 and rotate Y-axis by 90 degrees
myModelCF:TranslateCFrame(CFrame.new(30, 0, -10) * CFrame.Angles(0, math.pi/2, 0))
This might be hard.
You might want to look to free models for this one unless the people above get it to work.
I, however, do have a script to move a model:
game.Workspace.Model:MoveTo(Vector3.new(0,0,0))
Your code indeed needs fixing.
You should NOT use a never-ending loop to make your stuff work (unless that is the only way).
You should rather base actions on events.
Consider to use this:
Structure:
Door [Model]
DoorScript [Script]
Button [Part]
DoorOpen [BoolValue]
Top [Model]
Mabey [Part]
Comeon [Part]
Problynot [Part]
DoorScript:
local Model = script.Parent
local Door = Model.Top
local Button = Model.Button
local DoorOpen = Model.DoorOpen
local Offset = 0
local ToOffset = 100
local Direction = 1
local StepLength = 0.1
local Moving = false
function StartMoving()
if Moving then return end
Moving = true
while (DoorOpen.Value and Offset ~= ToOffset) or (not DoorOpen.Value and Offset ~= 0) do
local Change = Offset
Offset = math.max(0,math.min(ToOffset,Offset + StepLength * (DoorOpen.Value and 1 or -1)))
Change = Offset - Change
Top:TranslateBy(Vector3.new(0,Change,0))
wait()
end
Moving = false
end
StartMoving()
DoorOpen.Changed:connect(StartMoving)
local Debounce = false
Button.Touched:connect(function()
if Debounce then return end
Debounce = true
DoorOpen.Value = not DoorOpen.Value
wait(4)
Debounce = false
end)
You might want to adjust the speed tho.
This can be used to move models, try adding something like this into your code. It's more dynamic.
a = Workspace.Model
for i=0.1,40 do
for i,v in pairs(a:getChildren()) do
if v:IsA("Part") then
v.CFrame = CFrame.new(v.CFrame + Vector3.new(0,0.1,0))
else print("Not a part")
end
end
end
Related
I'm trying to make a procedurally generating game (ignore the goal of the game).
Currently I'm trying to make it so when you walk over a grid piece and collide with an invisible part it makes that specific grid piece the Current Global Object (CGO).
Using a BoolValue I made it possible to use this as some form of global variable, but whenever I try to use said value it only detects it as false even when it is indeed showing as true in the Explorer. It only works when I set the value to true BEFORE testing the game.
Here's the code in the script that's meant to detect the value:
it's a regular script and not a LocalScript fyi
local CGO = "0,0"
local tracker = game.Workspace.Tracker00
--local CGOa = tracker.CGOa
local trackerPos = tracker.Position
local trackerX = trackerPos.X
local trackerY = trackerPos.Y
local trackerZ = trackerPos.Z
while true do
while tracker.CGOa.Value == false do
tracker.YVal.Value = 7
print("Set TrackerYVal to 7")
wait()
if tracker.CGOa.Value == true then
break
end
end
while tracker.CGOa.Value == true do
tracker.YVal.Value = 14
print("Set TrackerYVal to 14")
wait()
end
tracker.CFrame = CFrame.new(trackerX, trackerY, trackerZ)
wait()
end
Any help would be much appreciated.
Rather than using infinite while loops, consider listening for the Changed signal. It's possible that the break command might be escaping both loops.
local tracker = game.Workspace.Tracker00
tracker.CGOa.Changed:Connect(function(newVal)
print("CGOa changed to ", newVal)
if newVal then
tracker.YVal.Value = 14
print("Set TrackerYVal to 14")
else
tracker.YVal.Value = 7
print("Set TrackerYVal to 7")
end
-- update the tracker position based on newly updated values
local x = tracker.XVal.Value
local y = tracker.YVal.Value
local x = tracker.ZVal.Value
tracker.CFrame = CFrame.new(Vector3.new(x, y, z))
end)
I made some assumptions about how you were positioning the tracker, because the old code would reset its position after every loop.
I'm creating a 2D platform game using Corona SDK and I'm stuck with collisions.
Basically there is a character that runs over this ground made of blocks. This is because sometimes there can be holes in the ground.
The game is an endless one, so as the character moves forward new blocks (and holes) are dynamically added - and also removed if they goes off screen.
It works nicely but this approach works against the collision system, let me explain how.
Now that I have the ground in place I want the character to jump but only if it is touching the ground, to avoid jumping while in air.
Whenever a collision is detected between character and ground an event is fired - two times. The first time when the character is entering a ground block and the second time when the character leaves it. So when the character lands on the ground a isGround Boolean is set to true. And when - after a jump - it leaves it the flag is set to false. The problem is that every time it exits a block to enter another - walking along the ground without jumping - the flag get updated. This makes the jump based on the isGround flag less reliable. Sometimes it happens that you can't jump because isGround == false though the character is on the ground.
Ground block creation snippet
-- init() method set the sprite of the ground block and physic to that sprite
function GroundBlock:init()
self.sprite = display.newImageRect(self.path, self.width, self.height)
self.sprite.x = self.x
self.sprite.y = self.y
physics.addBody(self.sprite, 'static', {
density = 0,
friction = 0,
bounce = 0,
box = {
halfWidth = self.width / 2,
halfHeight = self.height / 2,
y = 16,
x = 0
}
})
local collisionObj = {
name = 'ground'
}
self._collision = collisionObj
self.sprite._collision = collisionObj
self.isShow = true
end
Ground placing GroundBlocks snippet
-- init() method initialize the ground with a fixed number of blocks
function Ground:init()
self.offsetX = 0
while self.offsetX < self.camera.borderRight * 2 do
self._createBlock(1)
end
self.lastCameraPos = self.camera.borderRight
end
-- update() is called once per frame
function Ground:update()
if (self.camera.borderRight - self.lastCameraPos > self._blockWidth) then
local rand = math.ceil(math.random() * 10) % 2
if self._skippedBlock >= 2 or rand == 0 then
self._createBlock(1)
self._skippedBlock = 0
else
self._createBlock(0)
self._skippedBlock = self._skippedBlock + 1
end
self.lastCameraPos = self.camera.borderRight
end
for i, block in ipairs(self.blocks) do
if block.sprite.x < self.camera.borderLeft - block.width then
table.remove(self.blocks, i)
self.camera:remove(block.sprite)
block:delete()
end
end
end
Collision detection snippet
function Character:collision(event)
if ( event.phase == "began" ) then
if event.other._collision.name == "ground" then
self.isGround = true
end
elseif ( event.phase == "ended" ) then
if event.other._collision.name == "ground" then
self.isGround = false
print('nope')
end
end
end
A solution would be to make a ground as a single imgRect but how to make holes in it?
You could simplify your code and prevent this issue from ever occurring by tracking if the character can jump instead of tracking if the character is on the ground.
For instance,
function jump( event )
if event.phase == "began" then
if canJump then
canJump = false
-- your code that makes the player jump
end
end
end
You probably use touches to determine whether the player character jumps, right? This way, you'll trigger the jump when the touch starts as long as the character has not already jumped.
You could then reset this value in your collision function by editing it slightly:
function Character:collision(event)
if event.phase == "began" then
if event.other._collision.name == "ground" then
canJump = true
end
end
end
This way, the character's ability to jump is determined by whether or not the player has pressed jump already and if the character has hit the ground since the last jump.
This kind of approach also gives you the ability to pivot towards implementing mechanics like double jump. If instead of using a boolean canJump variable you chose to use a number variable, e.g. jumpsLeft, you could reduce the number of jumps left every time the character jumps and only let the character jump if jumpsLeft is larger than 0. Then you'd simply reset the value back to 1 (or whatever you'd want upon hitting the ground).
I've been trying to make a game where you're in a square and when you go to the sides, parts come up and block you.
I've gotten far to the point where it's working fine, except for a few problems:
the parts go below the square when not raised, I want them to be visible when they're not raised
the parts go down when you jump, making it easy to escape.
the parts go up too early
This is the code that deals with the wall positioning.
for _, v in pairs(model:GetChildren()) do
if string.sub(v.Name,1,4) == "Wall" then
local walls = {}
walls[v] = {v.CFrame,Vector3.new(1, 1, 1)}
game:GetService("RunService").RenderStepped:connect(function()
if(workspace[game.Players.LocalPlayer.Name]:FindFirstChild("HumanoidRootPart")) then
local mag = (v.Position - workspace[game.Players.LocalPlayer.Name]:FindFirstChild("HumanoidRootPart").Position).magnitude
sizeFactor = math.floor(mag)
v.CFrame = walls[v][1]*CFrame.new(0,-sizeFactor+(walls[v][1].Y*1.8),0)
end
end)
end
end
You can see my game here: https://www.roblox.com/games/400391033/Marble-walls
See commented code.
for _, v in pairs(model:GetChildren()) do
if string.sub(v.Name,1,4) == "Wall" then
local walls = {}
walls[v] = {v.CFrame,Vector3.new(1, 1, 1)}
game:GetService("RunService").RenderStepped:connect(function()
if(workspace[game.Players.LocalPlayer.Name]:FindFirstChild("HumanoidRootPart")) then
local mag = (v.Position - workspace[game.Players.LocalPlayer.Name]:FindFirstChild("HumanoidRootPart").Position).magnitude
if (mag <= 2) then --[[
Currently your issue is that you never actually do ANYTHING regarding magnitude
you essentially change the y-Axis as soon as the player spawns.. hence why it does it too early
kappa
]]
sizeFactor = math.floor(mag)
v.CFrame = walls[v][1]*CFrame.new(0,-sizeFactor+(walls[v][1].Y*1.8),0)
end;
end
end)
end
end
Im quite new to Lua so please pardon my ignorance but i cannot find the solution to my problem.
Whats going on
Im currently trying to move objects from A to B and once object is at B to restart at A and again move to B in a continuous cycle.
local function moveLeft(obj)
print("moving left")
local function resetObj(obj)
transition.to (obj,{ time = 10, x = obj.x + screenWidth + obj.width, onComplete=moveLeft })
end
transition.moveBy (obj,{ time = 3000, x = -screenWidth -obj.width, onComplete=resetObj })
end
and then called using
for idx1 = 1, 8 do
enemyRed = display.newImage("Images/enemyRed.png")
-- 7. Apply physics engine to the enemys, set density, friction, bounce and radius
physics.addBody(enemyRed, "dynamic", {density=0.1, friction=0.0, bounce=0, radius=9.5});
local xPositionEnemy = math.random() + math.random(1, screenWidth)
enemyRed.x = xPositionEnemy;
enemyRed.y = yPosition;
enemyRed.name = "enemyRed"..idx
moveLeft(enemyRed);
end
This is great and all objects are moving from A to B
Problem / issue
The issue here is that the onComplete is not called until ALL objects named "enemyRed" are at point B.
Question
What i want is for each individual object named "enemyRed" to reset to original position A once its reached its destination.
I can't answer the problem/issue because it is not clear (I added a comment). Re the question, you should probably add a A position field to each object, this way you can easily return to it (stylistic note: this is Lua, not c, you don't need semicolons). So In your loop do this:
enemyRed.x = xPositionEnemy
enemyRed.startPos = xPositionEnemy
then in your resetObj do this:
local function moveLeft(obj)
local function resetObj()
print("moving back")
transition.to (obj,
{ time = 10, x = obj.startPos, onComplete=function() moveLeft(obj) end })
end
print("moving left")
transition.moveBy (obj,
{ time = 3000, x = -screenWidth - obj.width, onComplete=resetObj })
end
The above also shows that when calling your moveLeft from the resetObj function, you have to give the obj to the moveLeft otherwise obj will be nil. The resetObjdoes not needobj` parameter since it is an upvalue already.
I am attempting to add/remove objects from the physics engine (addBody() and removeBody()) in an app I am working on. The app I am working on is modular so the issue is in one of two files.
The objects file (TransmitterObject) or the main file (main):
This is the relevant code for both:
main.lua
local physics = require("physics")
physics.start()
physics.setGravity(0,0)
physics.setDrawMode( "debug" )
local TransmitterObject = require("TransmitterObject")
function updateGame(event)
if(ITERATIONS % 100 == 0) then
tran1:activate() --create new physics object here
end
ITERATIONS = ITERATIONS + 1
--print(ITERATIONS)
end
Runtime:addEventListener("enterFrame", updateGame)
TransmitterObject.lua
function transmitter.new(props) --constructor
Transmitter =
{
x = props.x,
y = props.y,
receivers = props.receivers
}
return setmetatable( Transmitter, transmitter_mt )
end
function transmitter:activate()
local group = math.random(1, #self.receivers)
local receiver = math.random(1,#self.receivers[group])
local x , y = self.receivers[group][receiver][1], self.receivers[group][receiver][2]
local d = math.sqrt(math.pow((self.x-x),2) + math.pow((self.y-y),2))
local dx = math.abs(self.x - x)
local angle = math.deg(math.acos(dx/d))
local beam = display.newRect(self.x,self.y, d, 10)
beam:setReferencePoint(display.TopLeftReferencePoint)
beam.rotation = 180 + angle
beam:setFillColor(0,255,0)
beam.alpha = 0
local function add(event)
physics.addBody(beam, "static")
end
local function delete(event)
physics.removeBody(beam)
end
transition.to( beam, { time=1000, alpha=1.0, onComplete=add } )
transition.to( beam, { time=1000, delay=2500, alpha=0, onComplete=delete})
end
Now let me try to describe the issue a little better. basically every 100th time that 'enterFrame' fires I tell the transmitter object (tran1) to call its function 'activate'
which then preforms some basic math to get coordinates. Then it creates a rectangle (beam) using the calculated information and sets some properties. That is all basic stuff. Next I tell it to transition from not visible (alpha = 0) to visible over the span of 1 second. When does it is to call the function 'add' which adds the object to the physics engine. Likewise with the next line where it removes the objects.
That being said, when i set physics.setDrawMode( "debug" ) the beam object appears as a static body, but does not accept collisions. Does anyone know why the above code would not accept collisions for the beam object?
Keep in mind I have other objects that do work properly within the physics engine.
Wow, I'm answering super late!
On collisions, modifying bodies aren't supported.
What I propose you is to create a new function,
local function addBody ( event )
physics.addBody(ball, "static")
end
and in your collision event you have to add this,
timer.performWithDelay(500, addBody)
The only thing that may cause some problems it's the delay, but as the collision doesn't take too much time it should be ok.
Sorry for this necroposting,
It's just to help other people that may have that problem,
Fannick