removing object creted in loop help

Solar2D Lua Language
#1

Hi,

I have been messing with joints, I used Carlos’ ball on a rope code as a reference.

Now my problem is, how the heck do I destroy the rope ?

I have tried many things, using pairs and ipairs but no joy whatsoever.

I am missing something simple I feel can anybody help and point me in the correct direction please.

[lua]

local physics = require (“physics”)
physics.start ()
physics.setGravity (0, 10)
 
local ceiling   = display.newRect (0, 0, display.contentWidth, 1)
physics.addBody (ceiling, “static”, {bounce = 0.0, friction = 10})

local xCenter = 160
local hCeil = -5

local prevBody = ceiling

local w,h = 10,10
local halfW,halfH = 0.5*w,0.5*h

local x = xCenter
local y = hCeil - halfH
local yJoint = y - halfH

for i = 1, 20 do
    y = y + h
    yJoint = yJoint + h
    
     local chainlinks = display.newImage(“rope.png” ,x-halfW, y-halfH)
     physics.addBody( chainlinks, { density=15, friction=0.5, bounce = .2 })

     local chainjoint = physics.newJoint( “pivot”, prevBody, chainlinks, xCenter, yJoint )
     newBody = chainlinks
    
     function removeLinks()

       print(“removelinks called”) --CODE HERE TO REMOVE THE ROPE–
       
     end    
end

local key = display.newImage(“soccerball.png”, x,y -30)
physics.addBody( key, { density=2, friction=0.5, bounce=.2, radius=20  })

local joint = physics.newJoint( “pivot”, newBody, key, xCenter, y )

local function removeEverything()

   display.remove(ceiling)
   ceiling = nil
   removeLinks()
   display.remove(key)
   key = nil
end

timer.performWithDelay(2000,removeEverything,1)

[/lua]

#2

Store them somewhere :slight_smile:

#3

Haha,

yup, stored in a table and then removed like so:

for i = #chainlinks,1,-1 do
         chainlinks[i]:removeSelf()
         chainlinks [i] = nil
         end

thanks

#4

Keep in mind when you nil an object in the middle of the table, #tablename returns a count upto the nil and you loose access to the rest of them.  There is a table function (table.maxn I think) that gets you the full length of the table.

So you can do:

for i = table.maxn(chainlinks), 1, -1 do

     if chainlinks[i] and chainlinks[i].removeSelf then

           chainlinks[i]:removeSelf()

           chainlinks[i] = nil

     end

end

or you can to a table.remove() on the cell you just removed (though if its a big list that your checking a lot, its a bit pokey to do

#5

@Rob,

great, that takes care of things better.

Thanks for taking time to reply chaps :slight_smile:

#6

Rob, is there any performance disadvantage to using in pairs() instead of table.maxn() when you have gaps?

#7

You probably meant ipairs().

Pairs() is used to iterate through array with keys not indexes. Maxn() can give you wrong size if table has just keys.

#8

I don’t have any statistics to back things up, but we did a blog post that if I can summarize, ipairs is expensive where #table size is not.  I suspect that ipairs() is more expensive than .maxn() but .maxn() is a bit more costly than #array.  But if you have nil’s, you don’t have much choice.

#9

Store them somewhere :slight_smile:

#10

Haha,

yup, stored in a table and then removed like so:

for i = #chainlinks,1,-1 do
         chainlinks[i]:removeSelf()
         chainlinks [i] = nil
         end

thanks

#11

Keep in mind when you nil an object in the middle of the table, #tablename returns a count upto the nil and you loose access to the rest of them.  There is a table function (table.maxn I think) that gets you the full length of the table.

So you can do:

for i = table.maxn(chainlinks), 1, -1 do

     if chainlinks[i] and chainlinks[i].removeSelf then

           chainlinks[i]:removeSelf()

           chainlinks[i] = nil

     end

end

or you can to a table.remove() on the cell you just removed (though if its a big list that your checking a lot, its a bit pokey to do

#12

@Rob,

great, that takes care of things better.

Thanks for taking time to reply chaps :slight_smile:

#13

Rob, is there any performance disadvantage to using in pairs() instead of table.maxn() when you have gaps?

#14

You probably meant ipairs().

Pairs() is used to iterate through array with keys not indexes. Maxn() can give you wrong size if table has just keys.

#15

I don’t have any statistics to back things up, but we did a blog post that if I can summarize, ipairs is expensive where #table size is not.  I suspect that ipairs() is more expensive than .maxn() but .maxn() is a bit more costly than #array.  But if you have nil’s, you don’t have much choice.