System Rectwraps

[Azlier]

Rectwraps Bonus Plugin #1 operational.

Anyone have any rect-using systems I could make plugins for?

 

[Azlier]

Plugin #1 internalized, gaining efficiency. Also added the .setPathing method, which switches a pathingtype for an entire rectwrap on or off.

 

[Azlier]

Added a new method called .iterateThrough. I might use it someday.

 

[Azlier]

Update to make .replaceTerrain, .setPathing, and .iterateThrough have much less chance of hitting the op limit. [del]Not even the normal ReplaceTerrain system does that for you .[/del] Wait, yes it does.

 

[Azlier]

Bump?

 

[Dr.Jack]

Approved!
Thanks for contributing!

 

[Jesus4Lyf]

Congratulations, Azlier!
I know a lot of work went into this.

 

[Azlier]

At last, I can rest in peace . (I'll wake up sooner or later.)

 

[Azlier]

Update (How is that possible?!). The .minX and friends properties can now be set, and the rectwrap will conform to them. Also added four more methods.

 

[Troll-Brain]

Ok, i need to play with rects/reg so i've take a look on this script.
I've read it quickly and some things hurt me, but i could be wrong though, you will tell me.

1) Are you sure you recycle the region ?

For me you create a region all the time, or add rects to it.
You could try to use RegionClearRect with GetWorldBounds(), even if in fact you would need to use a rect variable instead of GetWorldBounds, since it creates a new rect at each call.
But i dunno if simply using RemoveRegion/CreateRegion(), instead would be better or not (benchmark ? ...).

2) You create one region for the enter and other for the leave event ?

One for both is enough.

3) When you change the size or whatever of the rect you don't change the eventual region associated to it as well ?

But maybe you don't need, currently i haven't played at all with dynamic regions, only static ones.

 

[Azlier]

I'm very sure that the stuff is recycled. And, yes. Excellent point. Exit and enter can share a region. Now I just need to figure out how to do that efficiently.

The replaceTerrain type methods seriously need a rewrite.

>For me you create a region all the time, or add rects to it.

Hmm? I only create the region once, and add the rect once. I only disable triggers when required.

 

[Troll-Brain]

I mean when an instance of a struct is recycled after a destroy, so the region already contains one rect, and you add a new one instead of clear it and add the new rect.

Maybe it would help you or not,you can keep the 2 variables and only one region created.
set handle1 = handle2

 

[Azlier]

Err... could you explain that again?

>and you add a new one instead of clear it and add the new rect.

Please... I don't understand that at all.

 

[Troll-Brain]

library Test initializer init uses rectwrap

private function Actions takes nothing returns boolean
    call BJDebugMsg("unit enter region")
    return false
endfunction

private function init takes nothing returns nothing
    local rectwrap s_rect = rectwrap.create(32.,64.,32.,64.)
    local trigger trig = CreateTrigger()
    local unit u = CreateUnit(Player(0),'hpea',0.,0.,0.)
    
    call TriggerAddCondition(trig,Condition(function Actions))
    
    call s_rect.registerEnter(trig)
    
    call TriggerSleepAction(1.)
    
    call BJDebugMsg("move unit")
    call SetUnitPosition(u,50.,50.)
    call SetUnitPosition(u,0.,0.)
    
    call TriggerSleepAction(1.)
    
    call s_rect.destroy()
    set s_rect = rectwrap.create(128.,256.,128.,256.)
    call s_rect.registerEnter(trig)
    
    call BJDebugMsg("move unit, the message 'unit enter region' shouldn't be displayed")
    call SetUnitPosition(u,50.,50.)
    call BJDebugMsg("move unit")
    call SetUnitPosition(u,200.,200.)
    
endfunction

endlibrary

It doesn't work at all, i do something wrong ?

 

[Azlier]

Yes. .create takes minX, minY, then maxX and maxY. Treat it exactly like the [lJASS]Rect[/lJASS] native.

So, [lJASS]rectwrap.create(32, 32, 64, 64)[/lJASS] is correct.

 

[Troll-Brain]

Ok really bad mistake, fixed :

library Test initializer init uses rectwrap

private function Actions takes nothing returns nothing
    call BJDebugMsg("unit enter region")
endfunction

private function init takes nothing returns nothing
    local rectwrap s_rect = rectwrap.create(32.,32.,64.,64.)
    local trigger trig = CreateTrigger()
    local unit u = CreateUnit(Player(0),'hpea',0.,0.,0.)
    
    call TriggerAddAction(trig,function Actions)
    
    call s_rect.registerEnter(trig)
    
    call TriggerSleepAction(1.)
    
    call BJDebugMsg("move unit")
    call SetUnitPosition(u,50.,50.)
    call SetUnitPosition(u,0.,0.)
    
    call TriggerSleepAction(1.)
    call BJDebugMsg(" ")
    
    call s_rect.destroy()
    set s_rect = rectwrap.create(128.,128.,256.,256.)
    call s_rect.registerEnter(trig)
    
    call BJDebugMsg("move unit, the message 'unit enter region' shouldn't be displayed")
    call SetUnitPosition(u,50.,50.)
    
    call TriggerSleepAction(1.)
    call BJDebugMsg(" ")
    
    call BJDebugMsg("move unit")
    call SetUnitPosition(u,200.,200.) // the message "unit enter in region" isn't showed
    
endfunction

endlibrary

Test the code and you will see what i want mean, i hope.

 

[Azlier]

Well, it seems that SetUnitPosition makes enter rect events fire again. How odd!

Nothing I can fix, though. Everything else seems to work perfectly.

 

[Troll-Brain]

Well, it seems that SetUnitPosition makes enter rect events fire again. How odd!

Nothing I can fix, though. Everything else seems to work perfectly.

No, it's nothing about SetUnitPosition, it's because the region isn't cleared, a better way would be to destroy/ create region or use RegionClearRect with the entire map rect.

Also, the last move doesn't fire the trigger, i wouldn't call it "works perfectly" xD

 

[Azlier]

Huh. But with my tests, it showed that moving a rect actually changed the region.

 

[Troll-Brain]

Huh. But with my tests, it showed that moving a rect actually changed the region.

Maybe you failed, share the code.