System DoubleClick

[Azlier]

It detects when you click the same unit twice within the provided time (a second by default).

A single function is available to you.

function TriggerRegisterDoubleClickEvent takes trigger whichTrigger returns nothing

DestroyTrigger friendly, leakless, uses no H2I or any attachment besides what is done at initialization.

Requires Event.

Your friend, Azlier. Oh, and I just found that Azlier is just Lazier jumbled up. Bah.

library DoubleClick initializer Init requires Event


globals
    private constant real CLICK_THRESHHOLD = 1.
    //How much time between clicks it takes to register a double click.
endglobals

//** System Code **

globals
    private boolean array Clicking
    private unit array Clicked
    private timer array ClickTimer
    private Event Ev
    private trigger array ClickTrig
endglobals

private function SetTriggerData takes trigger t, integer i returns nothing
    loop
        exitwhen i == 0
        call TriggerExecute(t)
        set i = i - 1
    endloop
endfunction

function TriggerRegisterDoubleClickEvent takes trigger whichTrigger returns nothing
    call Ev.register(whichTrigger)
endfunction

private function Callback takes nothing returns boolean
    set Clicking[GetTriggerExecCount(GetTriggeringTrigger())] = false
    return false
endfunction

globals
    private integer Id
    private unit Cu
endglobals
    
private function Actions takes nothing returns boolean
    set Id = GetPlayerId(GetTriggerPlayer())
    set Cu = GetTriggerUnit()
    if Clicking[Id] then
        call PauseTimer(ClickTimer[Id])
        if Clicked[Id] == Cu then
            call Ev.fire()
            set Clicking[Id] = false
        else
            set Clicked[Id] = Cu
            call TimerStart(ClickTimer[Id], CLICK_THRESHHOLD, false, null)
        endif
    else
        set Clicking[Id] = true
        set Clicked[Id] = Cu
        call TimerStart(ClickTimer[Id], CLICK_THRESHHOLD, false, null)
    endif
    return false
endfunction

private function Init takes nothing returns nothing
    local trigger t = CreateTrigger()
    local integer i = 11
    set Ev = Event.create()
    call TriggerAddCondition(t, Condition(function Actions))
    loop
        exitwhen i < 0
        call TriggerRegisterPlayerUnitEvent(t, Player(i), EVENT_PLAYER_UNIT_SELECTED, null)
        set ClickTimer<i> = CreateTimer()
        set ClickTrig<i> = CreateTrigger()
        call SetTriggerData.execute(ClickTrig<i>, i)
        call TriggerRegisterTimerExpireEvent(ClickTrig<i>, ClickTimer<i>)
        call TriggerAddCondition(ClickTrig<i>, Condition(function Callback))
        set i = i - 1
    endloop
endfunction

endlibrary</i></i></i></i></i></i>

 

[Builder Bob]

It looks very user friendly and efficient.

I just thought I'd mention that the struct member trig will not magically become null when the trigger it is pointing to has been destroyed. As far as I can see it will never become null.

if .trig == null then
    //Did the user destroy the trigger? Well. 
    //Let's stitch the list together.
    set .next.prev = .prev
    set .prev.next = .next
    set .temp = this
    set this = .next
    call .temp.destroy()
else

Edit: [del]Using GetTriggerExecCount() as attached data looks functional. However, couldn't you just have used playerId of triggering player? I'm just wondering why you chose a more complex solution. No need to change anything for my sake anyway. It makes no difference as far as I can see.[/del]

How am I to get a triggering player in a timer callback?

A very good question.

 

[Azlier]

Really? Destroyed triggers aren't null? How am I to detect whether it has been destroyed or not, then

Edit:
How am I to get a triggering player in a timer callback?

 

[Builder Bob]

Really? Destroyed triggers aren't null? How am I to detect whether it has been destroyed or not, then?

If I only knew. This always seem to complicate my code a lot more than I'd like...

Edit: GetTriggerExecCount() and GetTriggerEvalCount() will report 0 for destroyed triggers. It will also report 0 for newly created triggers though, so it's not going to be a safe way to determine if the trigger has been destroyed or not unless you do something clever.

 

[Azlier]

Ah well. A shame, that is.

See my last post as an 'answer' to your last question.

 

[Builder Bob]

Destroyed triggers report false on IsTriggerEnabled() calls.

Here's a sure way to check if a trigger has been destroyed without affecting the game in any way.

private function TrigState takes trigger trig returns nothing
	if IsTriggerEnabled(trig) then
		call BJDebugMsg("trigger is alive and enabled")
	else
		call EnableTrigger(trig)
		if IsTriggerEnabled(trig) then
			call DisableTrigger(trig)
			call BJDebugMsg("trigger is disabled")
		else
			call BJDebugMsg("trigger has been destroyed")
		endif
	endif
endfunction

 

[Azlier]

Perfect! Inserting into system now...

 

[gameman]

There was my way to make a double click, but it was made with gui and was a leaker despite my best efforts. I Have a hard time transitioning to jass but more and more useful stuff keeps popping up.

 

[Jesus4Lyf]

I might have to take that TriggerExist function and use it in GTrigger at some point. o.o

 

[Azlier]

Perhaps, and there was a major problem with this. If you clicked one unit, and double clicked another real quick, the system failed to register. This has been fixed.

 

[Jesus4Lyf]

Try using Event instead of trigger chains. Trigger chains wasn't really for public release, but Event is. Event should prove to be more efficient and make your code easier to maintain.

(I just released it after realising this could become an epidemic of dodgey code and should be somewhat standardised instead.)

Feel free to internalise Event in your system by copying it in and making it private. That's up to you.

 

[Romek]

private function Actions takes nothing returns boolean
    
    globals
        private integer Id
        private unit Cu
    endglobals

I'm surprised that doesn't syntax. You should use globals like a normal, sane person though. If that gets fixed, then this system won't work anymore. Why do you do that anyway? Why not use locals?

 

[Azlier]

Because I'm too lazy to null a local. If you really want that "fixed", I'll do that while I'm installing multiple click support.

 

[Romek]

> Because I'm too lazy to null a local.
But you're diligent enough to add a whole globals block?
You could've even put those globals in the globals block at the top, if locals are an issue.

> I'll do that while I'm installing multiple click support.
Do so.

 

[Jesus4Lyf]

>while I'm installing multiple click support.
Don't bother. It's madness.

If you have a triple click event and a double click event, one of two options will occur:

Both will fire, making the triple click redundant.
OR
The double click will fire delayed, which is annoying for coding.

Unless we hear a useful case for triple click (where double click would fire too), I strongly suggest you drop it.

Oh, and Romek, I'm pretty sure he did the globals for efficiency too. It's invisible to external users so its ok in my opinion, but at least he could take the globals block out of the function for the reason you stated about "what if its 'fixed'".

 

[Azlier]

>Both will fire, making the triple click redundant
It is indeed madness. But if TriggerHappy is copying my system and making it worse just for this madness, I'll just have to beat him.

>Both will fire, making the triple click redundant.
Not if I have 12 timers and triggers for each event. Madness.

>Unless we hear a useful case for triple click
Yeah, I want to hear one too.

>I'm pretty sure he did the globals for efficiency too
This is correct.

 

[Romek]

> I'm pretty sure he did the globals for efficiency too
Yeah, I realised. Hence why I provided a solution which avoids getting rid of them altogether.

Take that global block outside of the function, and this'll be approved.

 

[Azlier]

Alrighty then. Doesn't support multiple clicks (God knows who needs those). Can't change the demo map right now, though. I don't have NewGen here.

 

[Romek]

Approved.

 

[Azlier]

Thank ye.

Updated the demo map and shaved off a whopping 1 line of code. Thanks to Jesus4Lyf for the idea to shave off that 1 line.