Question about the JASS Compiler

[Arhowk]

so i had an error in my script and i noticed this

function sa__QuestInteractListener_enable takes nothing returns boolean
local integer this=f__arg_this
local player p=f__arg_player1
            call s__Table__setindex(s__QuestInteractListener_b[this],GetPlayerId(p), 1)
            call PingMinimap(GetWidgetX(s__QuestInteractListener_interact[this]), GetWidgetY(s__QuestInteractListener_interact[this]), 5)
            call ShowUnit(s__QuestInteractListener_exc[this], s__Table__getindex(s__QuestInteractListener_b[this],GetPlayerId(GetLocalPlayer())) != 99)
            call TimedTextBoolean(15 , s__Table__getindex(s__QuestInteractListener_b[this],GetPlayerId(GetLocalPlayer())) != 99 , Quest___beforeQuestNameAccepted + s__QuestInteractListener_questName[this] + Quest___afterQuestNameAccepted)
set f__result_integer= this
   return true
endfunction

wtf does it leak p?

 

[DysfunctionaI]

I've heard 2 things about that... 1 being that it leaks, and 2 that it doesn't because it points to a static player handle that will be used the rest of the game. Dunno, but that doesn't even need a local player var, since it could just be local integer p = GetPlayerId( f__arg_player1 ) and then replaced in the call to s__Table__setindex(). Also it uses GetPlayerId( GetLocalPlayer() ) twice :O

 

[Nestharus]

You could always move to a combo of SharpCraft and JASS. After all, SharpCraft works just fine on b.net : \.

You could do all of the triggers in JASS for the events and everything else in SharpCraft.

I'm starting to think that that's the way to go.

 

[Maker]

The p does not leak.

 

[Malhorne]

Only handle that get destroy have to be nulled.
So basically player can't be destroyed.
Even if I prefer to null them it shouldn't leak.