Snippet GUI-Friendly Damage Detection

[DuelPlayer]

If all spells in your map deal damage by triggers, you can temporarily set a boolean variable to true when dealing spell damage. Therefore, if the boolean is false when a unit takes damage, it must have come from an attack.

Otherwise, if you are comfortable triggering all buff-on-attack abilities, every unit can be given a hidden buff-on-attack ability which will let you determine whether the damage came from an attack by checking whether the damaged unit has that buff.

As far as I'm aware, these are the only reliable methods for discerning attack damage apart from spell damage in WC3.

This methods arent good...
And i tried once, i set the damage of cripple to 0
But the unit get knockback which is another trigger when it is casted

 

[Weep]

I've been using the GDD damage detection in my spell, and referencing the GDD_DamageSource as "triggering unit" in my hashtable Key Handle. I didn't think it would work because the custom variable cannot be referenced in a Handle, but it seems to being working just fine.

Is this a bug, or was I misinformed, or what?

Well, not being able to use a variable in a hashtable seems to be a World Editor shortcoming.

It's not a bug that you can use Event Response - Triggering Unit in place of GDD_DamagedUnit (though unless you want it for a hashtable or after a wait, using the variable might be marginally faster ); I just haven't documented it. Also, FYI, Event Response - Damaged Unit doesn't work with GDD.

 

[Inflicted]

nvm figured it out.

 

[Weep]

Updated to 1.2.1, with two user-configuration functions, minor code cleanup, and expanded documentation. Happy users of 1.2.0 don't really need to update.

 

[mapguy]

I just have to copy + paste it in trigger editor?

that jass code box?
just it?

how to enable my locusts to DEAL damage.

 

[HeX.16]

Download the test map and the instructions are REALLY easy.

 

[hgkjfhfdsj]

question(s) (more learning purposes.)

function GDD_LeaveMap takes nothing returns boolean
    local boolean cached = bj_slotControlUsed[5063]
    if(udg_GDD__Integers[2] < 64) then
        set udg_GDD__Integers[2] = udg_GDD__Integers[2]+1
    else
        set bj_slotControlUsed[5063] = true
        call ForGroup(udg_GDD__LeftMapGroup, function GDD_GroupRefresh)
        set udg_GDD__Integers[2] = 0
    endif
    call GroupAddUnit(udg_GDD__LeftMapGroup, GetFilterUnit())
    set bj_slotControlUsed[5063] = cached
    return false
endfunction

why set cached as bj_slotControlUsed instead of using bj_slotControlUsed directly, ie

function GDD_GroupRefresh takes nothing returns nothing
    if (bj_slotControlUsed[5063] == true) then
        call GroupClear(udg_GDD__LeftMapGroup)
        set bj_slotControlUsed[5063] = false
    endif
    call GroupAddUnit(udg_GDD__LeftMapGroup, GetEnumUnit())
endfunction
// . .
function GDD_LeaveMap takes nothing returns boolean
    //local boolean cached = bj_slotControlUsed[5063]
    if(udg_GDD__Integers[2] < 64) then
        set udg_GDD__Integers[2] = udg_GDD__Integers[2]+1
    else
        set bj_slotControlUsed[5063] = true
        call ForGroup(udg_GDD__LeftMapGroup, function GDD_GroupRefresh)
        set udg_GDD__Integers[2] = 0
    endif
    call GroupAddUnit(udg_GDD__LeftMapGroup, GetFilterUnit())
    //set bj_slotControlUsed[5063] = cached  //why set it to original bj_slotControlUsed?
    return false
endfunction

ty

 

[Weep]

It's because I'm co-opting a Blizzard global array instead of having to create my own global variable (it only matters without vJASS, which gives the freedom to declare globals in the code), and because I don't know whether some other system or spell might happen to use that exact same array variable (unlikely, but you never know ), I make sure to save its original value and put it back when I'm done.

 

[the Immortal]

Setting a (surely empty) array's 5063rd member should allocate at least 5062 (surely empty) booleans in the game's memory from what I understand? Since arrays are consecutive in memory and blahblahblahboringtalk, my point is: you reset it anyway, so simply use the 0 index of the array. It is highly improbable that someone could use a 'variable changed value' event on this array member, so why waste these few bytes of memory creating 5k useless booleans! =D


It's a pointless optimization.. but hey, it is something!

 

[Weep]

Setting a (surely empty) array's 5063rd member should allocate at least 5062 (surely empty) booleans in the game's memory from what I understand?

As far as I've read other people mention, WC3 allocates the entire array when declared, so this shouldn't make any difference. [edit] I guess it does...

 

[Opa]

and this really doesnt leak?

 

[Laiev]

100% leakless because this allow to add the event just one time, not 14923849273984 times for the same unit

 

[Sevion]

There is no leakless.

But as far as you're concerned (and any users of this system) you'll be fine.

 

[Opa]

100% leakless because this allow to add the event just one time, not 14923849273984 times for the same unit

nice system is nice...

 

[Sajin]

Would this system have a problem is the damage source and damaged unit are owned by the same player?

 

[Weep]

Would this system have a problem is the damage source and damaged unit are owned by the same player?

Nope, that works fine.

 

[Sajin]

ok thank you

 

[dianamod27]

Are there any conditions I can use to determine whether the damage taken by GDD_DamagedUnit is caused by a spell?
The problem is that I want certain triggers to fire only when a unit takes damage from a regular attack. I already have direct damage spells covered with

• Or - Any (Conditions) are true
  • Conditions
    • (Current order of GDD_DamageSource) Equal to (Order(<Empty String>))
    • (Current order of GDD_DamageSource) Equal to (Order(smart))

but said triggers still fire when a unit takes periodic damage from Acid Bomb and similar spells.
I have this feeling that I'm overlooking something really obvious here but I just can't figure it out.

 

[Ayanami]

Are there any conditions I can use to determine whether the damage taken by GDD_DamagedUnit is caused by a spell?
The problem is that I want certain triggers to fire only when a unit takes damage from a regular attack. I already have direct damage spells covered with

• Or - Any (Conditions) are true
  • Conditions
    • (Current order of GDD_DamageSource) Equal to (Order(<Empty String>))
    • (Current order of GDD_DamageSource) Equal to (Order(smart))

but said triggers still fire when a unit takes periodic damage from Acid Bomb and similar spells.
I have this feeling that I'm overlooking something really obvious here but I just can't figure it out.

I believe this System is only for detecting damage, and not damage type. You'd need to make your own damage system. For example, if you trigger all your spell-based damage, you can easily differentiate a physical attack and a spell attack.

 

[dianamod27]

I believe this System is only for detecting damage, and not damage type. You'd need to make your own damage system. For example, if you trigger all your spell-based damage, you can easily differentiate a physical attack and a spell attack.

I was thinking about that as a measure of last resort but I guess I have to.