IsUnitDeadBX

Yes. it's so simple but effective =) (i blame me what i just found it out not much earlier) ok here we go xD
yes the sense of this is it's 100% bullet proofed if unit is dead it will defenetly return true.
lol yes it's so simple... aka GetUnitStata(u,UNIT_STATE_LIFE)<.406 say bb (cause it may be also tricked for example by setting unit life after it's dead etc.)
gl hf

You make me lol, how come nobody brought it up xD

-Av3n

what ever... i havent see such way of detecting if unit is dead in any map.

unless someone has a problem with this method I will probably start using it over the GetWidgetLife method since it seems more sensible and accurate.

This method will fail if the unit has decayed.

Strilanc this will fail if this unit doesnt exist in the game =) if you need to do operation's with not existing units...

How can you possibly know the state of units over time without fully monitoring it? (Not worth the time as far as I'm concerned) That's reason enough for me to not use it.

There are situations, though rare, where you can end up not checking on a unit until after it decayed. A proper function should handle border cases (IsUnitDeadBJ returns true when they're decayed, and you do not want to do worse than a BJ).

Actually, for me, it's those border cases that wind up being the most important. If a unit with the artillery attack type explodes a unit and this method fails, it would be a catastrophe.

Because it doesn't really work either.

If you want a bullet proof method, just detect EVENT_UNIT_DEAD and mark units accordingly. Else I am gonna stick to GetWidgetLife or whatever, both checking the life and this one are equally ineffective.
--
I had thousands of issues with death detection when making a corpse explode spell, my conclussion there was that IsUnitType(u,UNIT_TYPE_DEAD) was ... the same to the unit life check. I don't think they changed it in a recent patch, but it is a possibility, it would be nice if blizz documented patch changes more.


--
Making this a function is absurd, it is easier to understand what it does without reading that whole BX stuff and you save a function call.

On a related note, here is a function you may find useful:

That'll return true on a null object.

GetWidgetLife has never been a problem for me (I strangely enough never set a dead unit's health up).

Good point, but the appropriate value for null is to be decayed, not alive. You will treat null as a decayed unit more often than as an active unit.

On the other hand, some people like to follow some sort of "always return 0/false/null for weird arguments" rule.

Ficksed.

You would need to constantly keep a jass reference to the unit, else the id would get recycled.

I guess this is fixed if you stop using I2H. But there might be other cases?