Or/and functions

[XeRo13g]

if udg_Boolean[1]==true and (udg_Int[2]==26 or udg_Int[2]==27 or udg_udg_Int[2]==28 or udg_udg_Int[2]==29 or udg_udg_Int[2]==31) then
    do something
    endif

Issue 1: Can boolean be used with integers in the same -IF- line? (I've tested it and it seems to work as far as I understand, yet there is a warning about comparing variables with different primitive types)

Issue 2: or/and functions are supposed to leak since they return a Boolexpr. (I think Blizzard claims that Boolexpr are now fixed with the latest patch). Does someone has knowledge on this?

Should I stop using that kind of -if- formats and go with something like this?

function Trig_Untitled_Trigger_001_Func001Func002C takes nothing returns boolean
    if ( ( udg_Int[2]== 26 ) ) then
        return true
    endif
    if ( ( udg_Int[2] == 27 ) ) then
        return true
    endif
    if ( ( udg_Int[2] == 28 ) ) then
        return true
    endif
    if ( ( udg_Int[2] == 29 ) ) then
        return true
    endif
    if ( ( udg_Int[2]== 31 ) ) then
        return true
    endif
    return false
endfunction

function Trig_Untitled_Trigger_001_Func001C takes nothing returns boolean
    if ( not ( udg_Boolean[1] == true ) ) then
        return false
    endif
    if ( not Trig_Untitled_Trigger_001_Func001Func002C() ) then
        return false
    endif
    return true
endfunction

function Trig_Untitled_Trigger_001_Actions takes nothing returns nothing
    if ( Trig_Untitled_Trigger_001_Func001C() ) then
    else
    endif
endfunction

Which looks terrible

 

[Dirac]

Issue 1: yes you can, as long as it returns boolean
Issue 2: [ljass]Or()[/ljass] and [ljass]And()[/ljass] do cause issues because they have to be destroyed, and the boolexpr destroy native is broken.

as for your code it would be a lot easier just to go like this
[ljass]if (udg_Int[2]>=26 and udg_Int[2]<=31) then[/ljass]

 

[XeRo13g]

Issue 1: yes you can, as long as it returns boolean

You mean the function to return boolean?
I've been using that kind of code in the body of the main trigger where if takes nothing and returns nothing...

 

[Dirac]

If anything returns boolean

[ljass]func() -> true[/ljass]
[ljass]var -> true[/ljass]
[ljass]1==1 -> true[/ljass]

 

[XeRo13g]

If "and/or" still cause issues, how should I go about?

The second code snippet is just how the editor converts into text and can get seriously messy when you use a lot of conditions.

PS. if (udg_Int[2]>=26 and udg_Int[2]<=31) then
Cannot but used because in this case udg_Int[2]!=30
...and it still uses "and"

 

[Bribe]

The and/or/not keywords do not cause issues at all. The issue that he mentioned is with three completely different things: And, Or and Not, same words but instead of keywords they are functions that take code arguments.

For that, you'd use [ljass]And(function ConditionFunc1, function ConditionFunc2)[/ljass], but what you are referring to here is just [ljass]condition1 and condition2[/ljass]

 

[Dirac]

that [ljass]and[/ljass] causes no issues, it's very different from the native [ljass]And()[/ljass] which returns boolexpr, same thing with [ljass]or[/ljass]. I think that clears a lot of your doubts.

[ljass]if udg_Int[2]>=26 and udg_Int[2]<=31 and udg_Int[2]!=30 then[/ljass]

 

[XeRo13g]

if udg_Boolean[1]==true and (udg_Int[2]==26 or udg_Int[2]==27 or udg_udg_Int[2]==28 or udg_udg_Int[2]==29 or udg_udg_Int[2]==31) then
    do something
    endif

So, since this is using the keywords, there is no need to change anything... (Let alone slightly improving the code as Dirac indicated)... I can use as many or/and as I like without leaks, yes?

 

[Bribe]

Absolutely, you got nothing to fear

Just a little misunderstanding you had was all. It happens.

 

[XeRo13g]

Thanks guys, +rep