JassHelper 0.A.2.B - A vJass and Zinc 2 Jass compiler

[Lord_Executor]

Is there a possibility to overload functions/methods?

JASS:

     function test takes integer i returns nothing
         call BJDebugMsg(I2S(i))
     endfunction
     function test takes string  s returns nothing overload
         call BJDebugMsg(s)
     endfunction
...
call test(3)
call test("test")
...

[Vexorian]

not yet.

[akolyt0r]

Originally Posted by Vexorian

not yet.

You will really add this ??
Thats quite a difficult job ...
think of:

JASS:

function test takes integer i returns nothing
function test takes somestruct o returns nothing
function test takes someotherstruct o returns nothing

[Vexorian]

not that, but something like that.

[Frozenhelfire]

Originally Posted by Vexorian

updating to one of the most recent versions.

Hmmm, this is not the case for me. Just to make sure, I copy the three files in the executable folder into the jasshelper folder in newgen? That is what I've been doing. I'll see if microsoft will fix it :D
Although if I try opening it directly, it says bin\SFmpq.dll is missing even though it is there. This may be the problem and I'll look at my paths once I get home.

[Earth-Fury]

JH error: "constant structs not supported"
on the line:
constant ARGB BonusBoard___COLOR_HEADINGS = 0xFFFF00
from the source:
private constant ARGB COLOR_HEADINGS = 0xFFFF00
Step 2 (Structs) is the phase the error happens in.

I'm guessing that's an error. If not, then consider this a feature request...

[Vexorian]

It's a feature request.

allow constant for array structs. Then allow constant for the time you can actually .create in globals.

Since ARGB fails without it, I'll do it in the next version.

[gekko]

Even with [forcemethodevaluate] in jasshelper.conf, this code compiled without errors:

JASS:

struct A
    method a takes nothing returns nothing
        call .b()
    endmethod
    
    method b takes nothing returns nothing
        call BJDebugMsg("")
        call BJDebugMsg("")
    endmethod
endstruct

[Rising_Dusk]

Yeah, that should work fine. Why wouldn't it?

[XieLong]

I think i have another feature request ._.

As there is inheritance of structs why isn't there also the keyword "protected"? I think it's really necessary cause otherwise we would have to declare all variables used by child structs as "public"...

[Vexorian]

If something is supposed to be used by code outside the struct, public is precisely the keyword to use.

I have always headed towards avoiding the use of implementation inheritance.

[XieLong]

Originally Posted by Vexorian

I have always headed towards avoiding the use of implementation inheritance.

Yes, I read this... but why do you offer interfaces and extensible structs if you do not want to aim inheritance? I don't get the point :\

See, when one struct extends another struct, all members of the parent struct are in some way not outside of the child struct, are they?

[Vexorian]

interfaces are interface inheritance (not implementation inheritance). Extensible structs were intended as a form of composition.

Anyway, the point of encapsullation is to protect code from outsiders. protected does not do that better than public, it just turns out to be a way to force implementation inheritance which actually breaks encapsulation...

[gekko]

Originally Posted by Rising_Dusk

Yeah, that should work fine. Why wouldn't it?

because I called a method before defining it which I guess would compile to an implicit evaluate which should be disabled with the setting in jasshelper.conf. So bug or me == fail?

[Bobo_The_Kodo]

Yes it's a bug. It compiles to using TriggerEvaluate()

><