The story of 0, 1 and null

[bowser499]

Contributions from SirNikolas and Volchachka were there aswell.
I've decided to compare NULL with integers. Here are the results:

// null<=0   -   true
// null==0   -   true
// null>=0   -   true
// null>0    -   false
// null<0    -   false
// null!=0   -   false

// null<=1   -   true
// null==1   -   false
// null>=1   -   false
// null>1    -   false
// null<1    -   true
// null!=1   -   true

The code for testing:

function TestByteCodingBug takes nothing returns nothing
    local string boolDesc = "false"
    if null <= 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null<=0   -   "+boolDesc)
    
    set boolDesc = "false"
    if null == 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null==0   -   "+boolDesc)
    
    set boolDesc = "false"
    if null >= 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null>=0   -   "+boolDesc)
    
    set boolDesc = "false"
    if null > 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null>0   -   "+boolDesc)
    
    set boolDesc = "false"
    if null < 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null<0   -   "+boolDesc)
    
    set boolDesc = "false"
    if null != 0 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null!=0   -   "+boolDesc)
    
    call BJDebugMsg("============================")
    
    set boolDesc = "false"
    if null <= 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null<=1   -   "+boolDesc)
    
    set boolDesc = "false"
    if null == 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null==1   -   "+boolDesc)
    
    set boolDesc = "false"
    if null >= 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null>=1   -   "+boolDesc)
    
    set boolDesc = "false"
    if null > 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null>1   -   "+boolDesc)
    
    set boolDesc = "false"
    if null < 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null<1   -   "+boolDesc)
    
    set boolDesc = "false"
    if null != 1 then
        set boolDesc = "true"
    endif
    call BJDebugMsg("null!=1   -   "+boolDesc)
endfunction

What can it be?

 

[Magtheridon96]

This is perfectly consistent if we assume null to be 0 :v
Try flipping the operands.

 

[bowser499]

// 0<=null   -   true
// 0==null   -   true
// 0>=null   -   true
// 0>null   -   false
// 0<null   -   false
// 0!=null   -   false

// 1<=null   -   false
// 1==null   -   false
// 1>=null   -   true
// 1>null   -   true
// 1<null   -   false
// 1!=null   -   true

Here.

 

[Magtheridon96]

Still consistent with null == 0 :/

 

[PurgeandFire]

Fun fact, null's definition relates to being associated with zero:
https://www.google.com/#q=define+null

Same for nil (just another term for 0). And in a lot of programming languages, null/nil is either equivalent or interchangeable with 0. So it is natural that null==0.

 

[muzzel]

Null is 0, except that its not of integer type.

But nice that it took 3 people to realize that 0 < 1, great job

Russians \o/

 

[Adiktuz]

yeah... one of the earliest things I learned in programming is that null is equivalent to 0...

null = nothing
0 = nothing

by one of the mathematical laws, if a = b and b = c then a = c

so null = nothing = 0

 

[bowser499]

null = void in C++?
Why function a takes null returns null won't compile then? XD

 

[Adiktuz]

not really... you can only convert/compare values if the target type has an equivalent for it...

null to 0 conversion is simply a value to integer value conversion... since integer type has an equivalent for null, the conversion/comparison proceeds

and since void doesn't have an equivalent for null (I don't think it even has any values inside it), then the conversion cannot proceed...

as for the functions take and returns part, you put types there not values... which is probably why it won't compile (since again, null is not a type)...

//doing this
function X takes null returns null

//is like doing this
function Y takes "RAWR" returns "RAWR"

//or this

function Z takes 0 returns 0

//All of them won't compile

This is my guess...

 

[Almia]

null is 0.
Reconstruct the results replacing all nulls with 0, and you will get the same result XD

 

[Daffa]

I believe null is for non integer related stuffs.
As null = 0, then null will have 0 value as integer (and real?), but others like handle can't accept 0 as value (because handle aren't numbers), thus null is the valid replacement. That's what I can think for null = 0.
anyway, this topic really interest me. Better subscribe before lost track.

 

[Malhorne]

Null is for non-number vars.

 

[muzzel]

yeah... one of the earliest things I learned in programming is that null is equivalent to 0...

null = nothing
0 = nothing

by one of the mathematical laws, if a = b and b = c then a = c

so null = nothing = 0

null is not always equivalent to 0, it depends on the language

Mathematical laws dont apply to programming, especially when it comes to pointers and stuff. "=" in the mathematical understanding is an equivalence relation (the property you refered to is known as "transitivity"), however that relation is defined for real numbers which are very different from variables that are used in programming in the way that variables have both a value and a type.

And being able to compare different datatypes doesnt mean they are fully compatible in all operations. Example: You can compare (<,>,=,!=) integers and reals, but you cannot use reals as arguments for a function that expects integers.

 

[Adiktuz]

that's why I said it is only applied if both types have an equivalent value for the comparison... like you can compare 123 to 123.0 but not A to 123...

And I never said anything about using them in operations outside of comparisons...

 

[Luorax]

null is not always equivalent to 0, it depends on the language

Or rather, the implementation (null can be 0 in a language on platform A, but something else on platform B, for example). It's a special value after all - even though it's usually implemented as "0", it's a good practice to just stick to "null".