WHY??

[quasi cedar]

why does odin have nullable pointers instead of option type, why?
and cant i enforce it somehow.
It could've prevented the most common bug by having an optional

[fossil summit]

https://www.gingerbill.org/article/2026/01/02/was-it-really-a-billion-dollar-mistake/
tl;dr nil pointers are the least of your concerns in terms of things that can go wrong when it comes to pointers 🙂

[proper crag]

Maybe is there if you want that behaviour.

[quasi cedar]

but still it makes it not clear if the value returned by this fn is guaranteed to have a valid reference and checking every single one of them is annoying

[woven lily]

most of nil pointers are bugs anyway

Odin also has the -> (res: ^T, ok: bool) pattern which is more ergonomic than checking the pointer

also, just don't return pointers

it's not that hard

[proper crag]

would you not being doing that with an optional type anyways? either you check it, or something does for you.

[quasi cedar]

the thing is if its not wraped in optional i can know for a fact that its valid if its not nullable

[quasi cedar]

this is exactly why pointers shouldn't be nullable by default

[woven lily]

nil pointer dereference is the easiest bug to spot and fix

[fossil summit]

In any case, the language's semantics pretty much don't allow them to not possibly be nil. ZII/zero is initialized; everything is default-initialized to 0

[woven lily]

having non nil pointer will literally solve nothing and make the language much more clunky to use

(since odin is not designed to be memory safe)

i think you are being unfair in using a language not even one day and then throwing yourself at the debate about its core design

[proper crag]

that is what Maybe(^T) gives. unions who only have a pointer in them drop the tag, and the tag check becomes a nil pointer check.

-# it still might not be valid memory, but that's besides the point.

[quasi cedar]

wait does that mean every language that allows zero initialization has a secret null value

[quasi cedar]

I think you are right i will read the article then