Dne st 23. 12. 2020 1:00 uživatel Steven Schveighoffer via Digitalmars-d-learn <digitalmars-d-learn@puremagic.com> napsal:
On 12/22/20 5:44 PM, Daniel Kozak wrote:
> On Tue, Dec 22, 2020 at 10:15 PM Andre Pany via Digitalmars-d-learn
> <digitalmars-d-learn@puremagic.com
> <mailto:digitalmars-d-learn@puremagic.com>> wrote:
>
>     Hi,
>
>     I am really confused, why is this valid:
>     void sample(string[string] s = string[string].init){}
>
>     while this causes syntax errors?
>
>     void sample_invalid1(double[string] s = double[string].init){}
>     void sample_invalid2(int[int] s = int[int].init){}
>
>     Kind regards
>     André
>
>
> As has been said this is an oddity in the grammar. But why would anyone
> need to use this anyway?
>
>    void sample_invalid2(int[int] s = int[int].init){}
>
> seems really awful to me anyway.

Yeah:

void sample_valid(int[int] s = null)

-Steve 

Yes AA.init is null per doc. 

https://dlang.org/spec/hash-map.html#construction_and_ref_semantic