Whoever released named argument: fuck you. Sincerly.

[WebFreak001]

On Wednesday, 28 February 2024 at 23:28:58 UTC, deadalnix wrote:
> These constant syntax changes being rolled out at random just 
> screw up tooling.
>
> The time was waste using D because of the lack of named 
> argument was approximately none at all.
>
> The time we waste maintaining constant breakage because of 
> bullshit like this is enormous.
>
> At some point, this community will have to realize that a 
> programming language is meant to solve people's problems, not 
> cause them.
>
> I thought this was understood with recent messaging we got from 
> leadership, but apparently not.
>
> Well, for as long as this is the case, then it is impossible to 
> recommend D to anyone for anything serious.

I gotta say I really like the change and the tooling (libdparse 
at least) has already been adopted to support it for one year 
already. I have no idea why you even write about this now that it 
has been in the language for so long.

Honestly if new syntax coming from a DIP that has been accepted 
already in September of 2020 (over 3 years ago) and the 
implementation landing in the compiler one year ago is not enough 
time for you to adopt things, I don't know why you even update 
your compiler installation.

Being able to use named initialization through the constructors 
everywhere now is huge syntax sugar benefit for the wider 
community compared to the half-baked solution we had before with 
struct initializers.

I haven't had any issues with named arguments being implemented 
in my use-cases, except for some beta bugs that were ironed out 
in the implementation while it wasn't publicized that much yet. 
Old code not using them certainly wasn't affected and gradually 
introducing them to existing code bases also wasn't a problem for 
me.