Re: Complicated Types: Prefer “alias this” Over “alias” For Easier-To-Read Error Messages

[Nick Sabalausky (Abscissa)]

On 05/21/2018 01:43 PM, Steven Schveighoffer wrote:
> Nice idea, I wonder if the compiler couldn't do this automatically with 
> alias, however. It already does this in some cases (e.g. string instead 
> of immutable(char)[]).
> 
> This would help solve the problem that error messages aren't going to 
> get better when you pass it into something that only accepts the aliased 
> type. However, for the most part, these are non-template functions anyway.

That was discussed and rejected some years ago. IIRC, Walter just didn't 
like it.

I'd certainly be in favor of it though. But in lieu of that...