D needs...

[Andrei Alexandrescu via Digitalmars-d-announce]

On 5/12/15 3:09 AM, Namespace wrote:
> On Tuesday, 12 May 2015 at 09:31:13 UTC, Marc Schütz wrote:
>> On Monday, 11 May 2015 at 17:49:32 UTC, Namespace wrote:
>>> I've read DIP69 and there were a few lines about scope ref. But I'm
>>> not sure whether I understand everything correct (because there is no
>>> concrete application example for my case): will DIP69 create the
>>> possibility to pass rvalues and lvalues alike to a function without
>>> the abuse of templates? Or did I got that wrong?
>>
>> DIP69 is more or less dead, we're going with a variation of DIP25
>> (although that one still has holes).
>>
>> As for rvalue ref, it would enable them safely, but it is not a part
>> of that DIP (or DIP69, AFAICS).
>
> And I thought for a moment, we have a solution in sight... :)

Knee-jerk response: if no "return" attribute on a function it should be 
safe to bind rvalues to ref parameters. Of course that's impractical as 
a default so explicit "auto ref" would be needed. -- Andrei