Owned pointer in Rust is a good design IMO, how about D
equivalent?

On 06/08/2013 06:26, Elvis wrote:
> Owned pointer in Rust is a good design IMO, how about D
> equivalent?

I think this is about the closest ATM:
http://dlang.org/phobos/std_typecons.html#.Unique

On Tuesday, 6 August 2013 at 16:32:43 UTC, Nick Treleaven wrote:
> On 06/08/2013 06:26, Elvis wrote:
>> Owned pointer in Rust is a good design IMO, how about D
>> equivalent?
>
> I think this is about the closest ATM:
> http://dlang.org/phobos/std_typecons.html#.Unique

would be good to have it as language feature like scoped reference. this clear code (as from c++11 experience abundant std::unique_ptr<T> get me irritated). And of course unique idiom  (ownership) get D usable for realtime (removed GC)

On Friday, 9 August 2013 at 18:16:57 UTC, Ruslan Mullakhmetov wrote:
> On Tuesday, 6 August 2013 at 16:32:43 UTC, Nick Treleaven wrote:
>> On 06/08/2013 06:26, Elvis wrote:
>>> Owned pointer in Rust is a good design IMO, how about D
>>> equivalent?
>>
>> I think this is about the closest ATM:
>> http://dlang.org/phobos/std_typecons.html#.Unique
>
> would be good to have it as language feature like scoped reference. this clear code (as from c++11 experience abundant std::unique_ptr<T> get me irritated). And of course unique idiom  (ownership) get D usable for realtime (removed GC)

While I'm convinced that unique pointers / types are a important concept, especially if combined with immutable. (What's transitive unique/immutable can be cast to immutable .. ), I don't think that just adding a new type modifier will lead D anywhere.

On Friday, 9 August 2013 at 19:05:26 UTC, Tobias Pankrath wrote:
> On Friday, 9 August 2013 at 18:16:57 UTC, Ruslan Mullakhmetov wrote:
>> On Tuesday, 6 August 2013 at 16:32:43 UTC, Nick Treleaven wrote:
>>> On 06/08/2013 06:26, Elvis wrote:
>>>> Owned pointer in Rust is a good design IMO, how about D
>>>> equivalent?
>>>
>>> I think this is about the closest ATM:
>>> http://dlang.org/phobos/std_typecons.html#.Unique
>>
>> would be good to have it as language feature like scoped reference. this clear code (as from c++11 experience abundant std::unique_ptr<T> get me irritated). And of course unique idiom  (ownership) get D usable for realtime (removed GC)
>
> While I'm convinced that unique pointers / types are a important concept, especially if combined with immutable. (What's transitive unique/immutable can be cast to immutable .. ), I don't think that just adding a new type modifier will lead D anywhere.

No, I didn't assert that. I just proposed a synthetic sugar =). Unique pointer/types are already in D (at the library level).