On 9/14/22 21:20, Rikki Cattermole wrote:
> Recently Walter has been wanting to remove language features that don't need to be kept to simplify the language.
> 
> It seems everybody agrees that binary literals are not a worthy candidate for removal, but one thing that I kept thinking about was @property. As far as I know, hardly anyone actually uses its semantic behaviors, and the ones that it does have which are incomplete are special cases that don't benefit us all that much.
> 
> I was going to post this yesterday, but it turns out I am not alone in thinking this can be removed safely (can be swapped to a UDA to prevent code breakage without any language additions).
> 
> So, who uses @property semantics?
> 
> https://dlang.org/spec/function.html#property-functions

I am pretty sure many (most?) people who rely on those semantics are oblivious to it because @property is in third-party code, so not much will come from this thread.

Anyway, if you want to break everyone's code over this why not just fix @property so it becomes actually useful instead?

On 16/09/2022 6:16 AM, Timon Gehr wrote:
> Anyway, if you want to break everyone's code over this why not just fix @property so it becomes actually useful instead?

I don't, but if something has to go... Better it be something that people out right don't intend to use or know what it does do.

On 9/26/22 11:50, Dom DiSc wrote:

> if x is a property
> (having a getter and a setter), ++x should work

That came up a couple of times on this thread.

Sounds easy to my ears. Is there a blocker for the implementation?

Ali

On Monday, 26 September 2022 at 18:51:41 UTC, Ali Çehreli wrote:
> On 9/26/22 11:50, Dom DiSc wrote:
>
> > if x is a property
> > (having a getter and a setter), ++x should work
>
> That came up a couple of times on this thread.
>
> Sounds easy to my ears. Is there a blocker for the implementation?
>
> Ali

I don't think so. Would be an easy lowering.
Forbidding to take the address is more controversial, but I don't understand why. It would be ambiguous (should it return a value-pointer or a function-pointer?). If you want something you can take the address of, make it a function or a value, not a property.

On 26.09.22 20:50, Dom DiSc wrote:
> On Wednesday, 14 September 2022 at 19:20:04 UTC, Rikki Cattermole wrote:
>> Recently Walter has been wanting to remove language features that don't need to be kept to simplify the language.
>>
>> So, who uses @property semantics?
>>
>> https://dlang.org/spec/function.html#property-functions
> 
> I do - or I would if only they would work - e.g. if x is a property (having a getter and a setter), ++x should work, &x should be an error etc. If it has only a getter or a setter, ++x should not work, but provide a meaningful error message instead of the garbage we get now.
> It would be easy to fix this. Makes me really sad.

Also see: https://wiki.dlang.org/DIP24