Hi!
Sorry,if it already was discussed,but

import std.stdio;

struct Foo
{
	static void bar()
	{
		writeln("static");
	}
	void bar()
	{
		writeln("non-static");
	}
}

int main()
{
    Foo gun;
    gun.bar();//fails here
}

Is it all right,that compiler dosn't prefer non-static function in ths case?

On Sunday, 13 January 2013 at 15:58:56 UTC, Zhenya wrote:
> Hi!
> Sorry,if it already was discussed,but
>
> import std.stdio;
>
> struct Foo
> {
> 	static void bar()
> 	{
> 		writeln("static");
> 	}
> 	void bar()
> 	{
> 		writeln("non-static");
> 	}
> }
>
> int main()
> {
>     Foo gun;
>     gun.bar();//fails here
> }
>
> Is it all right,that compiler dosn't prefer non-static function in ths case?

Main should be void in this case.

Yes, it is a problem - dmd allows to call static functions on instance. When both match, it issues ambiguity error.

On Sunday, 13 January 2013 at 16:18:36 UTC, Maxim Fomin wrote:
> On Sunday, 13 January 2013 at 15:58:56 UTC, Zhenya wrote:
>> Hi!
>> Sorry,if it already was discussed,but
>>
>> import std.stdio;
>>
>> struct Foo
>> {
>> 	static void bar()
>> 	{
>> 		writeln("static");
>> 	}
>> 	void bar()
>> 	{
>> 		writeln("non-static");
>> 	}
>> }
>>
>> int main()
>> {
>>    Foo gun;
>>    gun.bar();//fails here
>> }
>>
>> Is it all right,that compiler dosn't prefer non-static function in ths case?
>
> Main should be void in this case.
>
> Yes, it is a problem - dmd allows to call static functions on instance. When both match, it issues ambiguity error.
Hmmm...So it will remain as it is?
It hurts me a little bit(

Maxim Fomin:

> dmd allows to call static functions on instance.

I think that's a D design mistake (and I think Jonathan Davis agrees with me), but Walter prefers the current behavour.

Bye,
bearophile

On Sunday, 13 January 2013 at 16:39:22 UTC, bearophile wrote:
> Maxim Fomin:
>
>> dmd allows to call static functions on instance.
>
> I think that's a D design mistake (and I think Jonathan Davis agrees with me), but Walter prefers the current behavour.
>
> Bye,
> bearophile

Maybe you could suggest some workaround?
for example I tried to use global function,but
in that case it never will be executed...

On Sunday, 13 January 2013 at 16:39:22 UTC, bearophile wrote:
> Maxim Fomin:
>
>> dmd allows to call static functions on instance.
>
> I think that's a D design mistake (and I think Jonathan Davis agrees with me), but Walter prefers the current behavour.
>
> Bye,
> bearophile

Then I agree with both of you, however I found this issue not very dramatical (comparing with other problems), probably because I haven't hit some problems with it.

On Sunday, 13 January 2013 at 17:04:21 UTC, Maxim Fomin wrote:
> On Sunday, 13 January 2013 at 16:39:22 UTC, bearophile wrote:
>> Maxim Fomin:
>>
>>> dmd allows to call static functions on instance.
>>
>> I think that's a D design mistake (and I think Jonathan Davis agrees with me), but Walter prefers the current behavour.
>>
>> Bye,
>> bearophile
>
> Then I agree with both of you, however I found this issue not very dramatical (comparing with other problems), probably because I haven't hit some problems with it.

I just was writting something like n-dimensional dispatcher,described in Andrei's Modern C++ design,and I needed to write classinfo analog by hands.

On Sunday, 13 January 2013 at 16:23:27 UTC, Zhenya wrote:
> On Sunday, 13 January 2013 at 16:18:36 UTC, Maxim Fomin wrote:
>> Yes, it is a problem - dmd allows to call static functions on instance. When both match, it issues ambiguity error.
> Hmmm...So it will remain as it is?
> It hurts me a little bit(

Frankly speaking I do not know - I keep an eye on D project for some period of time and remember how some features I considered stable were easily changed. If Walter is for current behavior - than it will remain as it is (unless Walter is overpersuaded).

What I would do is not writing code which is looking for problems.

BTW, take look at this thread http://forum.dlang.org/thread/pkodcsxwwehumhtkrlty@forum.dlang.org (posts 1-3).

Definitely this is a bad design. Even in PHP you can't call static functions from class instance. :-)

Having two functions with the same name and argument list within one class is a bad idea too.

On Sunday, 13 January 2013 at 17:17:54 UTC, Maxim Fomin wrote:
> On Sunday, 13 January 2013 at 16:23:27 UTC, Zhenya wrote:
>> On Sunday, 13 January 2013 at 16:18:36 UTC, Maxim Fomin wrote:
>>> Yes, it is a problem - dmd allows to call static functions on instance. When both match, it issues ambiguity error.
>> Hmmm...So it will remain as it is?
>> It hurts me a little bit(
>
> Frankly speaking I do not know - I keep an eye on D project for some period of time and remember how some features I considered stable were easily changed. If Walter is for current behavior - than it will remain as it is (unless Walter is overpersuaded).
>
> What I would do is not writing code which is looking for problems.
>
> BTW, take look at this thread http://forum.dlang.org/thread/pkodcsxwwehumhtkrlty@forum.dlang.org (posts 1-3).

Thank you.