• Rob T (17/17) Jun 11 2013 struct S
• bearophile (5/7) Jun 12 2013 Probably no one thought on this case. Why do you think it's
• Rob T (23/30) Jun 12 2013 If we're to use UFCS as a means to externally extend classes and

"Rob T" <alanb ucora.com> writes:

struct S
{
}

void f(S)
{
}

void main()
{
   S s;
   with (s)
   {
      f(); // compiler error
   }
}

Rather disappointing that this fails. Anyone know if this is an 
expected limitation of UFCS or a bug?

--rt

"bearophile" <bearophileHUGS lycos.com> writes:

Rob T:

 Rather disappointing that this fails. Anyone know if this is an 
 expected limitation of UFCS or a bug?

Probably no one thought on this case. Why do you think it's 
useful?

Bye,
bearophile

"Rob T" <alanb ucora.com> writes:

On Wednesday, 12 June 2013 at 11:05:22 UTC, bearophile wrote:
 Rob T:

 Rather disappointing that this fails. Anyone know if this is 
 an expected limitation of UFCS or a bug?

 Probably no one thought on this case. Why do you think it's 
 useful?

 Bye,
 bearophile

If we're to use UFCS as a means to externally extend classes and 
structs, then UFCS should work as an extension under a with 
statement, otherwise UFCS is not living up to its full potential.

struct S
{
    void a(){ ... }
    void b(){ ... }
    void d(){ ... }
}

void c(S s){ ... }

void main()
{
   with (s)
   {
      a;
      b;
      c; // should be able to infer s.c just as it can infer s.b, 
etc
      d;
   }
}

--rt