• Jun (14/14) Nov 23 2012 I found some codes write
• Mike Parker (7/21) Nov 23 2012 FYI, you could even do this:
• Rob T (9/11) Nov 23 2012 More info here

"Jun" <ltalbsjea gmail.com> writes:

I found some codes write

toStringz(myString)

as

mystring.toStringz

So I tested this code myself and it worked.

int pow2(int i)
{
     return i*i;
}
int myint = 5;
int otherint = myint.pow2;
assert(otherint == 25);

I've never seen any documentation about this behaviour. I think 
it's a good feature, but I'm a bit confused.

"Mike Parker" <aldacron gmail.com> writes:

On Friday, 23 November 2012 at 14:08:05 UTC, Jun wrote:
 I found some codes write

 toStringz(myString)

 as

 mystring.toStringz

 So I tested this code myself and it worked.

 int pow2(int i)
 {
     return i*i;
 }
 int myint = 5;
 int otherint = myint.pow2;
 assert(otherint == 25);


FYI, you could even do this:

int otherint = 5.pow2;


 I've never seen any documentation about this behaviour. I think 
 it's a good feature, but I'm a bit confused.


It's called Universal Function Call Syntax (UFCS). The compiler 
essentially rewrites something.function to function(something), 
unless "something" is a struct or class with a method named 
"function", in which case it calls the method instead.

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

On Friday, 23 November 2012 at 14:08:05 UTC, Jun wrote:
 I've never seen any documentation about this behaviour. I think 
 it's a good feature, but I'm a bit confused.

More info here
http://www.drdobbs.com/cpp/uniform-function-call-syntax/232700394

UFCS is relatively new and has not been included in the language 
documentation yet. It was added to the language only a few months 
ago with the dmd 2.059 release.

IMO it should be in the docs by now, and I think that's a problem 
area being worked (I hope).

--rt