• "Saaa" <empty needmail.com> Oct 26 2008
• Johan Granberg <lijat.meREM OVEgmail.com> Oct 26 2008
• "Saaa" <empty needmail.com> Oct 26 2008
• "Jarrett Billingsley" <jarrett.billingsley gmail.com> Oct 26 2008
• "Saaa" <empty needmail.com> Oct 26 2008

"Saaa" <empty needmail.com> writes:

?
main.d(118): function std.math.pow called with argument types:
 (double,uint)
matches both:
 std.math.pow(real,uint)
and:
 std.math.pow(real,real)

Also, I use pow(x,2U). Is this the correct function to use or is there a 
dedicated x*x function? 

Johan Granberg <lijat.meREM OVEgmail.com> writes:

Saaa wrote:

 ?
 main.d(118): function std.math.pow called with argument types:
  (double,uint)
 matches both:
  std.math.pow(real,uint)
 and:
  std.math.pow(real,real)
 
 Also, I use pow(x,2U). Is this the correct function to use or is there a
 dedicated x*x function?


you can solve that by casting x into real before the call
pow(cast(real)x,2U);
but in this case why not just simply write x*x? personally I think thats both
clearer and more efficient.

"Saaa" <empty needmail.com> writes:

"Johan Granberg" <lijat.meREM OVEgmail.com> wrote in message 
news:ge1lob$1b2i$1 digitalmars.com...
 Saaa wrote:

 ?
 main.d(118): function std.math.pow called with argument types:
  (double,uint)
 matches both:
  std.math.pow(real,uint)
 and:
  std.math.pow(real,real)

 Also, I use pow(x,2U). Is this the correct function to use or is there a
 dedicated x*x function?


 you can solve that by casting x into real before the call
 pow(cast(real)x,2U);


Yes, but why is this necessary?
Isn't there a preference to more matching arguments if all arguments could 
be implicitly converted?

 but in this case why not just simply write x*x? personally I think thats 
 both clearer and more efficient.

"Jarrett Billingsley" <jarrett.billingsley gmail.com> writes:

On Sun, Oct 26, 2008 at 8:47 AM, Saaa <empty needmail.com> wrote:
 you can solve that by casting x into real before the call
 pow(cast(real)x,2U);


 Yes, but why is this necessary?
 Isn't there a preference to more matching arguments if all arguments could
 be implicitly converted?


No.  See the section on function overloading on this page:
http://www.digitalmars.com/d/1.0/function.html

So for pow(x, 2u), where x is a double, the match level for both
(real, uint) and (real, real) is "match with implicit conversions."
Therefore, it's an error.

"Saaa" <empty needmail.com> writes:

Thanks :)

 No.  See the section on function overloading on this page:
 http://www.digitalmars.com/d/1.0/function.html

 So for pow(x, 2u), where x is a double, the match level for both
 (real, uint) and (real, real) is "match with implicit conversions."
 Therefore, it's an error.