[C++] float, double & complex template non-type parms

[Nathan Sidwell]

Mark Mitchell wrote:

> The arguments in favor:
Another one is they're not implemented completely, for instance default
conversions (that happen for integral type template parms), don't happen
for floating ones.
	template <float T> void foo () {};
	template void foo<0> ();	//error doesn't match

>     template <class T, T t> void f(T t);
> 
>   is rejected as an overload candidate if `t' has type `float'.  If it
It is rejected with -pedantic, accepted otherwise.

> Nathan, if you like, go ahead and prepare a deprecation patch for the
> branch, and a patch to remove them altogether on the mainline
wilco,

Wolfram Gloger wrote:
> You forgot the very important:
> 
> - They enable you to write a C++ program that can compute a picture of
>   the Mandelbrot set _at compile time_.  With only integer arithmetic
>   that is much less fun..
<chuckle> I'm sure someone skilled will find it even more fun to implement
a fixed point implementation with integral arithmetic -- after all 30 or 62
bits of binary fraction is pretty good (IIR you need 2 bits of integer for
the set) - Heck you could even do something like
	template <int mantissa, int exponent> ...
probably ...

nathan

-- 
Dr Nathan Sidwell   ::   http://www.codesourcery.com   ::   CodeSourcery LLC
         'But that's a lie.' - 'Yes it is. What's your point?'
nathan@codesourcery.com : http://www.cs.bris.ac.uk/~nathan/ : nathan@acm.org