[boost] RE: Interval Library- providing intervals of constants

From: Paul A. Bristow (boost_at_[hidden])
Date: 2002-09-10 08:39:19

Sadly, according the MSVC .net (never the most reliable authority)

        Only static const INTEGRAL members can be initialised,

only static const int

     template <unsigned int decimal_precision>
     struct interval_pi
     {
       static const double lower = 3; // integer default version.
       static const double upper = 4; // 3 < pi < 4
     };

Or is this not true for 'real' compilers?
Other suggestions?

Meanwhile I have identified the following precisions
for which interval constants are probably needed:

Floating point Type	Common C++	Totalbits 	Significandbits
IEEE single        	 float           	32	           23 + 1 =
24
VAX F 	float                              	32             23 + 1 = 24
6	9
IBM G/390 short	      float	            32	24
IEEE single extended 	?           	>=43	>=32
IEEE double	double               	64          	52 + 1 = 53
VAX G 	double                              	64	52 + 1 = 53
IBM G/390 long	double               	64	56
VAX D	long double                     	64	56	15	17
IEEE double extended	long double	80	>=64
NTL quad	'quad'	                       128	106	
IBM G/390 extended	long double      128	112
VAX H	long double                        128	112 + 1 = 113	
IEEE quadruple	long double            128	112 + 1 = 113
signed fractional	?                     127	128
unsigned fractional	?                     128 128

Paul

Dr Paul A Bristow, hetp Chromatography
Prizet Farmhouse, Kendal, Cumbria, LA8 8AB  UK
[phone]  Mobile [phone]
mailto:pbristow_at_[hidden]

> -----Original Message-----
> From: boost-bounces_at_[hidden]
> [mailto:boost-bounces_at_[hidden]]On Behalf Of Herve Bronnimann
> Sent: Friday, September 06, 2002 11:11 PM
> To: boost_at_[hidden]
> Subject: Re: [boost] Formal Review for Interval Library- getting
> intervalsof constants
>
> 
> Just occured to me: If you want to take advantage of compile-time,
> template arguments of the form <unsigned int> would work, and
> specializations for various ints would give you the desired rounding.
> 
> Example (the constants are bogus, but you'll get my point):
> 
>      template <unsigned int decimal_precision>
>      struct interval_pi {
>        static const double lower = 3; // default version
>        static const double upper = 4;
>      };
>      
>      #define BOOST_INTERVAL_SPECIALIZE_PI_CONSTANT(n, l, u) \
>        template <> struct interval_pi<n> { \
>          static const double lower = l;  \
>          static const double upper = u;  \
>        }
>      
>      BOOST_INTERVAL_SPECIALIZE_PI_CONSTANT(1,  3.1          , 3.2 
>          );
>      BOOST_INTERVAL_SPECIALIZE_PI_CONSTANT(2,  3.14         , 
> 3.15         );
> ...

     std::cout << interval_pi<7>::lower << std::endl;
> 
>