www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - Initializing floating point types with explicit mantisa and exponent

reply =?UTF-8?B?Tm9yZGzDtnc=?= <per.nordlow gmail.com> writes:
How do I best initialize a D double to an exact mantissa and 
exponent representation?

I'm specifically interested in

     2^^i for all i in [min_exp, max_exp]
Jan 16
next sibling parent pineapple <meapineapple gmail.com> writes:
On Tuesday, 17 January 2017 at 00:08:24 UTC, Nordlöw wrote:
 How do I best initialize a D double to an exact mantissa and 
 exponent representation?

 I'm specifically interested in

     2^^i for all i in [min_exp, max_exp]
This mach module can do the job: https://github.com/pineapplemachine/mach.d/blob/master/mach/math/floats/inject.d
Jan 16
prev sibling next sibling parent Nicholas Wilson <iamthewilsonator hotmail.com> writes:
On Tuesday, 17 January 2017 at 00:08:24 UTC, Nordlöw wrote:
 How do I best initialize a D double to an exact mantissa and 
 exponent representation?

 I'm specifically interested in

     2^^i for all i in [min_exp, max_exp]
See std.bitmanip : FloatRep , DoubleRep;
Jan 16
prev sibling parent reply kinke <noone nowhere.com> writes:
On Tuesday, 17 January 2017 at 00:08:24 UTC, Nordlöw wrote:
 How do I best initialize a D double to an exact mantissa and 
 exponent representation?

 I'm specifically interested in

     2^^i for all i in [min_exp, max_exp]
If it doesn't have to be D ;), it can be as simple as `core.stdc.math.ldexp(1, exponent)`. No CTFE though.
Jan 17
parent reply =?UTF-8?B?Tm9yZGzDtnc=?= <per.nordlow gmail.com> writes:
On Tuesday, 17 January 2017 at 16:40:57 UTC, kinke wrote:
 If it doesn't have to be D ;), it can be as simple as 
 `core.stdc.math.ldexp(1, exponent)`. No CTFE though.
Isn't it a simple as 2.0^^exponent ?
Jan 17
parent reply kinke <noone nowhere.com> writes:
On Tuesday, 17 January 2017 at 17:56:13 UTC, Nordlöw wrote:
 On Tuesday, 17 January 2017 at 16:40:57 UTC, kinke wrote:
 If it doesn't have to be D ;), it can be as simple as 
 `core.stdc.math.ldexp(1, exponent)`. No CTFE though.
Isn't it a simple as 2.0^^exponent ?
It should and I looked into that as well, but I didn't like the implementation as loop: https://github.com/dlang/phobos/blob/master/std/math.d#L5988 A special case for base x == 2 wouldn't hurt.
Jan 17
parent =?UTF-8?B?Tm9yZGzDtnc=?= <per.nordlow gmail.com> writes:
On Tuesday, 17 January 2017 at 18:25:46 UTC, kinke wrote:
 It should and I looked into that as well, but I didn't like the 
 implementation as loop: 
 https://github.com/dlang/phobos/blob/master/std/math.d#L5988

 A special case for base x == 2 wouldn't hurt.
That seems strange. Why isn't that a builtin behaviour for operator ^^?
Jan 17