trunc

trunc

[phadreus]

what's the most efficient way within Ada
to obtain the whole-number portion Y of
a floating-point number X?

for example,

X = 5.5  Y = 5.0

X = 2.9  Y = 2.0

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: trunc

[Matthew Heaney]

phadreus@my-dejanews.com writes:

> what's the most efficient way within Ada
> to obtain the whole-number portion Y of
> a floating-point number X?
> 
> for example,
> 
> X = 5.5  Y = 5.0
> 
> X = 2.9  Y = 2.0

T'Floor

Re: trunc

[Tucker Taft]

Matthew Heaney wrote:
> 
> phadreus@my-dejanews.com writes:
> 
> > what's the most efficient way within Ada
> > to obtain the whole-number portion Y of
> > a floating-point number X?
> >
> > for example,
> >
> > X = 5.5  Y = 5.0
> >
> > X = 2.9  Y = 2.0
> 
> T'Floor

I suspect that T'Truncation is more what they had in mind.
T'Truncation(-3.5) = -3.0, whereas T'Floor(-3.5) = 4.0

-- 
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Distributed IT Solutions  (www.averstar.com/tools)
AverStar (formerly Intermetrics, Inc.)   Burlington, MA  USA

Re: trunc

[Michael Stark]

Tucker Taft wrote:
> 
> Matthew Heaney wrote:
> >
> > phadreus@my-dejanews.com writes:
> >
> > > what's the most efficient way within Ada
> > > to obtain the whole-number portion Y of
> > > a floating-point number X?
> > >
> > > for example,
> > >
> > > X = 5.5  Y = 5.0
> > >
> > > X = 2.9  Y = 2.0
> >
> > T'Floor
> 
> I suspect that T'Truncation is more what they had in mind.
> T'Truncation(-3.5) = -3.0, whereas T'Floor(-3.5) = 4.0

Tuck --

I assume you mean T'floor(-3.5) = -4.0 

I wouldn't want to confuse anyone more than they are already ;)

Mike

> 
> --
> -Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
> Technical Director, Distributed IT Solutions  (www.averstar.com/tools)
> AverStar (formerly Intermetrics, Inc.)   Burlington, MA  USA

-- 
Michael Stark
Goddard Research & Study Fellow
University of Maryland, College Park
e-mail: mstark@cs.umd.edu
phone: (301) 405-2721
"Where have you gone Joe DiMaggio, a nation turns its lonely eyes to
you"
    -- Simon & Garfunkel

Re: trunc

[Nick Roberts]

I suspect the 'Truncation' attribute is more likely to be what Phadreus is
looking for (see RM95 A.5.3), since it 'chops off' the fractional part,
whether the number is positive or negative.  As for efficiency, I would
guess that the use of the attribute is likely to be the most efficient
implementation for the vast majority of (native code) compilers.

-------------------------------------
Nick Roberts
-------------------------------------

Re: trunc

[Michael F Brenner]

The most efficient way is to go to URL http://www.adahome.com/rm95/
which is the Ada 95 Reference Manual which points you to the URL:

    http://www.adahome.com/rm95/rm9x-A-05-03.html

which defines the floor attribute which works like this:

   y:=x'floor;      -- Postcondition: Y is now the whole-number of X
   f:=x'fraction;   -- Postcondition: F is now the fractional value of X



> what's the most efficient way within Ada
> to obtain the whole-number portion Y of
> a floating-point number X?
 

Re: trunc

[Markus Kuhn]

mfb@mbunix.mitre.org (Michael F Brenner) writes:
|>    y:=x'floor;      -- Postcondition: Y is now the whole-number of X

GNAT 3.11p says "prefix of 'floor' attribute must be a type"
and the ARM agrees.

Correct would have been:

  y := float'floor(x);

Floor is a function, therefore it should look like one. Your
syntax looks like what might have been a convenient abbreviation,
but Ada is not about abbreviations.

Markus

-- 
Markus G. Kuhn, Computer Laboratory, University of Cambridge, UK
Email: mkuhn at acm.org,  WWW: <http://www.cl.cam.ac.uk/~mgk25/>

Re: trunc

[Howard W. LUDWIG]

Besides the issue, as Markus K. pointed out, that these attributes
are functions, the description below of the 'Fraction attribute is
likely misleading.

The inference I draw from the description below is that if I have
a floating-point number X of type T, then I can decompose X into
an integer part and a fractional part, whose sum is X, using these
two attributes.  However, using the URL given below, one can see
that 'Fraction is the counterpart of 'Exponent to decompose a
floating-point number into a fractional mantissa and an exponent, such that
     X := T'Fraction(X) * T'Machine_Radix**T'Exponent(X).

HWL

Michael F Brenner wrote:

> The most efficient way is to go to URL http://www.adahome.com/rm95/
> which is the Ada 95 Reference Manual which points you to the URL:
>
>     http://www.adahome.com/rm95/rm9x-A-05-03.html
>
> which defines the floor attribute which works like this:
>
>    y:=x'floor;      -- Postcondition: Y is now the whole-number of X
>    f:=x'fraction;   -- Postcondition: F is now the fractional value of X
>
> > what's the most efficient way within Ada
> > to obtain the whole-number portion Y of
> > a floating-point number X?