newbie ques - check double

[Apple]

is there any simple way to check if the value of a double is an integer?
like,

double d;
d = 15.5 => NOT an integer
d = 15.0 => IS an integer

if there's no standard function to do that, can anyone give me a hint on how
to write a right one?

thanks.

 

[John Harrison]

is there any simple way to check if the value of a double is an integer?
like,

double d;
d = 15.5 => NOT an integer
d = 15.0 => IS an integer

if there's no standard function to do that, can anyone give me a hint on how
to write a right one?

if (d == floor(d))
// d is an integer

john

 

[John Harrison]

#include <math.h> for the floor function.

john

 

[Mike Wahler]

if (d == floor(d))
// d is an integer

Another alternative is to use 'modf()'.

-Mike

 

[David Harmon]

On Tue, 28 Sep 2004 17:43:34 GMT in comp.lang.c++, "Mike Wahler"

Another alternative is to use 'modf()'.

And there are others. But testing any floating point result for any
exact value typically indicates wrong thinking in the design anyway.
Resist the temptation and save having to fix it in the end!

 

[Zian Smith]

is there any simple way to check if the value of a double is an integer?
like,

double d;
d = 15.5 => NOT an integer
d = 15.0 => IS an integer

if there's no standard function to do that, can anyone give me a hint on how
to write a right one?

thanks.

Try this:

double d;
bool isInt;

isInt = (int(d)-d) ? false : true;

 

[Mike Wahler]

"Apple" <[email protected]> wrote in message

Try this:

double d;
bool isInt;

isInt = (int(d)-d) ? false : true;

This can lead to possible undefined behavior (e.g. if
the value of 'd' is greater than 'INT_MAX' or less than
'INT_MIN'. Use the library functions, that's what they're
there for.

-Mike

 

[Mike Wahler]

On Tue, 28 Sep 2004 17:43:34 GMT in comp.lang.c++, "Mike Wahler"


And there are others. But testing any floating point result for any
exact value

OP is not looking for an 'exact value', but simply whether
a real number has a fractional part. Also note that types
'float' and 'double' can indeed represent integers exactly.

typically indicates wrong thinking in the design anyway.

I don't think it's 'wrong thinking' to try to determine
if a real number has a fractional part (unless of course
it's not really needed by the application being written).

-Mike