Fernando Cacciola wrote:

...

I just discovered (by debugging instead of reading :-() that a numeric_cast from a floating to an integral type always truncates by default, that is, 26.99999 is casted to 26. Is there a way to tell numeric_cast to use the nearest integer? Of course I can add 0.5 to my value and cast afterwards but that's definitely ugly.

If you're happy to use SVN Trunk, then you might want to take a look at the trunc and round family of functions here: http://svn.boost.org/svn/boost/trunk/libs/math/doc/sf_and_dist/html/math_too... Cheers, John.

Rainer Thaden wrote:

...

If you're happy to use SVN Trunk, then you might want to take a look at the trunc and round family of functions here: http://svn.boost.org/svn/boost/trunk/libs/math/doc/sf_and_dist/html/math_too...

Thanks John. From your answer I read that it doesn't seem to be possible to tell numeric_cast to return the nearest integer, right?

That I don't know, but you'd have to customise it for sure, where as these dedicated rounding functions are intended to do just what they say out of the box :-)

So I'd have to round the floating point type to the nearest integer with iround and catch rounding errors instead of the exceptions numeric_cast throws?

Yep. John.