|
Boost Users : |
From: Felipe Magno de Almeida (felipe.m.almeida_at_[hidden])
Date: 2008-02-07 10:59:15
On Feb 7, 2008 1:31 PM, Frank Mori Hess <frank.hess_at_[hidden]> wrote:
> On Thursday 07 February 2008 07:40 am, Fernando Cacciola wrote:
> > Then I suspect you don't really want to "convert" the float to int but to
> > get at the value-representation of the float.
> > If that's the case, you can do this:
> >
> > char* float_value_rep =
> > static_cast<char*>(static_cast<void*>(&your_float));
>
> Isn't this what reinterpret_cast is for? It wouldn't require bouncing through
> a void*. Also, to the original poster, don't succumb to the temptation to
> cast to anything other than a char* (like an int*) or you'll end up in strict
> aliasing violation hell.
For most uses of reinterpret_cast, it invokes implementation specific
behavior. static_cast is portable though.
So the latter is preferrable.
>
> --
> Frank
Regards,
-- Felipe Magno de Almeida
Boost-users list run by williamkempf at hotmail.com, kalb at libertysoft.com, bjorn.karlsson at readsoft.com, gregod at cs.rpi.edu, wekempf at cox.net