From: Paul Mensonides (pmenso57_at_[hidden])
Date: 2006-01-29 12:11:42
> -----Original Message-----
> From: boost-bounces_at_[hidden]
> [mailto:boost-bounces_at_[hidden]] On Behalf Of Tobias Schwinger
> Paul Mensonides wrote:
> > BOOST_PP_IF(
> > n,
> > BOOST_PP_TUPLE_EAT,
> > BOOST_PP_TUPLE_REM
> > )(1)( EXPENSIVE_MACRO_INVOCATION() )
> >
> > ...where EXPENSIVE_MACRO_INVOCATION() is only invoked if 'n' is 0.
>
> Guessing from the overtone of your comment it would be better to say
>
> BOOST_PP_IF(n, BOOST_PP_TUPLE_EAT(0), EXPENSIVE_MACRO_INVOCATION)()
>
> to be on the safe side with less compliant preprocessors. Correct?
Yes. It's also just better. Sorry, it is getting late in the morning here, and
I haven't slept yet.
> ---
>
> #define MACRO(x) x
> MACRO(BOOST_PP_EMPTY())
>
> BTW. is this special case safe (theoretically and
> practically, that is)?
On a good preprocessor, it's safe. Broken preprocessors tend to allow empty
arguments anyway, so it should be safe there too.
Regards,
Paul Mensonides