From: John Maddock (john_at_[hidden])
Date: 2004-01-24 06:11:20
> > What is that wrong in this case in circular dependency
> > (I do not advocate it is good)?
> It causes lots of recompilation; that's really frustrating when the
> test could be done just as well if not better at compile-time using
> some kind of static assert.
The tests check both the compile time value and the run time value (and yes
some implementation on some compilers are known to behave differently in
these two cases :-( ).
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk