20 Apr
2006
20 Apr
'06
10:03 a.m.
John Maddock wrote:
template<std::size_t TAlign, std::size_t Align> struct is_aligned { BOOST_STATIC_CONSTANT(bool, value = (TAlign >= Align) & (TAlign % Align == 0) ^^^ <<---- here! ); };
Should it be "&&" instead of "&"?
Logically yes: however when you think about it, it actually makes no practical difference, and I believe it was done that way deliberately as a workaround for certain broken compilers that choke on logical expressions "&&" inside integral constant expressions. So for now I'm going to leave it alone.
Perhaps a comment to this effect could be added to the code to show that it is intentional and not an error or oversight. Paul