Hi,

I have adapted a struct to be a fusion sequence like so:

struct s
{
    uint8_t i;
    uint8_t j;
    uint8_t k;
};

BOOST_FUSION_ADAPT_STRUCT(
    s,
    (uint8_t, i)
    (uint8_t, j)
    (uint8_t, k)
)

which works fine.  However, I pass the adapted struct to my MPL algorithm and it seems to always return false on mpl::is_sequence.  If I typedef boost::fusion::fusion_sequence_tag tag in the struct, then it all works, also if I pass in a proper fusion sequence, then that works as well.  I was wondering if there was a different way of doing things, I don't really want to require the tag to be typedef'ed in the structure.

Thanks,
Dave