https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120432
--- Comment #3 from Patrick Palka <ppalka at gcc dot gnu.org> --- (In reply to Arthur O'Dwyer from comment #2) > But I think a better answer in this case is "because using > a volatile index *will not actually compile*, so flat_map shouldn't > gratuitously advertise that signature to SFINAE." > https://godbolt.org/z/bdTK7nPKG Good point.. I think this also means the standard transparent-compare overload is underconstrained: https://godbolt.org/z/354GTTsxG
