https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127373
--- Comment #3 from Ivan Sokolov <sugrob9000aero at gmail dot com> --- (In reply to Drea Pinski from comment #2) > It does. Because if you used int rather than float, it is accepted. What I was trying to say that when the conversion operator is synthetized, the constrains don't seem to be present on it. As if your emulation were modified to replace the second occurence of std::integral with a mere typename, and same with std::floating_point.
