https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115218
--- Comment #4 from Jonathan Wakely <redi at gcc dot gnu.org> --- (In reply to Jonathan Wakely from comment #3) > (In reply to Patrick Palka from comment #1) > > Does this also mean the iterator's default ctor needs a > > default_initializable<_Vs...[0]> constraint? > > Yes, I think the default ctor should get deleted if the constraint isn't met. Sorry, that was meant as a reply to comment 2. I meant to say the variant's constraint should already affect the default ctor, so we don't need one there explicitly. So when I said "Yes" I meant "No" :-) Because we should already have some a constaint implicitly.
