Re: Disable some concept checks in C++11

[Jonathan Wakely <jwakely at redhat dot com>]

On 11/03/17 13:32 +0100, François Dumont wrote:
> On 10/03/2017 13:25, Jonathan Wakely wrote:
> > On 08/03/17 22:03 +0100, François Dumont wrote:
> > >   So slist also require it for its assign implementation. And a 
> > > slist of not assignable type instances is used in 
> > > libstdc++-prettyprinters/59161.cc.
> >
> > That test is ill-formed, strictly speaking. It violates the concept
> > requirements of the slist type. It works because it never tries to
> > assign them.
> >
> > If you're concerned you could forcibly disable the concept checks in
> > that testcase. But I'd just stop  using the concetp checks :-)
> >
> >
> > > Should I just forget about it ?
> >
> > Yes. They're broken. I don't think it's a good use of your time.
> >
> > > Go ahead with the patch except the slist part and then disable 
> > > concept checks ?
> >
> > If by "disable concept checks" you mean stop building with them
> > enabled, then yes, I think you should do that :-)
>
> Yes, this is what I meant so I'll do that.
>
> > The rest of the patch is OK, but please wait for Stage 1. Be sure to
> > test without concept checks enabled too.
> Ok I add it to my list of awaiting commits.

Actually, I've just looked at the patch again and it's OK to commit
now (as long as testing passes using the default configuration).

It only removes some unused private typedefs, so that's safe.

Thanks!