Re: libstdc++ not conforming with --enable-concepts-check?

[Jonathan Wakely <cow at compsoc dot man dot ac dot uk>]

On Wed, Sep 22, 2004 at 09:18:53AM -0700, Nathan Myers wrote:

> On Wed, Sep 22, 2004 at 06:03:42PM +0200, Paolo Carlini wrote:
> > Nathan Myers wrote:
> > 
> > >Maybe std::queue shouldn't require DefaultConstructible, but isn't its
> > >default container std::deque, and shouldn't deque be enforcing it?  

No.

> > >Why is it failing to do so?

Because std::deque's value_type needn't be DefaultConstructible, unless
using the default argument for the fill ctor.

Oh ... and unless you explicitly instantiate the container and the
implementation happens to have taken the option of implementing the
fill ctor with more than one function (see my other mail in this thread).

> > Because deque (and vector, and so on...) individually have really 
> > **minimal** concept checks, right now. We can endeavour working on 
> > this, if we (you?) consider it a worthy project...

Hmm, I don't know - I think the concept checks are pretty complete. This
is the second one this month that has been removed for being _too_
restrictive!

> Worthy?  That's up to whoever would do the work.  However, it seems 
> inconsistent to check so much else and fail to check these most 
> frequently-used library structures.  

Except we do check  :)

> Is this a good time to argue for making --enable-concept-checks the
> default configuration?  The distros can't really take that step 
> themselves.  

I've suggested at least running the testsuite with concept-checks
enabled, as I usually run the tests with it and have fixed all failures
recently.

jon

-- 
"Subvert the dominant paradigm!"
	- The I.O.D.