From: Juergen Hunold (juergen.hunold_at_[hidden])
Date: 2007-12-02 06:48:43
Hi Dave !
On Sonntag 02 Dezember 2007, David Abrahams wrote:
> Uh, wait: missing virtual d'tors? Sorry, I should have paid more
> attention. That warning is bogus and I certainly don't want anyone
> "fixing" it if there are no other virtual functions in the class.
No, they all have "real" virtual functions.
> And even if there are other virtual functions, I think it's a
> suspicious change to make. Please back out any added virtual dtors!
I agree that this warning is bogus. The evil thing is that you can't
disabled it, at least on gcc :-((
I've reverted revisions 41544 and 41549, rerun the tests and attached
the compressed test output. The nasty thing is that it is triggered
during template instation which will polute the output with even more
bogus messages. Please take a closer look at both the patch and the
compiler log. I'll revert the changes in the meantime.
I love atomic commits ;-))
> > The last warning ist a BOOST_ASSERT in
> > boost/python/converter/implicit.hpp
> > This could be fixed by using the new BOOST_VERIFY macro instead.
> > Patch attached.
> Uh, no thanks. I *want* the test to compile away in release mode.
> See http://lists.boost.org/Archives/boost/2006/06/106919.php . I
> think one of these already exists somewhere in Boost.Python.
I thought that BOOST_VERIFY
would do the "right thing" is this case. Yours template based solution
might be better, so I'll take a look at it.
-- * Dipl.-Math. Jürgen Hunold ! Ingenieurgesellschaft für * voice: ++49 511 262926 57 ! Verkehrs- und Eisenbahnwesen mbH * fax : ++49 511 262926 99 ! Lister Straße 15 * juergen.hunold_at_[hidden] ! www.ivembh.de * * Geschäftsführer: ! Sitz des Unternehmens: Hannover * Prof. Dr.-Ing. Thomas Siefer ! Amtsgericht Hannover, HRB 56965 * PD Dr.-Ing. Alfons Radtke !
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk