Boost logo

Boost :

From: Peder Holt (peder.holt_at_[hidden])
Date: 2005-03-30 08:04:28


On Wed, 30 Mar 2005 06:51:03 -0800, Arkadiy Vertleyb
<vertleyb_at_[hidden]> wrote:
> "Peder Holt" <peder.holt_at_[hidden]> wrote
> ...
>
> <off topic>
>
> Hey, welcome back!

Thanks :)

> How was your vacation?

Great! Travelling with a 3 month old baby to the other side of the
world is now proven to be doable :)

>
> </off topic>
>
> > Wow. Very impressive. What a pity noone has thought of that earlier...
>
> How is one supposed to think of something like this? The next thing we find
> out -- VC might already have an undocumented typeof -- just called
> differently :-)
>
> > I have modified the current vintage implementation to use this form of
> > typeof, allowing it to compile with VC6.5 as well.
>
> Did you remove __if_exists/if_not_exists stuff? This would mean it would
> also compile with the language extentions disabled...

Yep. VC6.5 does not implement __if_exists. I used the
compile-time-constants implementation that already exists with the
vintage implementation, so now it relies on two compiler bug/features
+ ODR in stead of one compiler bug/feature + a compiler extension :)

> >
> > I guess this deprecates most of the vintage code, since that was
> > implemented in order to allow support of VC6.5 and VC7.0.
> >
> And also removes half of the users from the compliant implementation.
> Really, how many are left if gcc has native typeof, and vc7 uses this
> stuff??
>
> Well, competition is competition...
>
> I think we should contact the guy and ask if he is interested in working
> with us. If not -- we can use his code, but I think this would be fair to
> ask. I'll do this later today.

Good idea. Do that

Regards
Peder

>
> Regards,
> Arkadiy
>
> _______________________________________________
> Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
>


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk