From: Anthony Williams (anthony.williamsNOSPAM_at_[hidden])
Date: 2002-10-30 06:22:10
Paul Mensonides writes:
>
> ----- Original Message -----
> From: "Anthony Williams" <anthony.williamsNOSPAM_at_[hidden]>
>
>
> > David Abrahams writes:
> > >
> > > I'm at a seminar right now where Steve Dewhurst has presented his
> > > technique for implementing typeof. It has the nice feature that you
> > > only need to register UDTs, and the system is then able to cope with
> > > them in all other type contexts including arrays, pointers, function
> > > (pointers/references), etc., etc. This technique was also covered in a
> > > recent CUJ article. I imagine it /might/ save some significant amount
> > > of code in Boost.Lambda and Spirit's Phoenix extension to take
> > > advantage of this approach. I think implementing it with MPL would
> > > make it a lot easier. Thoughts?
> >
> > >From my memory of the CUJ article, I think you would need the
> Preprocessor lib
> > more than MPL. Maybe implementing it with the preprocessor lib and MPL
> would
> > actually make it work on some compilers :-)
>
> Why would you need the PP lib? I can't immediately recall anything from the
> articles that suggests it off hand. What are you referring to?
The length of the identifying number is limited by the number of expansions
Steve was willing to do --- this strikes me as a task for the preprocessor lib.
Anthony
-- Anthony Williams Senior Software Engineer, Beran Instruments Ltd. Remove NOSPAM when replying, for timely response.