From: Moore, Paul (Paul.Moore_at_[hidden])
Date: 2000-01-19 05:07:40
From: Dave Abrahams [mailto:abrahams_at_[hidden]]
> I think I'm now quite confused and should shut up until I
> have a clue what's going on ;). I hope my previous posting
> outlined some of the issues, and that some other people
> will carry the discussion forward at least until my
> sanity returns.
Yes, I am also confused here. As I started this with the issue of abs() for
the rational class, I'm now stalled until I can see a way forward.
I think that in the interests of progress, I will submit an "interim"
version of rational with no abs() supplied (after all, it's easy for the
user to write one using the supplied interface), with a note in the
documentation that abs() is intended to appear in a future release.
[[Maybe if someone can provide correct code for a standards-conforming
compiler, I can put that in, but it will be untested as I don't have access
to anything other than MSVC and I now see no hope of getting anything clean
which supports both MSVC and the C++ standard :-(]]
Thanks for all the discussion and comments. I really thought than rational<>
was a nice simple class with no big issues...
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk