Re: [boost] generative geometry algorithms library idea

Subject: Re: [boost] generative geometry algorithms library idea
From: Bruno Lalande (bruno.lalande_at_[hidden])
Date: 2008-10-05 08:34:13

Hi Brandon,

I wouldn't have said better than Luke.What you are trying to do is
what we are all trying to do. Even though consensus is very difficult
to reach in that domain as Paul pointed out, it is wrong to say that
nothing has emerged from all the discussions that have occurred. The
fact that a C++ geometry library is definitely not all about providing
a point class, but proposing a set of algorithms generically adaptable
to any kind of point, is one of those consensuses. Another one is that
point coordinates selection should be made at compile-time (get<0>(p),
not get(0, p)), and be dimension-agnostic (get<0>, not get_x).

Barend's library wasn't made that way initially. It is now, even if
there remain some ways to acquire even more genericity. Points are now
handled generically (even a C-style array can be a point), linestrings
are now pairs of iterators, etc... And everything has been made
dimension-agnostic.

If I took a while to look at your library, it's obviously not for lack
of interest, but for lack of time. I finally glanced through it, it
sounds well written, well designed and clean to me. And like Luke
said, very similar to what we already have in terms of design, which
is very encouraging indeed. To the extent that at some places of your
files, I couldn't say directly if I'm browsing your library or
Barend's one :-) Everything is about coordinate_type, point_traits,
access_traits and stuff like that, which is absolutely normal. So we
shouldn't have troubles working together.

We are still working on cleaning everything and preparing the next
preview of Barend's library, it should come very soon.

Regards
Bruno