[OSM-talk] talk Digest, Vol 26, Issue 47

[David Earl]

> -----Original Message-----
> From: Robert Hart [mailto:Robert.Hart at BuroHappold.com]
> Sent: 19 October 2006 17:58
> To: David Earl
> Subject: RE: [OSM-talk] talk Digest, Vol 26, Issue 47
>
>
> >
> > I agree. This is known as the "winding rule" in computer graphics. But
> it
> > isn't necessary for rendering in general to have the directional
> winding
> > rule (though I can't speak to osmarender in particular) so long as you
> > know
> > some known point is "outside". A point nominally at infinity will do,
> then
> > you just count "in", "out", "in", "out" transitions fro rendering
> purposes
> > irrespective fo the path direction.
>
> Yep. Now answer the same question for a non-Euclidian space. There's no
> such thing as a "point nominally at infinity" when you are on the
> surface of the earth...
>
> Then again there's a whole bunch of OSM stuff that is going to fail
> spectacularly when people try and map Antarctica, and the eastern tip of
> Russia.... so it'll probably do for now.

Indeed. It's pedantic. In practice if you take a bounding box that contains
all the segments (including the bits that make up the holes) of the way that
makes up the areayou're interested in then any point just outside the
bounding box will do. That's why I said nominally. For the purpose of
determining inside and outside, the area (which is trivially small compared
to the Earth) can be considered, locally, to lie on plane not a sphere.

David