[OSM-dev] Relations without members?

Karl Newman siliconfiend at gmail.com
Fri Oct 12 23:53:41 BST 2007

On 10/12/07, Matthias Julius <lists at julius-net.net> wrote:
> "Karl Newman" <siliconfiend at gmail.com> writes:
> > Another thing to consider is if a way ventures outside the boundary
> > and then back inside the boundary, the way will have to be split into
> > at least 2 parts in order to cut out the "outside" portion. This has
> > the same id number problem as the synthetic nodes above.
> Do the sections really need to have unique ids?
> Matthias
Well, yes. If two disparate parts of the same way have the same id,
the results could be unpredictable. We may be able to avoid this with
a composite object that represents the original way with the original
id but contains all the different parts that are split by the bounding
box. If the output of this algorithm is going to be passed to another
process that is expecting osm-type data, the split sections of the way
will have to be given new, unique ids.


More information about the dev mailing list