[OSM-talk] Fwd: Is this relation really closed?

James Le Cuirot chewi at aura-online.co.uk
Wed Mar 31 22:23:44 BST 2010


On Wed, 31 Mar 2010 18:57:52 +0100
Emilie Laffray <emilie.laffray at gmail.com> wrote:
> 
> what you will find in PostGIS will fix the issue that you are having
> in your import of the database, but not in the raw data. It would
> actually consider rude not to fix it actually :)
> The problem was not obvious. As you have seen yourself, one of the
> relation tool didn't pick the issue. I don't think anyone is running a
> verification of all relations in general.
> You are more than welcome to try, and do not hesitate to ask for help
> if you don't see what the problem is. I am sure that plenty of people
> will be offering support if you need it.
> It is not uncommon for me to fix errors in places like Japan. There
> are errors that can be fixed just by looking at the geometry. It is
> the beauty of collaborative work, everyone can fix it.

Then fix it, I will! And I have tried actually but that site is still
reporting problems and PostGIS still isn't accepting it. I thought the
problem was that the ways went back on themselves, though they didn't
actually cross, so I merged with the nodes of the other nearby
boundary. Why there are two almost identical boundaries for this
stretch of coastline, I'm not quite sure. Is this likely to be a
mistake? And if this isn't the right way to fix a self-intersection,
what is? My edit is here.

http://www.openstreetmap.org/browse/changeset/4289465

James




More information about the talk mailing list