[OSM-talk] What's wrong with this multipolygon?

Zeke Farwell ezekielf at gmail.com
Sat Aug 28 15:04:30 BST 2010


Yeah I don't see anything about that Lake Nona multipolygon that would make
it render poorly.  I feel like ever since Mapnik began updating in a matter
of minutes or hours (rather than the week it used to take) Osmarender has
been atrophying.    The Southchase residential area has many other
multipolygons inside it, or maybe on top would be a better description since
they are not also tagged as inner ways of the Southchase multipolygon.  I
wouldn't be surprised if advanced multipolygons like these are just too
complicated for Osmarender to handle.

Zeke


On Sat, Aug 28, 2010 at 1:55 AM, Nathan Edgars II <neroute2 at gmail.com>wrote:

> On Sat, Aug 28, 2010 at 12:53 AM, Zeke Farwell <ezekielf at gmail.com> wrote:
> > As you know, Mapnik will render multipolygons
> > correctly as long as all the ways form complete rings and are tagged with
> > "outer" and "inner" as needed, but it turns out that Osmarender also
> needs
> > the ways making up each ring to be in consecutive order.
>
> I don't know if that's the problem for Southchase (I changed the
> order, but it will rerender when it rerenders) but Lake Nona is
> definitely in consecutive order, yet still renders badly:
> http://www.openstreetmap.org/?lat=28.3815&lon=-81.27844&zoom=17&layers=O
> (It also doesn't have any other multipolygons inside it, so that's not
> the problem.) Unless it's the adjacent Orlando International Airport
> order... I've changed that and will see if it fixes Lake Nona.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20100828/f18af017/attachment.html>


More information about the talk mailing list