[postgis-devel] error in ST_ModEdgeSplit

[Rémi Cura]

Solved by introducing precision (topology precision),
and another split function (st_split(line, point) is not reliable).

The edge is not moved in this case, it is the node that is moved,
thus the topology is safe

Cheers,
Rémi-C

2015-02-11 12:28 GMT+01:00 Sandro Santilli <strk at keybit.net>:

> On Wed, Feb 11, 2015 at 12:02:13PM +0100, Rémi Cura wrote:
> > Hey,
> >
> > I get an error with ST_ModEdgeSplit that looks like aa precision related
> > issue.
> >
> > When trying to split an edge with a point (the point is from
> > ST_ClosestPoint, so _very_ close to the edge),
> > st_modeedgesplit complains that the point is not on the edge.
>
> This is a known issue. I've mixed feeling about what the function
> should do because "ST_" prefixed functions supposedly come from
> the standard that makes no mention of "precision".
>
> The toTopoGeom function handles that by first snapping the edge against
> the point retrived by "ClosestPoint". Note that such snapping might
> move the edge in a way that it could end up colliding against another
> edge. This may happen if the intersection node you found is close to
> another existing node and between two indicent edges almost parallel.
> There's no primitive function handling such case, which would involve
> updating partially merging the two edges into one.
>
> --strk;
>
>   ()   Free GIS & Flash consultant/developer
>   /\   http://strk.keybit.net/services.html
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at lists.osgeo.org
> http://lists.osgeo.org/cgi-bin/mailman/listinfo/postgis-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-devel/attachments/20150211/75c526f2/attachment.html>