[postgis-users] Wrong result in spatial analysis methodslikeintersects due to limited coordinate precision

[Markus Schaber]

Hi, Nicolas,

Nicolas Ribot wrote:

> What about applying snapToGrid(geom, size) to your data to reduce
> their precision, adapting it to the real precision your data were
> produced with.

snapToGrid will not help him at all. If you view the sketch he posted,
it is clear that _any_ kind of grid / rounding of coordinates (be it
explicit or due to FP precision) will expose him to this problem.

It might, by accident, fix this particiular example when the grid is
chosen accordingly, but then there will exist other examples that fail,
but worked without snapping, or even fail in both cases.


HTH;
Markus

-- 
Markus Schaber | Logical Tracking&Tracing International AG
Dipl. Inf.     | Software Development GIS

Fight against software patents in EU! www.ffii.org www.nosoftwarepatents.org