[GRASS-user] v.generalize failing on some geoms

[Markus Metz]

On Mon, Feb 12, 2018 at 6:22 PM, Paolo Cavallini <cavallini at faunalia.it>
wrote:
>
> Il 11/02/2018 18:34, Markus Metz ha scritto:
>
> > This implies that the data have been reprojected at some stage somewhere
> > from one SRS to the other. The most likely explanation for the different
> > results, particularly the topological errors reported by v.in.ogr, is
> > that non-topological polygons were reprojected. Can you make these geoms
> > in EPSG:3857 and EPSG:3003 available for testing?
>
> Hi Markus,
> thanks for your thoughts. Here the data, original and simplified, in two
> CRS (3003 is simplified correctly, 3857 not):
> http://www.faunalia.eu/~paolo/test_generalize.tar.xz

The original data in EPSG:3003 and EPSG:3857 are not identical. The
original data in EPSG:3003 are topologically correct, while the original
data in EPSG:3857 are not. Further on, the original data in EPSG:3857
contain features that are not present in the original data in EPSG:3003
which is quite strange.

Reprojecting the original data in EPSG:3003 to EPSG:3857 within GRASS works
fine, also with subsequent v.generalize.

That means that the original data in EPSG:3857 are some reprojected version
of the original original data (which are these?). The reprojection step,
apparently performed on polygons, not GRASS areas (how did you reproject?),
introduced topological errors. Please use native GRASS v.in.ogr + v.proj to
reproject polygons.

Markus M

> All the best.
> --
> Paolo Cavallini - www.faunalia.eu
> QGIS & PostGIS courses: http://www.faunalia.eu/training.html
> https://www.google.com/trends/explore?date=all&geo=IT&q=qgis,arcgis
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20180212/9c7df5a3/attachment.html>