[GRASS-user] Snap distance to clean imported map?

[Markus Metz]

On Thu, Sep 12, 2019 at 2:24 AM Rich Shepard <rshepard at appl-ecosys.com>
wrote:
>
> Importing a large (4.4G) vector map of wetlands takes a long time even on
my
> 8-core/16-thread, 32G desktop. When it finally completed grass recommended
> re-importing with an additional 'snap' distance specified:
>
> Command line: > v.in.ogr in=OR_geodatabase_wetlands.gdb out=more_wetlands
loc=geo_wetlands
>
> WARNING: Number of incorrect boundaries: 88544
> -----------------------------------------------------
> WARNING: The output contains topological errors:
>           Unable to calculate a centroid for 329691 areas
>           Number of incorrect boundaries: 88544
> The input could be cleaned by snapping vertices to each other.
> Estimated range of snapping threshold: [1e-09, 1]
> Try to import again, snapping with 1e-05: 'snap=1e-05'
>
> In the past 1e-05 produced similar warnings.
>
> I'd like advice on two issues:
>
> 1) Is 1e-05 the most reasonable starting snap distance? (I assume it is a
> generic value but perhaps not for all maps.) If not, what would be a snap
> distance to use?

The estimated snap distance is estimated from the bounding box of the
imported data and thus specific for each import. In this particular case
the suggested snap distance is snap=1e-5. If results are not as expected,
increase or decrease by powers of 10, i.e. up to 1e-4 or down to 1e-6.
>
> 2) What would be a reasonable snap distance to stop re-importing when
> warnings continue to be displayed?

The upper limit of the estimated range, in this case of [1e-09, 1], i.e 1.

Markus M
>
> TIA,
>
> Rich
>
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/grass-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20190916/797132af/attachment.html>