[Qgis-user] Multipolygons with errors ?

[Hugues François]

Hello,

I'm back after few time because of work and because I had to look deeper at OGC spec. to give a right answer.

After I read it the implementation standard (http://www.opengeospatial.org/standards/sfa) I definitively think the union of two polygons sharing a limited number of points result in a valid multipolygon since a multipolygon is defined as their elements' *interior* may not intersect and may touch at only a finite number of points. Furthermore, fig 12 and 17, draw an example similar to my case study to illustrate respectively an invalid polygon and (17c) a valid multipolygon.

When QGis validity checker throw an error of this kind of multipolygon, I think it is wrong. Should I report a bug ?

Hugues.

-----Message d'origine-----
De : Goyo [mailto:goyodiaz at gmail.com] 
Envoyé : lundi 25 mars 2013 21:13
À : David J. Bakeman
Cc : Hugues François; Quantam GIS
Objet : Re: [Qgis-user] Multipolygons with errors ?

2013/3/25 David J. Bakeman <dbakeman at comcast.net>:

> OK.  So if I take your shape file and run Vector/Geometry 
> Tools/Singlepart to multipart.

Then if two parts of a feature touch each other I think it's invalid again (as for the OGC-SFS).

> Note grass doesn't report any topo errors on either the singlepart or 
> multipart shape files.

This may be due to GRASS using a different geometry model. It validates against different criteria.

Goyo