[postgis-devel] ST_Summary has no invalid info

[Kevin Neufeld]

Such a function would definitely be nice to see.   Perhaps a 
set-returning function that returns the problematic locations (ie. a 
geometry collection of the points or lines of intersections) and a brief 
text field describing the error (ie. "self-intersecting boundary", "hole 
outside polygon", "shared edge in multipolygon", ...)

Add it to the wish list :)

-- Kevin

Obe, Regina wrote:
> For some reason I always expected the summary function or at least have
> an accompanying function that does that would flag the details of
> invalid geometries as part of its inspection and give info similar to
> what ST_Valid NOTICE gives.  Why I would think such a silly thing is
> beyond me, or maybe such a function exists and I haven't noticed.
> 
> Is it a very costly check?  Seems like it would be useful.  The problem
> with the ST_IsValid notice is that it comes at you as one continuous
> stream - so if you have a whole slew of geometries, you have no idea
> which notice goes with what. I got one of those such cases from nowhere
> with 100 of 6000 someodd geometries being invalid and I was hoping to
> generate a report detailing to the provider why there dataset sucks and
> what exactly is wrong with it.
> 
> Thanks,
> Regina
> -----------------------------------------
> The substance of this message, including any attachments, may be
> confidential, legally privileged and/or exempt from disclosure
> pursuant to Massachusetts law. It is intended
> solely for the addressee. If you received this in error, please
> contact the sender and delete the material from any computer.
> 
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel