Hi Sean, Either design is perfectly adequate. The extra line of code is inconsequential, and so it's a matter of philosophy. It's your prerogative to decide if adherence to a standard is more important than simplicity (no pun), or if the standard is in the way of simplicity and clarity. Here, it's a wash.
For the moment, I've avoided needing PointOnSurface, but if/when I do, I'll send a patch. Nino On Jun 18, 2009, at 9:55 AM, Sean Gillies wrote: > On Jun 17, 2009, at 2:28 PM, Nino Walker wrote: > >> Hi all, >> >> I went looking for GEOSPointOnSurface in topology, which I didn’t >> find, and then found GEOSSimplify commented out: >> http://trac.gispython.org/lab/browser/shapely.geos/trunk/shapely/geos/topology.py#L92 >> >> Is there a technical reason GEOSSimplify and GEOSPointOnSurface are >> excluded from Unary Ops? And by extension the *Polygon interfaces? >> >> It’s been a while since wading through the shapely source, so I >> apologize if this is a rehash. >> >> Thanks, >> >> Nino Walker :: Urban Mapping Inc. > > > Hi Nino, > > This is oversight, mainly. I didn't have a use case for point on > surface or simplification (only since 3.0, IIRC), and no one has sent > any patches. > > The Shapely trunk (and 1.1 alpha) does provide GEOS simplification, > currently as a geometry method: > >>>> simple = complex.simplify() > > I'm reconsidering this. Simplification isn't part of the OGC's simple > feature access profile, and its possible parameterizations defy > pluggability. I think it might be a callable from an ops module > instead > >>>> from shapely.geos.ops import simplify >>>> simple = simplify(complex) > > Thoughts? > > Sean > > _______________________________________________ > Community mailing list > [email protected] > http://lists.gispython.org/mailman/listinfo/community > _______________________________________________ Community mailing list [email protected] http://lists.gispython.org/mailman/listinfo/community
