On Mon, Dec 08, 2008 at 07:23:01PM -0500, Heidt, Christopher M. wrote: > You are absolutly correct and my frustrations are more centered on > specifications than on OpenLayers. > > You are spot on that having a circle geometry whould mean nothing to the > community if it couldn't be represented in the standard specifications like > GML, WKT, KML, etc. > It just saddens me that the lack of circle support in the specs has led to > work-arounds and abuses that should never have needed to exist. > WKT's lack of circle support lead Oracle to create CURVEPOLYGON which is > completely unsupported by anything, even though Oracle outputs it as valid > WKT. > > Although you are correct that a RegularPolygon with many sides can look like > a circle, calculations based off that shape can be severly flawed when > treated as one. > if i were to make an app that fetched data within a drawn circle, my results > could prove to be highly inacurate by the lack of true curveature.
With a sufficient number of sides, I'm not convinced that's actually true. In any case, an OpenLayers.Geometry.Circle doesn't help that either. You have to serialize the 'circle' to something to query with it. If you want to, you can think of your RegularPolygons as circles, and treat your appilcation to do the same -- but that doesn't help hte server. REgards, -- Christopher Schmidt MetaCarta _______________________________________________ Users mailing list [email protected] http://openlayers.org/mailman/listinfo/users
