This problem is also causing our geometryfactoryfinder to fail because
it relies on finding a CRS using the equals comparator, which does not
work. I've written a small test here that shows the error. Basically I
create a CRS, then create a 2nd crs from the WKT of the first one. They
are the same thing, but not the same object. Then I create 2
geometrybuilders with those crs's. When I try to use builder2 to build
something from a builder1 object, it fails with a
mismatchedreferencesystemexception.
public void testPrimFactCRS() throws UnsupportedOperationException,
FactoryException {
CoordinateReferenceSystem crs = DefaultGeographicCRS.WGS84;
GeometryBuilder builder = new GeometryBuilder(crs);
CoordinateReferenceSystem crs2 = CRS.parseWKT(crs.toWKT());
GeometryBuilder builder2 = new GeometryBuilder(crs2);
// create a list of connected positions
List<Position> dps = new ArrayList<Position>();
dps.add(builder.createDirectPosition( new double[] {20, 10} ));
dps.add(builder.createDirectPosition( new double[] {40, 10} ));
dps.add(builder.createDirectPosition( new double[] {50, 40} ));
dps.add(builder.createDirectPosition( new double[] {30, 50} ));
dps.add(builder.createDirectPosition( new double[] {10, 30} ));
dps.add(builder.createDirectPosition( new double[] {20, 10} ));
// create linestring from directpositions
LineString line = builder.createLineString(dps);
// create curvesegments from line
ArrayList<CurveSegment> segs = new ArrayList<CurveSegment>();
segs.add(line);
// Create list of OrientableCurves that make up the surface
OrientableCurve curve = builder.createCurve(segs);
List<OrientableCurve> orientableCurves = new
ArrayList<OrientableCurve>();
orientableCurves.add(curve);
// create the interior ring and a list of empty interior rings
(holes)
Ring extRing = builder2.createRing(orientableCurves);
List<Ring> intRings = new ArrayList<Ring>();
// create the surfaceboundary from the rings
SurfaceBoundary sb = builder2.createSurfaceBoundary(extRing,
intRings);
// create the surface
Surface surface = builder2.createSurface(sb);
}
Graham.
Martin Desruisseaux wrote:
>Le mardi 11 septembre 2007 à 10:35 -0700, Jody Garnett a écrit :
>
>
>>Other than this the two look very similar... it is absolutely killing me
>>to have to cast to AbstractCRS in order to perform this check; we can do
>>this in all the GeometryFactory methods but I cannot do it in client
>>code - we need to figure out a way to make this functionality available
>>through the interfaces.
>>
>>
>
>You means a way to invoke CRS.equalsIgnoreMetadata(...) without the
>current requirement (in "equalsIgnoreMetadata" implementation) that both
>CRS are GeoTools implementations? Yes I realize that we have a strong
>limitation here.
>
>I see to possible approachs:
>
>1) Wraps non-geotools implementations into geotools ones under the
> hood. Hopefully the cost would not be that high.
>
>2) Promote the current AbstractIdentifiedObject.equals(..., boolean)
> method to GeoAPI. But I would be reluctant to do that since it
> would require defining its behavior in the contract for every
> IdentifiedObject subclasses, and can't pretend that we got them
> right in GeoTools implementation.
>
> Martin
>
>
>
>-------------------------------------------------------------------------
>This SF.net email is sponsored by: Microsoft
>Defy all challenges. Microsoft(R) Visual Studio 2005.
>http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>_______________________________________________
>Geotools-devel mailing list
>[email protected]
>https://lists.sourceforge.net/lists/listinfo/geotools-devel
>
>
--
Graham Davis
Refractions Research Inc.
[EMAIL PROTECTED]
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
