Simone Giannecchini a écrit :
> I was looking at the new classes in the coverage module for image
> geometry and gridrange 2d. I have been wonderinf for a while if we
> should somewhere report the tile layout for the coverages. I would
> think that gridrange2d could be a good candidate to contain info
> similar to ImageLayout for JAI. What do you think?

Possibly. Actually the legacy OGC 01-004 specification already have a method for
that:

http://geoapi.sourceforge.net/snapshot/javadoc/org/opengis/coverage/grid/GridCoverage.html#getOptimalDataBlockSizes()

I had hesitation about whatever we should keep this legacy method or not, but if
you believe that is has value, we will keep it. I don't remember if
GridCoverage2D implements it, but if it doesn't we could fix that.

GridRange2D may be a little bit too 2D specific. We could use GridRange, but I
believe that it would be a little bit misleading because a "range" contains a
"lower" and a "upper" value, which make no sense for tiling layout: only
"length" (i.e. width or height) make sense. A GridRange could apply to one
particular tile, but I don't think it would be a good fit for describing the
tiling of a GridCoverage in a whole.

The recent ImageGeometry brings nothing more than GridCoverage2D. Actually it
brings less, which is on purpose. It brings only what OGC 01-004 said, with the
additional restriction that the "gridToCRS" transform is garanteed affine. The
Envelope provided in GridGeometry2D is a Geotools extension, which work well for
GridCoverage2D but I realized was too much for ImageReader - hence the name
"ImageGeometry" for this simpliest GridGeometry implementation. Actually I had
hesitation about whatever "ImageGeometry" should be defined in
"org.geotools.image" or "org.geotools.coverage.grid" packages, but finally
though that it would be better to keep GridGeometry implementations together.

        Regards,

                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

Reply via email to