On 17/04/16 09:29, John Caron wrote:
The question I never answered for myself is what is the relationship
between
ucar.unidata.geoloc.projection.RotatedPole
which was contributed for CF:
[...]
and
ucar.unidata.geoloc.projection.RotatedLatLon
which was contributed for GRIB data:
John, as far as I can tell (and I added two unit tests to satisfy
myself), these two projections produce identical results, and differ
only in their projection parameters. Pull request #531 migrated rotated
lat/lon GRIB2 support to RotatedPole for conformance with CF Conventions.
The change to use RotatedPole improves interoperability; for example:
last week I merged a contributed Rotated_Pole projection into GeoTools
(from the COSMO community), and I have support for NetCDF CF Conventions
rotated pole in a GeoTools branch (soon to use the new Rotated_Pole
projection). The old RotatedLatLon did not conform to CF Conventions and
so lacked support.
PS: I presume youve given us some test GRIB files using template 32769
to add to our unit tests?
I have not yet done so. I think that adding a template 32769 file to the
Unidata test data set is an excellent idea. I will approach the data
owner, seek their permission, and follow up with Christian W-G.
Kind regards,
--
Ben Caradoc-Davies <ben@xxxxxxxxxxxx>
Director
Transient Software Limited <http://transient.nz/>
New Zealand
