NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Hi Eric: Eric Davies wrote: > I recently tried to read an OCCAM HDF5 file with the NetCDF Java > library. While I was able to read it, it didn't look like a regular > NetCDF file in that there were no global dimensions but each variable > had its own distinct set of nameless dimensions. > > Subsequent reading suggests this is a general difference between > NetCDF-3 and HDF5, the lack of global dimensions. yep > > This brings up the following questions: > Should the NetCDF Java library be trying to synthesize valid dimensions > using attribute information? we do try to do that for HDF-EOS. If the info is in the OCCAM files, we can do that also. > > If the NetCDF-4 library is based on HDF-5 (assuming one enables that > feature), does this imply that entire concept of named dimensions is > being discarded from NetCDF? What would be the effect on conventions > such as CF and COARDS? Ive been arguing that the lack of shared dimensions is not acceptable. NC4 adds some conventions to HDF5 to implement shared dimensions, and there's no chance that we will not continue to keep that in our data model. Its these additions that mean that you lose information when reading a netcdf4 file through the HDF5 interface, even though netcdf4 is build on hdf5. I advise anyone who cares about this to advocate to the HDF5 group that they should support shared dimensions. They do have "dimension scales" which implement a subset of the functionality of shared dimensions. > > Eric. > > ********************************************** > Eric Davies, M.Sc. > Barrodale Computing Services Ltd. > Tel: (250) 472-4372 Fax: (250) 472-4373 > Web: _http://www.barrodale.com_ <http://www.barrodale.com/> > Email: eric@xxxxxxxxxxxxx > ********************************************** > Mailing Address: > P.O. Box 3075 STN CSC > Victoria BC Canada V8W 3W2 > > Shipping Address: > Hut R, McKenzie Avenue > University of Victoria > Victoria BC Canada V8W 3W2 > ********************************************** > > > > ------------------------------------------------------------------------ > > _______________________________________________ > netcdf-java mailing list > netcdf-java@xxxxxxxxxxxxxxxx > For list information or to unsubscribe, visit: http://www.unidata.ucar.edu/mailing_lists/
netcdf-java archives: