Ansley-
I think I have finally solved the varm code problem.
Varm is rather more complicated than I thought.
Anyway, the fix should go in tonight, which means
that the snapshot available tomorrow should have the fix.
It works on the test case you sent and a couple of others
I created. Let me know if it fails inside ferret itself.
=Dennis Heimbigner
Ansley Manke wrote:
I'm resubmitting this, under a different subject line so it's not mixed
up with the "problems compiling" thread.
Ansley Manke wrote:
Hello,
I've built the netcdf-4.1-beta2 library, and am investigating its use
both by looking at ncdump and linking it with Ferret, to open local
netCDF datasets and remote URL's. We have previously used the OPeNDAP
library for this, but are now eager to switch over to netcdf-4.
I'm finding that for remote URL's, I can read data with a call to
"nc_get_vara_float", but if I try "nc_get_varm_float" which allows for
strides and specifying alternative mapping of the data structure, I'm
getting an error code of -36, which translates as "Invalid Argument".
For now I'm not trying to do any strides or permutations; just setting
the stride indices to 1 and the imap indices as we always have,
pointing to the location of each dimension of the data as it is stored.
For reading data from local netCDF files, calls to nc_get_varm_float
work fine.
Any ideas?
Ansley
_______________________________________________
netcdfgroup mailing list
netcdfgroup@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
_______________________________________________
netcdfgroup mailing list
netcdfgroup@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
