Hi again -
I've been using mexnc and the NetCDF Toolbox forever; it can be just
a little painful to port this around to different versions of Matlab on
different
versions of Mac OS X.
If I can write NetCDF without mexnc, I'd like to give it a try, but I was
surprised there wasn't more discussion about it on the list. Maybe all
the other Matlab users are happy with mexnc?
I can live without the redefine function, maybe because my datasets
are all in situ data, and are probably a lot smaller than what you might
be working with.
Thanks very much -
Nan
Nan,
We use our Netcdf-Java Toolbox for Matlab
(http://njtbx.sourceforge.net <http://njtbx.sourceforge.net/>) to
*access* NetCDF, OPenDAP and NcML, as well as utilize CF-Conventions,
and this toolbox relies on NetCDF-Java to do the heavy lifting.
(there will be a new release of these tools later today, BTW...)
For writing NetCDF files, we still use the NetCDF Toolbox or SNCTOOLS,
both of which depend on mex files. But if you have Matlab 2008b or
later, the required mex files are included with your Matlab
distribution from Mathworks -- you don't need to obtain them elsewhere.
The only thing you need for 2008b or later is the m-file distribution
of the NetCDF Toolbox or SNCTOOLS, which I get via SVN following the
instructions at:
http://mexcdf.sourceforge.net/
I know it's possible to write entire files using NetCDF-Java, but my
understanding is that it's still not possible to put an existing file
in redefine mode, so you can't open an existing netcdf file for
writing and start adding things.
-Rich
On Wed, Aug 26, 2009 at 10:41 AM, Nan Galbraith <ngalbraith@xxxxxxxx
<mailto:ngalbraith@xxxxxxxx>> wrote:
Hi All -
It seems that various people are now using a Matlab interface to
the Java
library to create NetCDF files. Is there updated documentation on
how to
do this, or is this 2007 email thread the last word on the subject?
The Unidata documentation still implies that Matlab->Java-NetCDF is
read-only, unless I'm looking in the wrong part of the site.
This may be a moot point soon, with the new version of Matlab
supporting
NetCDF, but we seem to be a long way from acquiring a site license for
that here at WHOI.
Thanks -
Nan
--
*******************************************************
* Nan Galbraith (508) 289-2444 *
* Upper Ocean Processes Group Mail Stop 29 *
* Woods Hole Oceanographic Institution *
* Woods Hole, MA 02543 *
*******************************************************