<DC0D3CA127373142BB40E5C6FA63B4D102140B29@xxxxxxxxxxxxxxxxxxxxxxxxxx>
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
Thread-Topic: [netcdfgroup] IDL and MATLAB - reading Classic NetCDF Data
Model
Thread-Index: AcmisxQ9UmyXcfIqQFSagdFUwX+NtwABEOCx
References: <894DF23E-5F96-4FF6-A94C-5F17968D0CDC@xxxxxxxx>
<49B91D3D.D209.00AE.0@xxxxxxxxxx>
From: "Mark Rivers" <rivers@xxxxxxxxxxxxxxxxx>
To: "Mark Hadfield" <m.hadfield@xxxxxxxxxx>,
"Joseph VanAndel" <vanandel@xxxxxxxx>
X-Loop: netcdfgroup@xxxxxxxxxxxxxxxx
A few corrections:
> IDL has a built-in netCDF interface. In the current version of IDL,
7.0.x, it does not support netCDF-4. In fact it doesn't even support
netCDF-3 fully, in that it chokes on files > 2 GiB, even those that
are handled by the netCDF 3.6.x library.
That is certainly not universally true. I routinely write and read netCDF
files >2GiB in IDL on 64-bit Linux. There may be some restrictions relative to
the netCDF 3.6.4 library, but certainly a single large array (even without an
unlimited dimension) works fine.
> Matlab does not have a built-in netCDF interface
That is no longer true. Matlab added a native netCDF interface in version 7.7,
according to the Unidata netCDF Web pages and Matlab's Web page.
Mark Rivers
________________________________
From: netcdfgroup-bounces@xxxxxxxxxxxxxxxx on behalf of Mark Hadfield
Sent: Wed 3/11/2009 8:33 PM
To: Joseph VanAndel
Subject: Re: [netcdfgroup] IDL and MATLAB - reading Classic NetCDF Data Model
> 1) Is it true that any program written with the netCDF3 API
> needs minimal (no?) changes to read such a netCDF-4 compressed
> file, once the application has been relinked with the netCDF-4
> library?
Yes that is true. *No* changes required. (Unless I am very much
mistaken, which does happen from time to time.)
> 2) Can anyone tell me if programming languages like IDL or MATLAB
> have been linked with the netCDF-4 library? (I suspect not, but
> I haven't been able to determine this by running 'nm' or 'ldd').
> 3) Has anyone heard discussions about IDL supporting netCDF-4 or
> using HDF5 1.8.1?
IDL has a built-in netCDF interface. In the current version of IDL,
7.0.x, it does not support netCDF-4. In fact it doesn't even support
netCDF-3 fully, in that it chokes on files > 2 GiB, even those that
are handled by the netCDF 3.6.x library.
The next version of IDL, presumably 8.x, was in beta several months
ago, so I expect a release in a few months. I do not know if there
has been any advance in its netCDF support. I submit feature requests
from time to time requesting such advances. You will have to ask ITTVis
about their plans.
Doubtless it would be possible for a third party to write a netCDF
plug-in for IDL to read netCDF-4. As far as I know, no-one has.
> 4) Has anyone heard discussions about MATLAB supporting netCDF-4
> or using HDF5 1.8.1?
Matlab does not have a built-in netCDF interface, but there are (or
have been) several open-source add-ons that provide such an interface.
I do not know if any of these has been updated to support netCDF-4.
Relinking one of these against the netCDF-4 library should be trivial
(see response to question 1).
> 5) What's the "market penetration" of netCDF-4? Does anyone have
> a feel for how many netCDF-3 users have at least relinked with the
> netCDF-4/HDF5 libraries, even if they aren't using the more advanced
> netCDF-4 features, like groups?
Not many, I suspect, but it will come soon. As an example, an ocean
model that I use, ROMS, has recently acquired support for netCDF-4 in
order to access the compression facilities. The ROMS developers are
probably early adopters, however.
My question:
Are you aware that the NetCDF Operators (NCO) suite provides easy
conversion between netCDF-3 and netCDF-4 formats? This can be very
useful in heterogeneous environments.
Mark Hadfield "Kei puwaha te tai nei, Hoea tahi tatou"
m.hadfield@xxxxxxxxxx
NIWA
NIWA is the trading name of the National Institute of Water & Atmospheric
Research Ltd.
