In the interests of avoiding possible confusion, I'd like to point out
that there are already existing R packages on CRAN that implement these
functions. The package "ncdf" interfaces to the netCDF library, and
"udunits" interfaces to the udunits library. (Disclaimer: I'm the author
of both those packages.) These packages have been available for a year or
so, and I update them when I get bug reports. I presume the new RNetCDF
package is not compatible with either of the existing packages, so I
suggest people not try to mix calls to them.
Regards,
--Dave
---------------------------------------------------------------
David W. Pierce / Climate Research Division
Scripps Inst. Oceanog. / (858) 534-8276 (voice)
dpierce@xxxxxxxx / (858) 534-8561 (fax)
---------------------------------------------------------------
Pavel Michna said:
> Dear all,
>
> I just released version 1.0 of the RNetCDF R package on CRAN
> (Comprehensive R Archive Network, http://cran.at.r-project.org/)
> under the terms of the GNU General Public License (GPL).
>
> This package provides an interface to Unidata's NetCDF library functions
> (version 3) and furthermore access to Unidata's udunits calendar
> conversions. The routines and the documentation follow the NetCDF and
> udunits C interface, so the corresponding manuals can be consulted for
> more detailed information.
>
> The main aims of this package are:
> - to provide full read/write access to all NetCDF datasets (supporting
> all external NetCDF data types)
> - to provide a low level interface which allows the user to write his
> own (customized) NetCDF read/write functions in a simple way
>
> This package requires Unidata's NetCDF and the undunits library already
> installed on the system.
>
>
> Feedback is greatly appreciated.
>
> Pavel
>
>
> I would like to thank Juerg Schmidli for his excellent comments and
> suggestions during the development process of this package and This
> Rutishauser for testing the package from the user's perspective.
>
>
> ------------------------------------------------------------------------
>
> Pavel Michna (PhD Student) Tel: +41 (0)31 631 85 42
> Climatology and Meteorology (KLIMET) Fax: +41 (0)31 631 85 11
> Institute of Geography E-Mail: michna@xxxxxxxxxxxxx
> University of Bern
> Hallerstrasse 12
> CH-3012 Bern/Switzerland
>
> ------------------------------------------------------------------------
>
>
> ------------------------------------------------------
> This mail was sent through IMP at http://mail.unibe.ch
>
>
>From owner-netcdfgroup@xxxxxxxxxxxxxxxx Tue 3 2004 Aug 00:51:01
Message-ID: <1091487061.410ec555813ca@xxxxxxxxxxxxxxx>
Date: Tue, 3 Aug 2004 00:51:01 +0200
From: Pavel Michna <michna@xxxxxxxxxxxxx>
In-Reply-To: <3105.132.239.114.8.1091464186.squirrel@xxxxxxxxxxxxxxxx>
To: dpierce@xxxxxxxx
Subject: Re: RNetCDF -- R Interface to NetCDF Datasets
Received: (from majordo@localhost)
by unidata.ucar.edu (UCAR/Unidata) id i72MqlN4003306
for netcdfgroup-out; Mon, 2 Aug 2004 16:52:47 -0600 (MDT)
Organization: UCAR/Unidata
Keywords: 200408022252.i72MqjaW003215
Cc: netcdfgroup@xxxxxxxxxxxxxxxx
References: <1091441292.410e128ccbadf@xxxxxxxxxxxxxxx>
<3105.132.239.114.8.1091464186.squirrel@xxxxxxxxxxxxxxxx>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-15
Content-Transfer-Encoding: 8bit
User-Agent: Internet Messaging Program (IMP) 3.1
X-Originating-IP: 212.254.40.162
X-Virus-checked: by University of Berne
Sender: owner-netcdfgroup@xxxxxxxxxxxxxxxx
Precedence: bulk
Reply-To: Pavel Michna <michna@xxxxxxxxxxxxx>
Quoting David Pierce <dpierce@xxxxxxxx>:
> In the interests of avoiding possible confusion, I'd like to point out
> that there are already existing R packages on CRAN that implement these
> functions. The package "ncdf" interfaces to the netCDF library, and
> "udunits" interfaces to the udunits library. (Disclaimer: I'm the author
> of both those packages.) These packages have been available for a year or
> so, and I update them when I get bug reports. I presume the new RNetCDF
> package is not compatible with either of the existing packages, so I
> suggest people not try to mix calls to them.
That's correct. I'm sorry that I forgot to mention this. However both packages
(ncdf and netCDF) are acknowledged in the manual.
Of course the new package is not compatible with the existing two -- otherwise
it wouldn't make sense to create a new one.
To avoid more confusion, I just want to point out the important differences:
- The "RNetCDF" package starts strightforward from the NetCDF C interface,
whereas the "ncdf" package goes a more object-oriented way.
- All external data types are covered for variables and attributes in the
"RNetCDF" package, whereas the "ncdf" package does at least not support
attributes of type NC_BYTE for read/write. To my knowledge, the "netCDF"
package does not support NC_BYTE at all.
- The new package contains functions for renaming, copying and deleting
attributes and for renaming dimensions and variables, which is not covered
by the two other packages.
- Regarding the "udunits" package: I just implemented the calendar conversions
because that is often needed. All other udunits functionality is covered
exclusively by the "udunits" package.
Regards,
Pavel
------------------------------------------------------------------------
Pavel Michna (PhD Student) Tel: +41 (0)31 631 85 42
Climatology and Meteorology (KLIMET) Fax: +41 (0)31 631 85 11
Institute of Geography E-Mail: michna@xxxxxxxxxxxxx
University of Bern
Hallerstrasse 12
CH-3012 Bern/Switzerland
------------------------------------------------------------------------
------------------------------------------------------
This mail was sent through IMP at http://mail.unibe.ch
