Re: python interface


Dear Magnus,

I'll try to clear up the PyNIO issue, as it is my team that develops
and supports this software. I am not familiar enough with ScientificPython
to know what the status is of this for numpy.

You're right in that PyNIO is only available in binary form. We are
currently in the process of making it open source so that you can
download and install the software on any version of python you want.

In order to build PyNIO from source, you will need to install some
external software, like NIO (which is our I/O library), netCDF and HDF
(HDFEOS/gctp and/or grib2c/jasper are needed if you want support for
HDF-EOS2 and/or GRIB2 files).

I can provide you with the source code for PyNIO, NCL, and NCARG if you
want to try building these from source code. You will need to install
netCDF version 3.6.x, HDF4.2r1, JPEG-6b, and zlib 1.2.3 first.

Let me know if you'd like to give this a try.

--Mary

On Mon, 21 May 2007, Magnus Hagdorn wrote:

Hi all,
I am in the process of reviewing the python interface to netCDF I use
for accessing our data files. I am extremely puzzled now.

Up to now I have always used the ScientificPython module, which does
more or less what I want it to. However, the website seems to have
disappeared. Some searching around revealed the download page:
http://sourcesup.cru.fr/projects/scientific-py/
ScientificPython seems to be under active development.

I am also a user of some form of numerical array extension to python. I
started with Numeric, then ported my code to numarray. However, it seems
Numpy is the extension to use.

Further looking around it seem the only netCDF interface supporting
Numpy is PyNIO. However, this package seems to only come as a binary and
is compiled only using Python 2.4. This is not really acceptable since I
really don't want to depend on some package which I cannot change if I
have to and which forces me to use a specific python version.

So I am wondering what python package for netCDF I/O other can
recommend.

Regards
magnus


==============================================================================
To unsubscribe netcdfgroup, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================



==============================================================================
To unsubscribe netcdfgroup, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================