Hi Matt and Charlie,
Correct me if I am wrong, but when I compile netcdf-c v.4.9.0 with the option
--with-plugin-dir, then this is the default location for the HDF5 plugins. Only
when a user has its own plugins it should extent the search path using
HDF5_PLUGIN_PATH. Then when I compile netcdf-fortran or nco (the latter
compiled without the option --with-hdf5-plugin-path) against this version of
the netcdf-c library, they will also look at the same default location or use
the HDF5_PLUGIN_PATH search path.
As a maintainer, I would discourage (non-expert) users to use set the
environment variable HDF5_PLUGIN_PATH. And keep all the necessary plugins at
one place.
Greetings, Richard
On 04-08-2022 17:25, Charlie Zender wrote:
[EXTERNAL EMAIL]
Matt, Richard:
I mis-spoke about NCO's support for --with-hdf5-plugin-path
I thought this was a custom option implemented by netCDF.
(I have always used the environment variable method for NCO).
Now I realize it is a generic option provided by configure,
and thus is available to NCO at build time as well.
There is no harm in building NCO with --with-hdf5-plugin-path.
It is probably more robust to build with --with-hdf5-plugin-path,
that way users who do not know about the environment variable
will get predictable behavior from your builds.
Charlie
--
Charlie Zender, Earth System Sci. & Computer Sci.
University of California, Irvine 949-891-2429 )'(
_______________________________________________
NOTE: All exchanges posted to Unidata maintained email lists are
recorded in the Unidata inquiry tracking system and made publicly
available through the web. Users who post to any of the lists we
maintain are reminded to remove any personal information that they
do not want to be made public.
netcdfgroup mailing list
netcdfgroup@xxxxxxxxxxxxxxxx<mailto:netcdfgroup@xxxxxxxxxxxxxxxx>
For list information or to unsubscribe, visit:
https://www.unidata.ucar.edu/mailing_lists/
