> Thanks for your help on Fridays question: sorry for pestering you once more!
> (1) I've had ncl installed on an IBM RS6000 cluster running AIX.
> Everything appeared to be ok until I tried loading the gsn_*.ncl files, the
> shea_util.ncl and contributed.ncl files from the generic location (if I
> recall correctly it's $NCARG_ROOT/lib/ncarg/nclscripts/csm/). The ncl
> interpeter then gave me all kinds of non-fatal error messages related to
> those files being read incorrrectly.
> However - when using ncl from the command line the same files loaded without
> any complaints from the ncl interpreter.
In situations like this, it helps if you send us the code and the
error messages. There could be any number of things happening: the
scripts could be getting loaded in the wrong order, there could be
something wrong with the scripts (funny characters), there could be a
bug in the scripts, the load command may be in the wrong place, etc.
> The following problem occurs on my server running red-hat linux:
> (2) A second, and potentially related issue, occurs when I try to point the
> NCL_DEF_SCRIPTS_DIR environmental variable to a directory containing the
> latest versions of the gsn_*.ncl etc. files that I want to "auto-load" into
> my scripts. I recieve all kinds of non-fatal complaints from the ncl
> interpreter as in problem (1). This also fails when I point
> NCL_DEF_SCRIPTS_DIR to the versions of the gsn*.ncl scripts installed in the
> generic $NCARG_ROOT/lib etc location.
Again, we would need to see the script and the error messages. You
have to be careful that you don't have duplicate versions of files in
the same directory. For example, if you have "gsn_code.ncl" and
"gsn_code.new.ncl", and one is a newer version of the other, you may
run into problems because they both contain the same functions.
ncl-talk mailing list
ncl-talk AT ucar.edu
This archive was generated by hypermail 2b29 : Tue Feb 18 2003 - 07:54:39 MST