[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20000814: NetCDF::put_att_text not found

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: 20000814: NetCDF::put_att_text not found
  • Date: Mon, 14 Aug 2000 14:57:15 -0600
Ron,

>Date: Mon, 14 Aug 2000 14:43:21 -0600
>From: Ron Kahn <address@hidden>
>Organization: System Development Division
>To: Steve Emmerson <address@hidden>
>Subject: Re: 20000814: NetCDF::put_att_text not found
>Keywords: 200008141705.e7EH5NN04980

The above message contained the following:

> Thanks for your quick response, however I've run into an additional
> inconsistency.  I retrieved the netcdf-2 manual from the link you
> provided in your e-mail and looked at the interface to ncattput as it
> is described in version 2.4 of the NetCDF Users's Guide.  It indicates
> for the c interface, 6 parameters as the interface to the call the
> 5th being the length of the character string and the 6th being a
> pointer to said character string.  When I ran it as the documentation
> seemed to indicate I got a usage error message from the netCDF perl
> module.  Upon experimentation I got the call to work by eliminating
> the length parameter and not making the variable holding the value
> a pointer.  My question is in regards to whether or not I have the
> correct documentation for netcdf-2 or whether I have missed something
> obvious.

You haven't missed anything obvious.  The correspondence between
the netCDFPerl API and the netCDF version 2 API isn't exact -- and
you've found one of the inexact spots.  Since Perl strings know their
length, it's not necessary to supply a string-length argument to
NetCDF::attput().

Unfortunately, this is inconsistent with the way the netCDFPerl API
handles character variables -- where each character is interpreted as a
element in an array and the length must explicitly specified.  In
practice, however, users tend to have string-valued attributes and
byte-valued variables (for the most part) so things are as inconvenient
as they might, otherwise, be.

Regards,
Steve Emmerson   <http://www.unidata.ucar.edu>