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

[Support #KOM-880811]: chunksize, NC_STRING questions

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [Support #KOM-880811]: chunksize, NC_STRING questions
  • Date: Wed, 13 May 2009 10:48:59 -0600
Charlie- So I did some checking and here is the
situation as I understand it.

The original ncgen program treats
single char constants as bytes, not chars.
so :x = 'a', 'b'; will have type NC_BYTE.
The same is true for ncgen4.

The ncgen4 program, when creating a netcdf-4 file,
currently treats :x="abc"; as creating an attribute
with type NC_STRING. In the same situation, the
original ncgen assumes the type NC_CHAR.
This is a backward compatability problem
and I will fix it shortly so ncgen4 acts like ncgen.

The consequence of this is that if one wants to create
an attribute whose type is NC_STRING, then one
must explicitly type the attribute like so:
    string :x="abc";

It turns out however that there is an error in ncdump
related to this. If I create a .nc file using the above
    string :x="abc";
and then dump it, it appears that ncdump outputs
    :x="abc";
instead of the typed
    string :x="abc";
If the ncdump output were then passed again to ncgen4,
x would end up typed as NC_CHAR instead of NC_STRING.

=Dennis Heimbigner


Ticket Details
===================
Ticket ID: KOM-880811
Department: Support netCDF
Priority: Normal
Status: Open