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

[netCDF #HLY-318623]: Combining conventions

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [netCDF #HLY-318623]: Combining conventions
  • Date: Thu, 04 Feb 2010 12:51:47 -0700
Hi Mike,

> We are developing a NetCDF convention which will combine the CF-1.4
> convention with a radar convention, called CfRadial.
> 
> Is there a standard way to represent this in a string. For example,
> would it be
> 
> CF-1.4/CfRadial
> 
> or
> 
> CF-1.4:CfRadial
> 
> or what?

Documentation for the global "Conventions" attribute specifies the
first alternative above, both on the conventions web page:

  http://www.unidata.ucar.edu/netcdf/conventions.html

and also in the Attribute Conventions chapter of the Users Guide in
the current snapshot release:

  
http://www.unidata.ucar.edu/netcdf/docs_snapshot/netcdf.html#Attribute-Conventions

If you instead just listed the two conventions using

  :Conventions = "CF-1.4 CfRadial";

it would still mean that the data conformed with both conventions, but
wouldn't convey the extra information that the CfRadial conventions
are purely an extension of CF-1.4.  Maybe you don't need to convey
that information anywhere except in the CfRadial conventions document,
in which case I would recommend this alternative.  It also allows for
more flexibility in evolution of the CfRadial conventions, in case
something is added in the future that overrides something in the
CF-1.4 conventions.  But note that there are no rules about which of
multiple conventions are supposed to be followed if the specified
conventions are incompatible in any way.

So this is an issue that needs some clarification.  Because it
involves multiple conventions, it's not really an issue that can be
decided at the CF level, for example.  I think that means we have to
decide how to handle it.  I propose specifying something simple, such
as: later conventions override earlier ones whenever there is a 
conflict.

--Russ

Russ Rew                                         UCAR Unidata Program
address@hidden                      http://www.unidata.ucar.edu



Ticket Details
===================
Ticket ID: HLY-318623
Department: Support netCDF
Priority: Normal
Status: Closed