Re: Another conversion question for Russ

Quincey,

> > You're right, loss of precision is not an overflow.  Again, the
> > C/Fortran assignment model is followed, in which there is no error or
> > warning when trying to assign a float value to an int, merely
> > truncation.
>     Ok, I'll include a parameter in the callback call that indicates why the
> callback routine was called (overflow, loss of precision, etc.), which should
> allow the nc4 overflow routine to make the correct judgement call.

Great, that's ideal.  Glenn and I originally argued about how netCDF
should handle this and eventually concluded that in the rare case that
users request floating point data as integers, they know what they're
doing and want it truncated.

--Russ

>From owner-netcdf-hdf@xxxxxxxxxxxxxxxx 25 2003 Oct -0600 06:43:04 
Message-ID: <wrxismdxzc7.fsf@xxxxxxxxxxxxxxxxxxxxxxx>
Date: 25 Oct 2003 06:43:04 -0600
From: Ed Hartnett <ed@xxxxxxxxxxxxxxxx>
To: netcdf-hdf@xxxxxxxxxxxxxxxx
Subject: how often does HDF binary format change?
Received: (from majordo@localhost)
        by unidata.ucar.edu (UCAR/Unidata) id h9PCh6PT026481
        for netcdf-hdf-out; Sat, 25 Oct 2003 06:43:06 -0600 (MDT)
Received: from rodney.unidata.ucar.edu (rodney.unidata.ucar.edu 
[128.117.140.88])
        by unidata.ucar.edu (UCAR/Unidata) with ESMTP id h9PCh5Ob026477
        for <netcdf-hdf@xxxxxxxxxxxxxxxx>; Sat, 25 Oct 2003 06:43:05 -0600 (MDT)
Organization: UCAR/Unidata
Keywords: 200310251243.h9PCh5Ob026477
Lines: 12
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-netcdf-hdf@xxxxxxxxxxxxxxxx
Precedence: bulk


Greetings to HDF HQ from NetCDFville.

I wonder how often you guys have to change the underlying binary
format of the data in support of new releases of HDF5?

And is there some format version number in the data file? I presume
there is.

Thanks,

Ed