Re: timestamp and Plain

• To: Steve Emmerson <steve@xxxxxxxxxxxxxxxx>
• Subject: Re: timestamp and Plain
• From: "John J Brecht" <john.brecht@xxxxxxx>
• Date: Fri, 21 Apr 2000 15:05:43 -0500

Thanks!

This appears to be working.  I'll let you know if anything goes wrong.

-john


Steve Emmerson wrote:
> 
> John,
> 
> >Date: Thu, 20 Apr 2000 10:51:23 -0500
> >From: "John J Brecht" <john.brecht@xxxxxxx>
> >Organization: SRI
> >To: VisAD List <visad-list@xxxxxxxxxxxxx>
> >Subject: timestamp and Plain
> >Keywords: 200004201752.LAA24216
> 
> In the above message, you wrote:
> 
> > when I open netCDF files I've create using Plain, I get this error:
> >
> > Couldn't decode attribute Time:units="s @ -9.783072E8": time unit with
> > non-timestamp origin
> 
> This has been a problem with the "units" subset of the VisAD package for
> a while now.  Units such as the above were something of second class
> citizens.  I've been waiting for a push to correct this situation -- so
> thank you for giving it to me. :-)
> 
> Please find 3 files enclosed:
> 
>     Unit.java
>     TimeScaleUnit.java
>     data/netcdf/units/UnitParser.java
> 
> You should compile them and make them available to your JVM.  They
> should enable the VisAD package to correctly parse unit specifications
> such as the above -- but, on output, will create unit specifications
> like this:
> 
>     s since 1970-01-01 00:00:00.000 UTC
> 
> So, your old netCDF files should still work and any new one you create
> should have the "better" format.
> 
> Be advised that this is alpha-test software (I wanted to get you a
> solution quickly).  Please let me know of any problems.
> 
> These new files will be in the next release of the VisAD package.
> 
> Regards,
> Steve Emmerson   <http://www.unidata.ucar.edu>
>