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

Re: 20031003: gribtocdl not writing data (fwd)




===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Fri, 3 Oct 2003 12:10:26 -0400 (EDT)
From: Lori Thompson <address@hidden>
To: Robb Kambic <address@hidden>
Subject: Re: 20031003: gribtocdl not writing data

Hi Robb. I think I understand gribtocdl a little better now, thanks! I
believe I received the segmentation fault because I did not reset
Lo2 and La2 to the new values after I deleted the additional lat/lon
points. I'm not sure if that was the reason but I reran it and the data looks
good. The only thing is the vertical structure of the 3D variables may not
have been retained, but I'm looking into that now.

Thank you for your help.

Lori

On Fri, 3 Oct 2003, Robb Kambic wrote:

> On Fri, 3 Oct 2003, Lori Thompson wrote:
>
> > So the cdl file only contains information on the coordinates and not
> > the data for the variables?
>
> Lori,
>
> That's correct, cdl are only template files for the netCDF files.
>
>
> I have done ascii to NetCDF conversions
> > where the cdl file takes the form:
> >
> > data:
> > lev = ;
> > lon = ;
> > lat = ;
> > var1 = ;
> > var2 = ;
> > .
> > .
> > .
> > varn = ;
> > }
> >
> > So are you saying that gribtocdl will create a cdl file with only:
> > data:
> > lev = ;
> > lon = ;
> > lat = ;
> > }
> >
> > I used gribtonc following gribtocdl and received the error message:
> > ncgen: isabel13l.2003091100.grib3rd.f00.cdl line 258: too many values for
> > this variable, 225 >= 225
> >
> > This is referring to the lon points which have 226. The same is true for
> > lat. I manually removed the extra lon/lat points and reran gribtonc and
> > received a segmentation fault. I did this for 225 and 224 point.
> >
>
> I hate to say this but gribtocdl isn't perfect becasue there are so many
> variations of GRIB files, it's used as a starting point as you found out.
> Deleting the extra lat/lon points etc was the correct thing to do.
>
> If you send me the cdl file and a grib product, I'll take a look at it.
> Also what version of the decoders package are you using?  One version did
> do segmentation fault on some GRIB products but I thought it was fixed.
> This could be a good test case for me.
>
> Robb...
>
>
> >
> > Thanks,
> > Lori
> >
> >
> > I did have a misunderstanding of gribtocdl. I thought the cdl
> > file was supposed to contain
> >
> > On Fri, 3 Oct 2003, Robb Kambic wrote:
> >
> > > On Fri, 3 Oct 2003, Unidata Support wrote:
> > >
> > > >
> > > > ------- Forwarded Message
> > > >
> > > > >To: address@hidden
> > > > >From: Lori Thompson <address@hidden>
> > > > >Subject: gribtocdl not writing data
> > > > >Organization: NOAA
> > > > >Keywords: 200310031240.h93CeHk1011045 gribtonc gribtocdl
> > > >
> > > > Hello. I'm using gribtocdl and gribtonc to convert grib files to NetCDF.
> > > > The grib data files have 7 variables, 5 of which have a vertical
> > > > structure. gribtocdl creates the cdl file and returns the message:
> > > >
> > > > Oct 03 12:38:02 gribtocdl[10036]: Starting Up
> > > > Oct 03 12:38:04 gribtocdl[10036]: EOF on input
> > > > Oct 03 12:38:04 gribtocdl[10036]: Exiting
> > > > Oct 03 12:38:04 gribtocdl[10036]: 111 WMO msgs, 111 GRIBs decoded, 111
> > > > written
> > > >
> > > > which contains the correct number 111. The cdl file writes out the 
> > > > lat/lon
> > > > and time data correctly and has the correct metadata for the variables.
> > > > However, the cdl file does not write out any data for the variables and
> > > > ends with lat data.
> > >
> > > Lori,
> > >
> > > I think you have a slight misunderstanding, gribtocdl just creates cdl
> > > files with only data of the projection/lat/lon/etc.  To actually create a
> > > netCDF file one has to use gribtonc with the cdl created by gribtocdl. ie
> > >
> > > % gribtonc -l - your.cdl your.nc < your.gribdata
> > >
> > > This will create a your.nc netCDF file. gribtonc also needs the ncgen
> > > utility to create the template your.nc from the your.cdl. ncgen is
> > > included in the netCDF package.  I wasn't quite sure of your question, let
> > > me know if I misinterpreted it.
> > >
> > > Robb...
> > >
> > > >
> > > > I am running the utility on a Linux machine.
> > > >
> > > > Any suggestions on why this is occuring?
> > > >
> > > > Thanks,
> > > > Lori
> > > >
> > > >
> > > >
> > > > ------- End of Forwarded Message
> > > >
> > > >
> > >
> > > ===============================================================================
> > > Robb Kambic                                  Unidata Program Center
> > > Software Engineer III                        Univ. Corp for Atmospheric 
> > > Research
> > > address@hidden               WWW: http://www.unidata.ucar.edu/
> > > ===============================================================================
> > >
> >
> > --
> > ----------------------------------------------------
> > Lori A. Thompson
> > Applications Programmer/RS Information Systems, Inc.
> > Geophysical Fluid Dynamics Laboratory/NOAA
> > Princeton University Forrestal Campus/Route 1
> > PO Box 308
> > Princeton, NJ 08542
> > Phone: 609-452-6563
> > Fax:   609-987-5063
> > http://www.gfdl.noaa.gov/~lat
> >
> >
>
> ===============================================================================
> Robb Kambic                              Unidata Program Center
> Software Engineer III                    Univ. Corp for Atmospheric Research
> address@hidden                   WWW: http://www.unidata.ucar.edu/
> ===============================================================================
>

-- 
----------------------------------------------------
Lori A. Thompson
Applications Programmer/RS Information Systems, Inc.
Geophysical Fluid Dynamics Laboratory/NOAA
Princeton University Forrestal Campus/Route 1
PO Box 308
Princeton, NJ 08542
Phone: 609-452-6563
Fax:   609-987-5063
http://www.gfdl.noaa.gov/~lat