[Fwd: Re: [netcdf-java] announce: stable release netcdf-java 4.0] (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, 27 Feb 2009 13:54:15 -0700
From: Bill Moninger <address@hidden>
To: Robb Kambic <address@hidden>
Subject: [Fwd: Re: [netcdf-java] announce: stable release netcdf-java 4.0]

Hi Robb,

this is a different issue than the parameter 189 issue, which I've solved on my 
end by using a new fsl_2_0.tab. I'm checking with folks here about why 
fsl_2_0.tab has two parameters 142, and will get back to you separately on 
that.

The issue with the operational RUC grib2 file (from NCEP, as you noticed) is 
that netcdf-java-4 doesn't seem to see the hybrid (vertical) level, although it 
*seems* to read the file fine. Please see the ncfile.toString() output below, 
and note that the variable

variable float Pressure(time=1, y=225, x=301);

should have an additional dimension ('hybrid'), according to the wgrib2 dump.

Sorry you got hit with two separate issues from us at the same time.

Feel free to call me (202-497-6435) if you wish, and maybe we can get this 
resolved faster.

-Bill

On 2/27/2009 11:53 AM, Robb Kambic wrote:
> Bill,
>
> this is not a FSL file, it decoded fine on our decoder
> and it doesn't have parameter 189.
>
> What am I missing here?
>
> Robb...
>
> Fri Feb 27 11:48:59 MST 2009 ... Start of GribIndexer
> index_version = 6.0
> grid_edition = 2
> location = 0905712000000
> length = 17512783
> created = 2009-02-27T18:48:59Z
> center = 7
> sub_center = 0
> table_version = 1

-------- Original Message --------
Subject: Re: [netcdf-java] announce: stable release netcdf-java 4.0
Date: Thu, 26 Feb 2009 10:50:24 -0700
From: John Caron <address@hidden>
To: Bill Moninger <address@hidden>
CC: address@hidden
References: <address@hidden> <address@hidden>

Hi Bill:

Usually the problem is in tyhe tables. Can you send us a sample file it fails 
on? Our Grib expert,
Robb Kambic, can have a look.Ill

Ill fix the download problem, thanks for reporting.

Bill Moninger wrote:
> Hi John,
>
> two items: a small one and a (maybe) big one.
>
> First, on page
> http://www.unidata.ucar.edu/downloads/netcdf/netcdf-java-4/index.jsp
> the link to netcdfAll-4.0.jar actually points to netcdf-4.0.jar.
>
> Second, the latest netcdfAll-4.0.jar does not read the RUC grib2 files
> correctly, at least as I'm using it. It doesn't find the hybrid dimension.
>
> Here's a snip from ncfile.toString():
> ----------------------------------------------------
> ncfile is netcdf /public/data/grids/ruc/hyb_A252/grib2/0905622000003 {
>  dimensions:
>    time = 1;
>    y = 225;
>    x = 301;
>    depth_below_surface = 5;
>    depth_below_surface1 = 1;
>    depth_below_surface2 = 2;
>    depth_below_surface3 = 1;
>    height_above_ground = 1;
>    height_above_ground1 = 1;
>    pressure_difference_layer = 1;
>    bounds_dim = 2;
>  variables:
> ...
>  float Pressure(time=1, y=225, x=301);
>      :units = "Pa";
>      :long_name = "Pressure @ hybrid";
>      :missing_value = -9999.0f; // float
>      :grid_mapping = "Lambert_Conformal";
>      :GRIB_param_discipline = "Meteorological_products";
>      :GRIB_param_category = "Mass";
>      :GRIB_param_name = "Pressure";
>      :GRIB_param_id = 2, 0, 3, 0; // int
>      :GRIB_product_definition_type = "Analysis/forecast at horizontal
> level/layer";
>      :GRIB_level_type = 105; // int
> --------------------------------------------------------
>
> And here's some output from wgrib2 on the same file:
> > wgrib2 /public/data/grids/ruc/hyb_A252/grib2/0905622000003|more
> 1:0:d=2009022522:PRES:1 hybrid level:3 hour fcst:
> 2:70425:d=2009022522:PRES:2 hybrid level:3 hour fcst:
> 3:140762:d=2009022522:PRES:3 hybrid level:3 hour fcst:
> 4:211202:d=2009022522:PRES:4 hybrid level:3 hour fcst:
> 5:281657:d=2009022522:PRES:5 hybrid level:3 hour fcst:
> 6:351773:d=2009022522:PRES:6 hybrid level:3 hour fcst:
> 7:421147:d=2009022522:PRES:7 hybrid level:3 hour fcst:
> 8:489799:d=2009022522:PRES:8 hybrid level:3 hour fcst:
> 9:557601:d=2009022522:PRES:9 hybrid level:3 hour fcst:
> 10:624657:d=2009022522:PRES:10 hybrid level:3 hour fcst:
> 11:691379:d=2009022522:PRES:11 hybrid level:3 hour fcst:
> 12:758346:d=2009022522:PRES:12 hybrid level:3 hour fcst:
> 13:825473:d=2009022522:PRES:13 hybrid level:3 hour fcst:
> 14:892604:d=2009022522:PRES:14 hybrid level:3 hour fcst:
> 15:959705:d=2009022522:PRES:15 hybrid level:3 hour fcst:
>
> Is this a problem with our grib tables not having information about
> hybrid levels?
>
> -Bill
>
> On 2/19/2009 5:28 PM, John Caron wrote:
> > Version 4.0 is now the the stable release for the Netcdf-Java library.
> > It is available at
> >
> >   http://www.unidata.ucar.edu/software/netcdf-java/
> >
> > Version 4.0 continues to evolve with new features. However, a stable
> > release means that we will keep
> > future changes to the API backwards compatible to the public API at:
> >
> >  http://www.unidata.ucar.edu/software/netcdf-java/v4.0/javadoc/index.html
> >
> > Version 2.2 is on "maintenence only" mode. We encourage you to switch
> > to 4.0 to take advantage of
> > many significant improvements and bug fixes.
> >
> > Please report problems, bugs and suggested improvements to
> > address@hidden.
> >
> > Thanks!
> >
> > ------------------------------------------------------------------------
> >
> > There are API changes that you may see when you start using 4.0 with
> > an application that has been
> > using 2.2. These are mostly minor and in all cases have equivalent or
> > improved functionality in 4.0.
> > Most of the changes are in implementation classes not often used by
> > applications. More info is at:
> >
> >   http://www.unidata.ucar.edu/software/netcdf-java/transition4.html
> >
> > The license for netcdf-java 4.0 has been changed from LGPL to the
> > "MIT-style" license used by the
> > netCDF C library. This is intended to remove perceived barriers to use
> > by some commercial companies.
> > We anticipate continued contributions of bug fixes and enhancements by
> > the community.
> > _______________________________________________
> > netcdf-java mailing list
> > address@hidden
> > For list information or to unsubscribe, visit:
> > http://www.unidata.ucar.edu/mailing_lists/ 

--
William R. Moninger         http://www-frd.fsl.noaa.gov/~moninger/
NOAA / Earth Systems Research Laboratory / Global Systems Division
325 Broadway, R/GSD1                           voice: 303-497-6435
Boulder, CO 80305                              fax:   303-497-3329