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

20000218: gninfo for SUN

>From: address@hidden
>Organization: .
>Keywords: 200002182111.OAA13004

Glenn,

My directory structure on disk looks like:

WEST-CONUS/
           1km/VIS/
                   VIS_000222_1045
                   VIS_000222_1100
                   VIS_000222_1115
           4km/IR/
                   IR_000222_1045
                   IR_000222_1100
                   IR_000222_1115
           4km/12.0/
                   12.0_000222_1045
                   12.0_000222_1100 
                   12.0_000222_1115
           4km/3.9/
                   3.9_000222_1045
                   3.9_000222_1100
                   3.9_000222_1115
           8km/WV/
                   WV_000222_1045
                   WV_000222_1100
                   WV_000222_1115

The program which injects the GINI imagery into the product queue creates the 
product
names as outlined below. The ch2 refers to noaaport channel two, not the band 
"vis" as 
referenced below. Satellite imagery is found on NOAAport channels 1,2 and 4. The
platform ID is given based on the table 4.5, the time from bytes 9-15, the 
sector
ID from table 4.6, and the band from table 4.7.

With the products inserted as shown below, the pqact I use for all products is 
simply:
^sat/ch[0-9]/.*/(.*)/([12][0-9])([0-9][0-9])([01][0-9])([0-3][0-9]) 
    ([0-2][0-9])([0-5][0-9])/(.*)/(.*km)/
(line wrapped above)

        FILE    data/gempak/nport/\8/\9/\1/\1_\3\4\5_\6\7

As I mentioned before, if you use the WMO type entries for your pattern 
matching,
you can use separate entries for every product like:
^TIGW01 ([0-3][0-9])([0-2][0-9][0-5][0-9])
        FILE    data/gempak/nport/WEST-CONUS/1km/VIS/VIS_(\1:yy)(\1:mm)\1_\2
^TIGE01 ([0-3][0-9])([0-2][0-9][0-5][0-9])
        FILE    data/gempak/nport/EAST-CONUS/1km/VIS/VIS_(\1:yy)(\1:mm)\1_\2
etc.

The thing here being that WMO headers aren't very useful for describing the
directory structure, so you have to code every GW01, GW02, GW03, GW04, GW05 for
VIS, IR, WV, etc. Rather, I have done that work at the begining when the 
products 
are placed into the LDM product queue as they are read from the NOAAPORT 
receiver
so that end processing is easier.

Or, you can file all the products with the WMO names, and have a script which
moves the products into the directory structure using the tables from the
NESDIS interface control document. The problem with this approach is that 
the system becomes brittle when changes are made to the data stream.

Steve Chiswell

>
>
>Thanks Steve,
>Not sure I understand.... So, your directory structure on disk looks like:
>sat/ch2/GOES-10/VIS/20000218 1930/NHEM-COMP/24km/   plus the TIG sat file
>(even tho ch2 is not vis!  Is it rather near IR?)
>
>And this is done in your pqact? Or do you ID and give 'better' names
>upstream from pqact?
>Thanks for your patience here......
>
>
>
>
>
>Unidata Support <address@hidden> on 02/18/2000 03:43:03 PM
>                                                              
>                                                              
>                                                              
> To:      Glenn Rutledge/NCDC                                 
>                                                              
> cc:      address@hidden                     
>                                                              
>                                                              
>                                                              
> Subject: 20000218: gninfo for SUN                            
>                                                              
>
>
>
>
>
>
>
>
>>From: address@hidden
>>Organization: .
>>Keywords: 200002181846.LAA08727
>
>>
>>
>>I see that Jim Cowie et.al., wrote a nice script to covnert GINI to the
>>filenaming conventions expected in GARP called gdinfo.  Is this avbl for a
>>SUN OS?
>>
>>
>>        ****** Signature Tag ******
>>
>>        National Climatic Data Center
>>        Climate Data Division
>>        (828) 271-4097
>>        address@hidden
>>
>>
>>
>
>Glenn,
>
>You probably intended this to go to COMET, which is where Jim
>worked, and probably where his script was.
>
>I am injecting the GINI images into our LDM here using the octets from
>the GINI header to give the products more informative names so that
>pur pqact just files the data directly without having to rename things-
>our LDM products look like:
>
>sat/ch2/GOES-10/VIS/20000218 1930/SUPER-NATIONAL/8km/ TIGN01 KNES 181930
>sat/ch2/GOES-10/VIS/20000218 1930/NHEM-COMP/24km/ TIGF01 KNES 181930
>So all of the info needed for the $SAT directory structure is
>available in the product names.
>
>If COMET uses a script, it probably uses the WMO header and matches
>the values from the tables 4.4-4.12 from the NESDIS interface control
>document.
>
>Steve Chiswell
>***************************************************************************
>Unidata User Support                                    UCAR Unidata
>(303)497-8644                                                  P.O. Box
>address@hidden                                   Boulder, CO
>---------------------------------------------------------------------------
>Unidata WWW Service                        http://www.unidata.ucar.edu/
><
>***************************************************************************
>
>
>