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

20011130: KWSI station of origin and arbitrary directories in $LDMHOME



Stonie,

I sent my 2 cents to the ldm users list.

One option for the FPCN00 products is to use the system time in
the file action rather than trying to use the mangled product header.

For example:

#
# use wmo bulletin time here
FPCN(0[^0]|[1-9][0-9]) .... ([0-3][0-9])([0-2][0-9])([0-6][0-9])
        FILE    data/(\2:yyyy)(\2:mm)\2\3_FPCN\1

#
# use system time to file these products
FPCN00
        FILE     data/%Y%m%d%H_FPCN00

Chiz




>From: "Stonie R. Cooper" <address@hidden>
>Organization: Planetary Data, Incorporated
>Keywords: 200111302218.fAUMHRN08556

>Stephen,
>
>I will be surprised if you get any bites on the ldm-users list.
>
>Five years ago, while at TWC, we feed LDM (I believe version 5.04) via a WSI 
>DAS, and later, an HCSN.  To cut to the chase, WSI creates it's own products, 
>also, and adds them to their DAS/HCSN stream.  Being internal products, I 
>guess they don't feel the need to conform to WMO standard . . . 
>
>I could chase the code to see if I am correct, but I surmise that the LDM 
>engine is passing through the WMO header, and checking for validity.  In the 
>example you show, it starts off O.K. - assigning FD to the TT, CN to the AA, 
>but . . . whoa, O is not a digit, which is illegal.  So, as the pqcat 
>indicates, a 00 is put in the ii place, and the O is assumed to be part of 
>the CCCC field of the WMO.  Followed by a space, the O is taken as the 
>complete CCCC.  Also note that the DDGGgg section is now all off, also, so 
>LDM appears to assign system time to the DDGGgg group - suggested by the 
>301828 given by the pqcat, even though the header has 302000.
>
>That doesn't necessarly help you . . . but it poses two questions:
>
> 1) are the KWSI products important to your operations?
> 2) can you put up with the extraneous messages in the LDM dir?
>
>If the KWSI messages have no value to you, you could filter them in your 
>pqact entry.  If they are valuable, you may want to try more robust pqact 
>entries to try to make sure the errant messages don't end up in your LDM home.
>
>I copied support@unidata for archive sake (in case someone comes upon this 
>later), but I would be glad to discuss this further betwix us if you like.
>-- 
>Stonie R. Cooper,
>Science Officer
>Planetary Data, Incorporated
>3495 Liberty Road
>Villa Rica, Georgia  30180
>ph. (770) 456-0700; pg. (888) 974-5017; fx. (770) 459-0016
>
>
>On Friday 30 November 2001 21:22, Stephen Sinnis wrote:
>> I have recently noticed many new subdirectories being created in the
>> $LDMHOME directory.  These directories are all one (1) and two (2) alpha
>> characters in length. EXAMPLE:'+2' ';1' 'Y' '0'
>>
>> Upon examination of the data they all seem to have the station of origin as
>> KWSI, for example:
>>
>> ^A^M
>> 867 ^M
>> FDCNO KWSI 302000^M
>> ^^FDC !FDC 1/2850 N14 FI/T FLYING W  LUMBERTON, NJ.^M
>>      VOR-A AMDT 2 ...^M
>>        MSA FROM ROBBINSVILLE (RBV) VORTAC 025-270 1700, 270-025 2100.^M
>> ^^FDC !FDC 1/2852 DCA FI/T RONALD REAGAN WASHINGTON NATIONAL, ^M
>> WASHINGTON, DC. ^M
>>   VOR/DME OR GPS RWY 19, AMDT 9...^M
>>      S-19 MDA 940/HAT 926 ALL CATS. VIS CATS A/B RVR 6000,^M
>>      CAT C 2 3/4, CAT D 3.^M
>>      CIRCLING MDA 940/HAA 924 ALL CATS. VIS CATS A/B 1 1/4,^M
>>      CAT C 2 3/4, CAT D 3.^M
>>      MINIMUM ALTITUDE AT DCA 5.00 DME (FAF) 1600.^M
>>      DCA 3.0 DME STEPDOWN FIX NA.^M
>>      VDP DCA 2.2 DME NA.^M
>>      ALTERNATE MINIMUMS: CATS A/B 940-2, CAT C 940-2 3/4, ^M
>>      CAT D 940-3.^M
>> ^^^M
>> ^C^A^M
>> 916 ^M
>> FDCNO KWSI 302000^M
>> ^^FDC !FDC 1/2853 FDC CANCEL 1/2742 DCA^M
>> ^^^M
>> ^C
>>
>> This bulletin gets placed into a directory called ~ldm/O/20.
>>
>> pqcat -v reveals:
>>
>> Nov 30 21:16:07 pqcat: Starting Up (6337)
>> Nov 30 21:16:07 pqcat:      357 20011130182856.123  DDPLUS 242  FDCN00    O
>> 301828
>>
>> I get my data directly from a WSI HCSN feed and from the Canadian Met.
>> Center's Anikom Satellite service.
>>
>> Is anyone else getting these randomly created subdirectories in the
>> $LDMHOME directory?
>>
>> Stephen Sinnis
>> Pelmorex Inc.
>> address@hidden
>