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

Re: 19990313: LDM modification for TTAA00 K... support-ldm <address@hidden>

> > >From: Gilbert Sebenste <address@hidden>
> > >Subject: To: Robb Kambic et al...
> > >Organization: Northern Illinois University
> > >Keywords: 199903132055.NAA24527 LDM
> >
> > Hi Robb,
> >
> > Is there anything I can modify in the LDM software to put the
> > TTAA00 K... header back on the first line of the AFOS PIL?

Gilbert & Robb:

If an AFOS product has a "WMO ID" like TTAA00 K...,
I believe it is specifically telling you that there is
no FOS/NOAAPORT product that matches this. It is excluded
from printing because no valid WMO Abbreviated Id begins
with "TTAA00". Perhaps someone in the weather service can
comment.

The salient difference in the source code between the releases in
question is in the test.

The old test was:

/* $Id: afos_message.c,v 1.33 1995/11/11 00:47:27 davis Exp $ */

        if(get_wmo_header(buf, mess->wmo_hdr) == NULL)
        {
                /* reject it */
        }
        else if(isascii(mess->wmo_hdr->TT[0]) &&
                 isupper(mess->wmo_hdr->TT[0]) &&
isupper(mess->wmo_hdr->TT[1]))
        {
                /* use it */
        }

The new test is more stringent:

/* $Id: afos_message.c,v 1.38 1998/10/16 19:28:35 steve Exp $ */

                if(get_wmo_header(buf, &wmo_hdr) != NULL
                                && wmo_hdr.ii != 0
                                && isupper(wmo_hdr.TT[0])
                                && isupper(wmo_hdr.TT[1])
                                && isupper(wmo_hdr.AA[0])
                                && isupper(wmo_hdr.AA[1]) /* number okay? */
                                && isupper(wmo_hdr.CCCC[0])
                        )
                {
                        /* use it */
                }

The specific clause that is dropping the "TTAA00" is "wmo_hdr.ii != 0".

Returning to your original question:

> Date: Mon, 8 Mar 1999 23:27:54 -0500 (EST)
> From: Gilbert Sebenste <address@hidden>
> To: "Glenn P. Davis" <address@hidden>
> Subject: Re: More on AFOS
>
> OK. That's cool. But, my problem comes when I want to match the names
> so that the AFOS products go into the same files as the products off of
> WMO headers produced by NOAAPORT. Like this:
>
> > AFOS        ^MKCWSTW        FILE    somesigmet_file
>
> Let's say I want the output to be:
>
> convective_sigmets.mar10
>
> How do I do this? With WMO headers, piece of cake. the date follows the
> header. But on an AFOS PIL, it doesn't. Glenn, you said that I thought
> pqing was...well, I can't read that message and compose at the same time
> here at home, but I was insuating that the current version of pqing was
> truncating the end of the header of an AFOS PIL and LDM 5.0.0 was not. I
> think the current version is. I know it is, because when my test site went
> from 5.0.0 to 5.0.6, poof! At precisely 10 Am when he switched it over, it
> disappeared. And all was working fine until he did that. Now, only certain
> AFOS PILs...with the WMO headers under the AFOS PIL...still work.
> The others do not.

Changing the source code to solve this is like cracking a walnut
with a sledge hammer. From your training at the ldm workshop, you know
that it supports date substitution based on the ingest time of the product.
You could achieve your goal by using the following pqact entry:

AFOS    ^MKCWSTW        FILE    convective_sigmets.%b%d

See the subheading "strftime()-based encoding"
on page 38 of the site manager's guide.

----

If there is a consensus among AFOS ldm users that they want to see the
bogus headers, then we can consider changing the source. I personally think
this is a bad idea, since naive pqact users converting from FOS will think that
they can just use the parenthesized string for selection, and they will be
mistaken.

-glenn