Re: pqinsert

• To: Christian Page <page@xxxxxxxxxxxxxxxxx>
• Subject: Re: pqinsert
• From: Robb Kambic <rkambic@xxxxxxxxxxxxxxxx>
• Date: Mon, 6 Aug 2001 15:03:30 -0600 (MDT)

On Thu, 2 Aug 2001, Christian Page wrote:

Hi,

Yes I am willing to test your new code of pqsurf. Just tell me when it's ready
to test.

Christian,

I attached surf_split.c , put it in the src/pqsurf directory and then type

% make

You might want to save the original pqsurf/surf_split.c and the
bin/pqsurf  before making the changes.   After creating pqsurf, put it
into the bin directory.

The fix is to check for the Pils in the bulletins before processing
starts. ie

       if( strstr( header, "METAR\r" ) || strstr( header, "SPECI\r" ) ||
               strstr( header, "MTR" ) ||
               strstr( header, "METAR \r" ) || strstr( header, "SPECI \r"
)) {
               usePil = 1;
       } else {
               usePil = 0;
       }


Let me know how this works.

Thanks,
Robb....

ALso, I got some errors in my ldmd.log, but only once, about pqsurf:

ldmd.log.3:Aug 01 00:03:31 3Q:io pqsurf[1912802]: HDR + PIL: 981
ldmd.log.3:Aug 01 00:33:30 3Q:io pqsurf[1912802]: HDR + PIL: 981

The 981 seems related to my local product I am inserting (see email downhere).
But no errors since.

Greetings,

Christian Page
UQAM

On Thu, 2 Aug 2001, Robb Kambic wrote:

> On Tue, 31 Jul 2001, Christian Page wrote:
>
> >
> > Thanks for the reply, it works perfectly if rename the file "SAFR31 LFPW 
311230"
> > Another question, related to pqsurf, I noticed that products are duplicated 
,
> > e.g. by example :
> >
> > SAFR32 LFPW 311900^M^M
> > LFBT 311900Z 10003KT CAVOK 27/20 Q1022 NOSIG=^M^M
> > (...)
> >
> > and
> >
> > SAFR31 KWBC 311900^M^M
> > METAR^M^M
> > LFBT 311900Z 10003KT CAVOK 27/20 Q1022 NOSIG=^M^M
> > (...)
> >
> > In the second case, I have a METAR, and in the first case, not.
> > For pqsurf.conf, I have:
> > WMO     ^(metar|speci) (....) ([0-3][0-9])([0-2][0-9])
> >         DBFILE  data/surface/gdbm/(\3:yyyy)(\3:mm)\3\4.metar_gdb \2
> >
> > So, in the first case, the entry is then not processed by pqsurf?
> > It seems the case, because in my try to insert prods, I don't have a METAR 
line,
> > and the product is not processed by pqsurf. Should I have a non-restrictive
>
> Christian,
>
> I've been fighting with the pqsurf code for awhile now. I did not write it
> and it's not very flexible code to change. The problem is that some METAR
> bulletins have PILs and some don't have PIL.  The code has problems
> handling this situation. As a kludge :( I designater all Canadian products
> to not have PILs and all the others to be processed as having a PIL. Off
> course this is not correct....  I'll try to correct the problem.  WOuld
> you be willing to test the new solution?  I might have time to work on it
> soon.
>
> I believe the rest of your configurations are correct.
>
> Robb...
>
>
>
>
>
> > pqsurf.conf, like:
> > WMO     (*) (....) ([0-3][0-9])([0-2][0-9])
> >         DBFILE  data/surface/gdbm/(\3:yyyy)(\3:mm)\3\4.metar_gdb \2
> >
> > My ldmd.conf is:
> > (...)
> > exec    "pqsurf -p ^(S[AP]) -Q /io/ldm/data/pqsurf.pq -d /io/ldm 
/io/ldm/etc/pqsurf.conf"
> > (...)
> >
> >
> > Christian Page
> > UQAM
> >
> > On Tue, 31 Jul 2001, Robb Kambic wrote:
> >
> > > On Tue, 31 Jul 2001, Christian Page wrote:
> > >
> > > >
> > > > Hi,
> > > >
> > > > I want to insert a few supplemental METARs into the local product queue 
of LDM
> > > > and have them processed by pqact, pqsurf. I tried pqinsert:
> > > >
> > > > pqinsert -vx -q /io/ldm/data/ldm.pq -f IDS /io/ldm/data/nws/LF.TXT
> > >
> > > > The products are in the queue, but they are not processed. If I try 
pqinsert
> > > > again:
> > >
> > > Christian,
> > >
> > > Your process is correct but with one small detail missing. The product
> > > identifier in the above example is "/io/ldm/data/nws/LF.TXT" so your
> > > pqact entries must be expecting that identifier.  Do you have an
> > > pqact entry for that identifier under feedtype IDS? Another problem is
> > > that pqsurf is looking for the standard WMO identifier too. ie "SAFR31
> > > LFPW 311230" Actually pqinsert should have an id flag to set the
> > > identifier. One solution would be to create a filename with the desired
> > > WMO indentifier and then use pqinsert.  The other problem about the
> > > product already being in the que is that the LDM sees it as a duplicate
> > > product and so it ignores it. One can wait unitl the product is removed
> > > from the que, about an hour on a LDM receiving data or create a product
> > > that is slighty different.  The product is considered the same from the
> > > WMO heading down to the end of the data. One quick way to change the
> > > product would be to duplicate a internal report.
> > >
> > > Robb...
> > >
> > > >         Mapping 267993088
> > > >         PQUEUE_DUP
> > > > pqinsert: Product already in queue: 91d8a5146b2ede24afaae708b94975f9    
  398
> > > > 20010731124810.406     IDS
> > > >
> > > > of course the product has already been inserted.
> > > >
> > > > The file itself is (where ^M ^C and ^A are control characters):
> > > > ^C^A^M^M
> > > > 981 ^M^M
> > > > SAFR31 LFPW 311230^M^M
> > > > LFJL 311230Z AUTO 26010KT 9999 SCT058 SCT070 30/15 Q1022=^M^M
> > > > LFBG 311230Z 36006KT 320V040 CAVOK 31/16 Q1024=^M^M
> > > > ^M^M
> > > >
> > > >
> > > > I put a dummy number 981, and a dummy product header SAFR31 LFPW.
> > > > The real SAFR31 LFPW product has already been processed by LDM.
> > > >
> > > > What is missing?
> > > >
> > > > Thanks!
> > > >
> > > > Christian Page
> > > > UQAM
> > > >

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