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

Re: missing SPECI reports and the db file entries

hiya,

if there was a decoder that stored the individual reports in a gdbm
database then no need for pqsurf?  How about a dir on a daily basis that
has one file/stn with all the reports for the day?  it would probably be
faster  the gdbm.

robb...

On Mon, 20 Sep 2004, Neil R. Smith wrote:

> Our only requirement for pqsurf is the DBFILE action to generate output
> for Peter Neilly's weather program. Our ldm was set up that way when I
> took this job, and I've just left it that way. I see that DBFILE is an
> available pqact.conf action, so I don't know why I couldn't move it into
> a pqact.conf file.
>
> Often times I've wondered about pqsurf and what it's advantage is but
> never found a discussion of it in online ldm documentation. All I have
> is the description I have in printout of the LDM Workshop I attended in
> '98:
>
> pqsurf: "Bursts surface products into products containing individual
> observations".
>
> My margin note says "splits 'bulletins' into reports".
>
> The only use I find in creating a DBFILE output is that using it as
> input to the weather program generates output with only the METAR
> station reports that have a Remarks section. So we don't see both
> reports for that station at that time: the one with Remarks and the one
> without. Since I can find both reports in the output db file, it must be
> a filtering done in Peter's programming for db file processing, because
> if I do a weather program 'FLATMETAR' query from standard metar decoded
> files, the weather program will output both reports, which is something
> we don't want. The speed of db index'd lookups is really not needed
> anymore with the power of 2.x GHz Xeon processors these days, at least
> with these file sizes.
>
>
> On Mon, 2004-09-20 at 16:39, Robb Kambic wrote:
> > hiya,
> >
> > i did a quick grep of the last 3 days for reports that have ".... SPECI"
> > as the first part. the only station that came up was:
> >
> > 2004091718_sao.wmo:SBBE SPECI  ...
> >
> > is this the only station or are there other?  this is not the standard
> > format of according to the WMO 306  on codes.  unidata has been trying to
> > deprecate pqsurf processing because it's a fragile piece of code and it
> > really shouldn't be part of the ldm because it's a decoder. With that
> > being said, is there some other filter to replace the pqsurf code? thus
> > eliminating pqsurf.conf and the pqsurf queue. i would like to open a
> > discussion to find out the available options and the main reasons for
> > using pqsurf.
> >
> > thanks,
> > robb...
> >
> >
> > On Mon, 20 Sep 2004, Kevin R. Tyle wrote:
> >
> > > I have noticed this too; it appears to be a bug in the way METARs
> > > are displayed when read from a METAR file that is in the gdbm database
> > > format.  I have never investigated it further, but have typically
> > > fallen back to using "flatmetar" which reads from METARs written
> > > in ASCII format.  You sacrifice a bit of speed, but make up for
> > > it in the completeness of the data.
> > >
> > > ______________________________________________________________________
> > > Kevin Tyle, Systems Administrator               **********************
> > > Dept. of Earth & Atmospheric Sciences           address@hidden
> > > University at Albany, ES-235                    518-442-4578 (voice)
> > > 1400 Washington Avenue                          518-442-5825 (fax)
> > > Albany, NY 12222                                **********************
> > > ______________________________________________________________________
> > >
> > > On Mon, 20 Sep 2004, Neil R. Smith wrote:
> > >
> > > > All,
> > > > It was pointed out to me recently that our weather program(v4.10) was
> > > > not printing out all SPECI special reports that are available in the
> > > > bulletins.
> > > >
> > > > While looking at both a text and dbfile output from pqsurf, where both
> > > > pqsurf.conf entries have the same header request, and comparing with
> > > > output from weather queries, it seems that reports that are preceded
> > > > with a line that looks like xxxxSPECI, where xxxx is the 4-character
> > > > station id, are not output by METAR, METARDECODE, or METARFULLDECODE,
> > > > whereas those preceded with just a simple line that has only the
> > > > characters SPECI are output by these weather commands.
> > > >
> > > > Is this a known issue with weather, or is it a problem with the DBFILE
> > > > action in the LDM, or perhaps a bad header request entry in pqsurf.conf?
> > > >
> > > > Is there a known solution?
> > > >
> > > > (I can't find a way to search the weatherbud mailing list at unidata, so
> > > > I don't know if this has been treated before.)
> > > >
> > > > Our pqsurf ldmd.conf entry looks like
> > > >
> > > > exec    "pqsurf -p ^S(A....|P....|XUS8.|XUS91) -Q /que/pqsurf.pq
> > > > /unidata/ldm/etc/pqsurf.conf"
> > > >
> > > > Our pqsurf.conf looks like
> > > >
> > > > WMO     ^metar (....) ([0-3][0-9])([0-2][0-9])
> > > >         DBFILE  /data/ddplus/db_metar/(\2:yy)(\2:mm)\2\3.METAR_DB \1
> > > >
> > > > WMO     ^speci (....) ([0-3][0-9])([0-2][0-9])
> > > >         DBFILE  /data/ddplus/db_metar/(\2:yy)(\2:mm)\2\3.METAR_DB \1
> > > >
> > > > Thanks,
> > > > -Neil
> > > > --
> > > > Neil R. Smith, Comp. Sys. Mngr.         address@hidden
> > > > Dept. Atmospheric Sci., Texas A&M Univ. 979/845-6272 FAX:979/862-4466
> > > >
> > > >
> > >
> >
> > ===============================================================================
> > Robb Kambic                            Unidata Program Center
> > Software Engineer III                          Univ. Corp for Atmospheric 
> > Research
> > address@hidden                 WWW: http://www.unidata.ucar.edu/
> > ===============================================================================
> >
> --
> Neil R. Smith, Comp. Sys. Mngr.               address@hidden
> Dept. Atmospheric Sci., Texas A&M Univ.       979/845-6272 FAX:979/862-4466
>

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