NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.

To learn about what's going on, see About the Archive Site.

RE: bug in 'UNITS' field in metar2nc ... clue II

NOTE: The decoders mailing list is no longer active. The list archives are made available for historical reasons.

Claude,

i made a typo in the metar.cdl file, that's the problem. i attached the
new metar.cdl file.

i also attached a perl script makeWMO.pl to convert the format to the wmo
network format.

% makeWMO.pl < wmo.text.format > wmo.network.format

the wmo.text.format is from the site you mentioned, ie

2005/12/14 04:45
KARM 140445Z AUTO 15007KT 10SM OVC050 18/15 A2993 RMK AO2

consisting of 3 lines per report, last line is blank, the converted file
looks like:

^A
000
SAMX00 KDEN 140445
METAR
KARM 140445Z AUTO 15007KT 10SM OVC050 18/15 A2993 RMK AO2
^C^A

so now you can use wmo.network.format file as direct input to the decoder,

%  metar2nc metar.cdl data < wmo.network.format

if you are missing stations, add them to the sfmetar_sa.tbl file.

hopefully this should solve your problems.

robb...




On Wed, 14 Dec 2005, DEROGNAT, Claude wrote:

> Thank you,
>
> Do you want me to do test or something ???
>
> Last clue for today ... the decoder tools uses netcdf.3.5 from Debian
> distrubution
> and I try to read the netcdf file on windows.
>
> Claude
>
>
> -----Message d'origine-----
> De : Robb Kambic [mailto:rkambic@xxxxxxxxxxxxxxxx]
> Envoyé : mercredi 14 décembre 2005 17:54
> À : DEROGNAT, Claude
> Objet : RE: bug in 'UNITS' field in metar2nc ... clue II
>
>
>
> i'm looking into it now.  i made a perl script to convert to wmo network
> format but it's not working yet.
>
> robb...
>
>
> On Wed, 14 Dec 2005, DEROGNAT, Claude wrote:
>
> > another clue,
> >
> > -1- I obtain the same result with the testdata.wmo
> > -2- over the whole string (character) fields only the first and second
> > one (rep_type and stn_name) are different to nothing in the netcdf file
.
> >
> > Claude.
> >
> > -----Message d'origine-----
> > De : DEROGNAT, Claude
> > Envoyé : mercredi 14 décembre 2005 14:55
> > À : 'Robb Kambic'
> > Objet : RE: bug in 'UNITS' field in metar2nc ... clue
> >
> >
> > Dear Rob,
> >
> > If I put the verbose option (-v), in the logfile,
> > the 'UNITS' values are correct but there are nothing
> > in the netcfd output file.
> >
> > Claude
> >
> > -----Message d'origine-----
> > De : DEROGNAT, Claude
> > Envoyé : mercredi 14 décembre 2005 12:36
> > À : 'Robb Kambic'
> > Objet : bug in 'UNITS' field in metar2nc
> >
> >
> > Dear Rob,
> >
> > I finally find the way to use metar2nc with the wmo raw format.
> > Everything is working properly but the 'UNITS' field (wind speed unit)
> > seem to be empty.
> >
> > -1- The number of character string (given by nf90_inquire_variable) is
> > strangely
> > equal to 5. I thought that the wind speed unit is KMH, KT, MPS, or MPH
> > which are less or equal to 3 characters.
> > -2- the string (given by nf90_get_var) is always empty.
> >
> > thank you,
> >
> > Claude
> >
> > -----Message d'origine-----
> > De : Robb Kambic [mailto:rkambic@xxxxxxxxxxxxxxxxxxxxxxxx]
> > Envoyé : jeudi 1 décembre 2005 21:12
> > À : DEROGNAT, Claude
> > Cc : decoders
> > Objet : Re: 20051130: Please Help about metar2nc
> >
> >
> > Ce message utilise un jeu de caractères qui n'est pas pris en charge
par
> le
> > service Messagerie Internet. Pour visualiser le contenu du message
> original,
> > ouvrez le message joint. Si le texte n'est pas affiché correctement,
> > enregistrez la pièce jointe sur le disque, et ouvrez-le avec une
> application
> > capable d'afficher le jeu de caractères original.
> > --
>
  • 2005 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the decoders archives: