phpweather-devel

SV: Strange METAR

[Funder, J. O. <jh...@NA...>]

Hi Martin,

You are right 0200KT is wrong. Looking at the current weather in England my
guess is that the last number of the velosity is missing should most likely
read 02007KT og something like that.

The reason the decoding is fooled, is (as I have mentioned before) because
each group is decoded as a group by it self and not in the context of the
message. Normally when decoding metars we make sure that the sequence of the
groups are correct and that eg. there must be a cloud group and if a
weathergroup is present it must be placed correctly in the metar message.

rgds
Johnny


-----Oprindelig meddelelse-----
Fra: Martin Geisler [mailto:gim...@gi...]
Sendt: 26. marts 2002 10:45
Til: PHP Weather
Emne: Strange METAR


Hi again,

I just saw this METAR:

  EGBE 260820Z 0200KT 340V060 CAVOK 06/02 Q1033

There's an error in the windgroup (0200KT), isn't there? Johnny?

It fools the get_winddir_image() method in pw_images.php because the
'340V060' makes the 'wind' entry non-empty, but still, there's no
'deg' entry because '0200KT' is malformed.

The fix is trivial in pw_images.php, but I just wanted to make sure,
that it's the METAR that's wrong, and not the parsing.

-- 
Martin Geisler                       My GnuPG Key: 0xF7F6B57B

See my homepage at http://www.gimpster.com/ for:
PHP Weather => Shows the current weather on your webpage.
PHP Shell   => A telnet-connection (almost :-) in a PHP page.

Re: SV: Strange METAR

[Martin G. <gim...@gi...>]

"Funder, Johnny                  OTP" <jh...@NA...> writes:

Hi Johnny
=20
> You are right 0200KT is wrong. Looking at the current weather in
> England my guess is that the last number of the velosity is missing
> should most likely read 02007KT og something like that.

OK, but that would be rather difficult for the script to figure that
out by itself :-)
=20
> The reason the decoding is fooled, is (as I have mentioned before)
> because each group is decoded as a group by it self and not in the
> context of the message.

I know - it should probably be reworked a bit.

> Normally when decoding metars we make sure that the sequence of the
> groups are correct and that eg. there must be a cloud group and if a
> weathergroup is present it must be placed correctly in the metar
> message.

Hmm, we should probably have such logic in PHP Weather too. Let's look
at that when we have a version 2 ready...

=2D-=20
Martin Geisler                       My GnuPG Key: 0xF7F6B57B

See my homepage at http://www.gimpster.com/ for:
PHP Weather =3D> Shows the current weather on your webpage.
PHP Shell   =3D> A telnet-connection (almost :-) in a PHP page.