[INN] #4: Add encoding to checkgroups processing

[Julien ÉLIE]

Hi Andrew ,

>> ukr.comp.hardware Проблеми апаратно╖ частини обчисювально╖ техн╕ки
>
>  This one is in koi8-u, not -r (fir Ukranian language).

Oh, thanks!
It now looks better:

ukr.comp.hardware Проблеми апаратної частини обчисювальної техніки

http://usenet.trigofacile.com/hierarchies/index.py?see=UKR


>  Thank you for your patience - we have an awful situation with national
>  encoding, e.g. M$ Internet Explorer has 4 charsets for us :-))

Thanks to you too.  Your information about charsets is very valuable!




Russ, I have just noticed a badly encoded description in ftp.isc.org:
    ee.nali   Huumoris�radele.

According to ftp://ftp.isc.org/pub/usenet/control/ee/ee.nali.gz, it is:
    Huumorisõpradele.

It is more consistent with the other descriptions available for ee.* (with
the use of õ).  Could you please fix it?

The only remaining problem is the one for tw.bbs.org.cycluga (which I had
already told you it was not properly encoded in Big5).  I will try to
ask in a Chinese forum to see if they know what it should be.  This
page http://cbs.ntu.edu.tw/threadread.php/board=bbsconfig&nums=399 has also
the problem.  And the official message
ftp://ftp.isc.org/pub/usenet/control/tw/tw.bbs.org.cycluga.gz is also bad
(and shorter!).

-- 
Julien ÉLIE

« -- Redde Caesari quae sunt Caesaris.
  -- Eh ?
  -- Y en a rendre à César ce qui y en a appartenir à César. » (Saugrenus)