[Tlf-devel] Re: TLF country resolution

[Rein Couperus]

Hi Martin, nice to work you from TS7N...

you are right, tlf does not know about prefixes at the moment, it just looks 
at the length, assuming the shorter part before or after the slash is the 
prefix. Of course this fails when both parts are the same length. I can even 
think about P3A/CE0Y coming on the band and tlf thinks it is Cyprus.

I get your point about calls and prefixes. I had not looked at it that way, 
and the concept is interesting. One problem is that cty.dat does not make the 
difference either. And I would hate to use a country file I have to maintain 
myself, as tlf takes enough of my time already :-)
Anyway, the cty.dat parser is going to be completely rewritten this winter, 
and I will take your ideas into account. Maybe I must hard code these 
exceptional prefixes into tlf. 

Now we hope that WP2E/VP2E refrains from contesting in the mean time...

73, Rein PA0R

BTW, we made 10k qso's in the M2 class, which means LOTS of log checking :-)


On Wednesday 03 December 2003 19:28, Martin Kratoska wrote:
> It sounds rather unlikely that a call like WP2E/VP2E appears but,
> surprisingly, such call is incorrectly resolved. Of course, both
> WP2E/VP2E and VP2E/WP2E are Anguilla and not Brit. Virgin Isl.
>
> More interesting seems that TLF does not know what is a 'call' (which
> can be used as 'slashed') and what is a prefix. Actually, VP2 is NOT a
> valid prefix but VP2E, VP2M and VP2V are. The same applies to VK9, CE0
> etc. Thus, I introduced so called 'extended prefix' into my country
> tables (used in YPLOG) which can be accurate. Remember that both 'CEPT
> style' calls like J3/G4ABC and 'classic' like W4ABC/J3 are possible.
> We need probably to search for both strings '/extended_prefix' and
> 'extended_prefix/'.
>
>
> 73,
>  Martin
>  address@hidden

-- 
address@hidden or address@hidden
http://www.couperus.com
http://savannah.nongnu.org/projects/tlf