[Xastir] ?APRSD queries reporting all rx trafic including adsb, ais...

[Tom Russo]

On Fri, Jun 22, 2018 at 07:56:23AM -0700, we recorded a bogon-computron collision of the <curt.we7u at gmail.com> flavor, containing:
> On Fri, 22 Jun 2018, F1MHV wrote:
> 
> > ?Bonjour? Gilles,
> >
> > Don?t stop-it but may be there is way to restrict the adsb forward from YAAC to xastir, or? I thought you had the adsb-rx connected directly to Xastir as per http://xastir.org/index.php/HowTo:Display_Aircraft_and_Ships
> 
> Agreed, don't stop it!
> 
> I wrote the ADS-B and AIS scripts which convert the other protocols to APRS packets and inject them on Xastir's UDP port. It may be a bit tricky figuring out how to mark them as "special" so the APRS query ignores them. It probably should be looked at to see whether it's possible.

This is a little puzzling, because the handling of APRSD *already* checks
if the object/station has been received over a TNC, is not third party,
and has never passed through a digi.  If all of these are true, 
th record has the ST_DIRECT bit set, which gets set in a block of
code beginning around 13290 of db.c.   APRSD is only supposed to respond
with stations that have ST_DIRECT set.

What's puzzling me is how any packet received through the UDP interface could
be getting marked as direct.  It sounds like something in the receive chain
is getting confused, or being lied to.  I'm trying to dig through the chain
of custody of packets to see where it could possibly be getting "DATA_VIA_TNC"
set, but I can't find it quickly.  But that must be what is happening 
somewhere.  It's the only way that APRSD could possibly be thinking it's 
appropriate to send this data on.

Or perhaps this is some oddity of how Gilles is using YACC to feed data to 
Xastir.  He has already said he's not using the adsb/ais scripts to do it.

-- 
Tom Russo    KM5VY
Tijeras, NM  

 echo "prpv_a'rfg_cnf_har_cvcr" | sed -e 's/_/ /g' | tr [a-m][n-z] [n-z][a-m]