Improved Routing Information

[ACW367]

I recently flew on a on sector (KBUF-KBWI) of a route that possibly has the most daily sectors and winding route.

Southwest Airlines SWA364
http://flightaware.com/live/flight/SWA364

KBUF-KBWI-KMDW-KLIT-KDAL-KMSY-KTPA
Buffalo-Baltimore-Chicago-Little Rock-Dallas-New Orleans-Tampa

I don't think the system can cope with that one!

[DaveReid]

We need the list in xml format, not in pdf.

That wouldn't really help you - the XML comes from a relational database that can handle multiple (i.e. more than two) legs per flight:

  <?xml version="1.0" encoding="utf-8" ?>
- <Flight>
  <FlightNumber>UAL840</FlightNumber>
  <Callsign>United 840</Callsign>
- <Leg>
- <Origin>
  <ICAOCode>YMML</ICAOCode>
  <IATACode>MEL</IATACode>
  <AirportName>MELBOURNE</AirportName>
  <Country>AUSTRALIA</Country>
  </Origin>
- <Destination>
  <ICAOCode>YSSY</ICAOCode>
  <IATACode>SYD</IATACode>
  <AirportName>SYDNEY (KINGSFORD SMITH)</AirportName>
  <Country>AUSTRALIA</Country>
  </Destination>
  <GCDistanceNM>381</GCDistanceNM>
  </Leg>
- <Leg>
- <Origin>
  <ICAOCode>YSSY</ICAOCode>
  <IATACode>SYD</IATACode>
  <AirportName>SYDNEY (KINGSFORD SMITH)</AirportName>
  <Country>AUSTRALIA</Country>
  </Origin>
- <Destination>
  <ICAOCode>KLAX</ICAOCode>
  <IATACode>LAX</IATACode>
  <AirportName>LOS ANGELES (INTERNATIONAL)</AirportName>
  <Country>UNITED STATES</Country>
  </Destination>
  <GCDistanceNM>6508</GCDistanceNM>
  </Leg>
- <Leg>
- <Origin>
  <ICAOCode>KLAX</ICAOCode>
  <IATACode>LAX</IATACode>
  <AirportName>LOS ANGELES (INTERNATIONAL)</AirportName>
  <Country>UNITED STATES</Country>
  </Origin>
- <Destination>
  <ICAOCode>KORD</ICAOCode>
  <IATACode>ORD</IATACode>
  <AirportName>CHICAGO (O'HARE INTERNATIONAL)</AirportName>
  <Country>UNITED STATES</Country>
  </Destination>
  <GCDistanceNM>1512</GCDistanceNM>
  </Leg>
  </Flight>

As the previous poster points out, the simplistic From/To/Via structure of the NavData flights table can't handle that - UAL840 never gets as far as Chicago in your database.

[DaveReid]

B772 G-VIIW operating BAW9262 on MyFlights this morning, route showing as London City to Gatwick.

Now that's a departure I'd like to have seen !!

[bratters]

B772 G-VIIW operating BAW9262 on MyFlights this morning, route showing as London City to Gatwick.

Now that's a departure I'd like to have seen !!

The route date in my database explorer for BAW9262 is 11 August 2010 so presumably it was a Flightstats jobbie rather than a result of the new Up&Down system.

[tarbat]

The route date in my database explorer for BAW9262 is 11 August 2010 so presumably it was a Flightstats jobbie rather than a result of the new Up&Down system.

I suspect this came from the new TO/LND system.

An engineering/test flight? - see http://www.kineticavionics.co.uk/phpBB3/viewtopic.php?f=5&t=12706#p97275

I would presume that Airnav ran this new TO/LND routine for a few days before announcing it and activating the feed into the routes database.

[bratters]

Bit of a red herring here - that Kinetic posting on June 3rd was DR himself.
Doesn't seem remotely connected with matters in hand.

[tarbat]

Okay, I thought it was relevant to mention that BAW9262 was thought to be a test flight back in June, so may still be a test flight.  And it's not on Flightstats, so may have come from somewhere other than Flightstats.

If it did come from the TO/LND system, then that would suggest some flaws in the TO/LND system.

Or it might be an old, out-of-date positioning FlightID from earlier in the year - see http://www.airnavsystems.com/forum/index.php?topic=4558.msg45828#msg45828

[DaveReid]

The route date in my database explorer for BAW9262 is 11 August 2010 so presumably it was a Flightstats jobbie rather than a result of the new Up&Down system.

Somebody correct me if I'm wrong, but AFAIK the query that RadarBox sends to the server to get a flight routing only returns a bit of pseudo-XML like this:

<ANLVStart>
BAW285,EGLL,,KSFO,
<ANLVEnd>

So there is no way of knowing how old that route data is, or whether it has come from FlightStats or is derived from the new algorithm.

All that the date string in Database Explorer tells you is when it was downloaded, not when the record on the server was created.

[bratters]

So can the download to Database Explorer pre-date when the record on the server was created?

I thought it was only recently that TO/LND system was introduced - like three days ago. But if that system provided the routing, how can the Explorer entry be dated June?

Ugh! I'm getting seriously CONFUSED here.  Only "Dev" has the definitive answers (?) and he, she or it is keeping decidedly stum.

[AirNav Support]

That is correct Dave.

The dates in your local database only show when you downloaded the data from the server.

[bratters]

So BAW9262 routing London City to Gatwick has nothing to with TO/LND then.

[tarbat]

I thought it was only recently that TO/LND system was introduced - like three days ago. But if that system provided the routing, how can the Explorer entry be dated June?

I thought it was 11th August?  Anyway, my guess is that this routing came from an old flight earlier this year, possibly during re-positioning flights.  I guess that once a route gets into the database, until it's replaced with something more recent, it stays there forever.

[bratters]

Correction on my part - my Explorer entry 11th august. Dave's Kinetic entry in June.

Neverthless, reading Airnav's posts I gain the impression that TO/LND is fresh out of the box and I'm sure you're right Tarbut.

[DaveReid]

I guess that once a route gets into the database, until it's replaced with something more recent, it stays there forever.

Yes, see my earlier post re yesterday's (and today's) Heathrow flights which are wrong on the server and are clearly not being validated by the algorithm.

That strikes me as a fairly fundamental shortcoming of the system - if the network detects an aircraft landing at Heathrow, e.g. this morning's UAL948, why persist in pushing out the route as KLAX-KDEN ?  Even FlightStats will tell you that the flight continues to LHR.

[bratters]

How complicated would it be to enable the database to be instantly over-written by the latest data from the server?