6 hours ago, DaveS said:

I wish in these instances the update wasn't forced so we could stay on an earlier version

One of the new data providers does not provide for a tail number in the traffic data and some of the models in the data are wromng as well, that's why in the generation of the archive a lot of , sorry for that. I will have to look into this how to fix that, maybe via the call sign.

For the time being I have placed the archive for week 19 back at my server, with version number 23.2.0. So although in your main wondow it will still read week 23, it actually is week 19.

The update at your end will either take place when you log in at a new day, or after you have deleted the PSXT\airports\archive folder, or if you have removed the DATE_CHECKED parameter from parameters.xml

Nico (there) goes my Friday :-[

 

 

39 minutes ago, kiek said:

. I will have to look into this how to fix that, maybe via the call sign.

Was gonna suggest exactly this 

1) first, check if registration matches 1:1, if not

2) check first three characters (strip the _, assume it’s the airline code). If airline code is not in dictionary, fallback to PVT

Thanks again for the interim rollback, and for your future efforts on aligning this

Edited June 14, 20241 yr by AeroMaster12

6 hours ago, kiek said:

One of the new data providers does not provide for a tail number in the traffic data and some of the models in the data are wrong as well, that's why in the generation of the archive contains a lot of PVT, sorry for that. I will have to look into this how to fix that, maybe via the call sign.

For the time being I have placed the archive for week 19 back at my server, with version number 23.2.0. So although in your main window it will still read week 23, it actually is week 19.

The update at your end will either take place when you log in at a new day, or after you have deleted the PSXT\airports\archive folder, or if you have removed the DATE_CHECKED parameter from parameters.xml

Nico, there goes my Friday :-[     

 

 

Interesting.

However, these new streams reflect live traffic correctly (I mean live active and live parked aircraft). Does this issue relate to historical data only?

And thanks for your efforts, Nico!

Interesting - 2.

Here is the live learning from ESSA yesterday. Whenever you have underscore in front of a flight number, you'll get PVT code for airline. I'm afraid that entire live stream might be affected as well.



<parkpos id="g12">
        <opt reg="" airl="PVT" type="A320" day="4" hh="15" cs="_DLH5XY" />
        <opt reg="" airl="DLH" type="A320" day="2" hh="26" cs="DLH1PL" />
    </parkpos>
   
    <parkpos id="g31">
        <opt reg="" airl="PVT" type="B738" day="4" hh="15" cs="_RYR5NR" />
        <opt reg="" airl="RYR" type="B738" day="1" hh="19" cs="RYR7692" />
    </parkpos>
   
    <parkpos id="g5">
        <opt reg="" airl="AUA" type="A320" day="2" hh="16" cs="AUA312C" />
        <opt reg="" airl="PVT" type="A320" day="4" hh="15" cs="_AUA312C" />


Here is KJFK on 06/06/2024.

<parkpos id="D2">
        <opt reg="N712JB" day="4" hh="26" cs="N712JB" />
        <opt reg="" airl="PVT" type="A320" day="4" hh="35" cs="_JBU2725" />
        <opt reg="" airl="PVT" type="B736" day="3" hh="42" cs="_JBU1118" />
    </parkpos>

Edited June 14, 20241 yr by G-YMML1

2 hours ago, G-YMML1 said:

Interesting - 2.

Here is the live learning from ESSA yesterday. Whenever you have underscore in front of a flight number, you'll get PVT code for airline. I'm afraid that entire live stream might be affected as well.



<parkpos id="g12">
        <opt reg="" airl="PVT" type="A320" day="4" hh="15" cs="_DLH5XY" />
        <opt reg="" airl="DLH" type="A320" day="2" hh="26" cs="DLH1PL" />
    </parkpos>
   
    <parkpos id="g31">
        <opt reg="" airl="PVT" type="B738" day="4" hh="15" cs="_RYR5NR" />
        <opt reg="" airl="RYR" type="B738" day="1" hh="19" cs="RYR7692" />
    </parkpos>
   
    <parkpos id="g5">
        <opt reg="" airl="AUA" type="A320" day="2" hh="16" cs="AUA312C" />
        <opt reg="" airl="PVT" type="A320" day="4" hh="15" cs="_AUA312C" />


Here is KJFK on 06/06/2024.

<parkpos id="D2">
        <opt reg="N712JB" day="4" hh="26" cs="N712JB" />
        <opt reg="" airl="PVT" type="A320" day="4" hh="35" cs="_JBU2725" />
        <opt reg="" airl="PVT" type="B736" day="3" hh="42" cs="_JBU1118" />
    </parkpos>

Yeah there are simply to many errrors, JetBlue with a B736?  It must be an A321 ...

Pff, I have enough of it, worked on this the whole Friday and tonight the European Soccer champion ship starts, so, to enter a reliable state, both for the program as the archive:

I have suppressed traffic data coming from the new source that lacks a registration code (tail number). Data from that source that does have a tail number is still present.

See the release notes of version 7.4.1

 

@G-YMML1You better delete your Learned file, and start again.

Edited June 14, 20241 yr by kiek

Well, that's a bummer 😞

Two questions though.

1. Is all the data from the new sources corrupted or just a part of it?

2, May we expect a fix for this situation, as without this fix all the benefits of adding these two sources would be literally wiped out and we're back to square one.

Thanks!  

 

9 minutes ago, G-YMML1 said:

1. Is all the data from the new sources corrupted or just a part of it?

2, May we expect a fix for this situation, as without this fix all the benefits of adding these two sources would be literally wiped out and we're back to square one.

@1 part of it.

@2  Do not expect a fix from my part, it has to come from the data provider.

 

3 minutes ago, kiek said:

@1 part of it.

@2  Do not expect a fix from my part, it has to come from the data provider.

 

Got it.

For @2, it apparently means "never". Very sad.

45 minutes ago, kiek said:

@1 part of it.

@2  Do not expect a fix from my part, it has to come from the data provider.

 

Hi Nico, thank you for your work to try to solve the problem. I have little hope that the data provider will fix this, so it looks like the RT update is not going to provide the service we expected. With this news I do not expect that I will renew my license when the time comes, such a shame.

Nico mentioned that he would try to implement "callsign" approach for proper recognition. I don't know what it means, perhaps Nico may shed some light onto it.

2 hours ago, G-YMML1 said:

Nico mentioned that he would try to implement "callsign" approach for proper recognition. I don't know what it means, perhaps Nico may shed some light onto it.

The callsign approach is that if the callsign starts with 3 letters followed by a digit you check if these 3 letters is a valid ICAO airline code. 

However, that does not help here, because the problem lies in the model info (F100 and so on) that is not always correct with that new data provider. Normally a tail number is given in the traffic data and with the tail number I look up in my dbase what the airline and model and operator are. But there is no tail number...

 

26 minutes ago, kiek said:

The callsign approach is that if the callsign starts with 3 letters followed by a digit you check if these 3 letters is a valid ICAO airline code. 

However, that does not help here, because the problem lies in the model info (F100 and so on) that is not always correct with that new data provider. Normally a tail number is given in the traffic data and with the tail number I look up in my dbase what the airline and model and operator are. But there is no tail number...

 

Is it something that might be fixed in a casual way by asking a provider to adjust the data parameters in the stream OR it's just a fundamental issue with how data is structured and provider most likely would be reluctant to honor such request?

 

A message to MOD - I'm sorry Ray. I really tried not to post twice.

8 hours ago, G-YMML1 said:

Is it something that might be fixed in a casual way by asking a provider to adjust the data parameters in the stream OR it's just a fundamental issue with how data is structured and provider most likely would be reluctant to honor such request?

 

I don't know, that's outside of my scope. Ask Balthasar.

7 hours ago, kiek said:

I don't know, that's outside of my scope. Ask Balthasar.

Hmmm...surely i can ask, but pardon my naivety, I though that you are both in a constant contact regarding technical aspects of PSXT/RT integration and it's in your mutual interest to keep the software running smoothly. In the meantime, the announcement that Balthasar made two weeks ago regarding "improvement' is nothing but hoax. 

 

Edited June 15, 20241 yr by G-YMML1