Board index FlightGear Development AI Traffic

ErrorLog cxx:268 exception updating AI model  Topic is solved

Intelligent, computer controlled vehicles that drive/fly over the planet!

Re: ErrorLog cxx:268 exception updating AI model

Postby wkitty42 » Wed May 23, 2018 1:02 am

see? this is kinda like what i/we were doing with the performancedb stuff... when the pdb was fixed the traffic files were, too... in this case, go ahead and fix that one file and get it committed... then we don't have to worry about it any more... later on, when that traffic file is cleaned up and adjusted, then the fix won't matter but in the mean time, why not go ahead and fix it? in this case, it is only one line in the traffic file and possibly another in the pdb... holding back on these things and adding them to a TODO list doesn't make any sense when they can be done and over with in 2 seconds...
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 9148
Joined: Fri Feb 20, 2015 4:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 20.04

Re: ErrorLog cxx:268 exception updating AI model

Postby BecOzIcan » Wed May 23, 2018 4:52 am

wkitty42 wrote in Wed May 23, 2018 1:02 am:why not go ahead and fix it? in this case, it is only one line in the traffic file and possibly another in the pdb... holding back on these things and adding them to a TODO list doesn't make any sense when they can be done and over with in 2 seconds...


Hi wkitty,

It's not really clear why you didn't follow your own advice and updated <actype> in the S7 traffic file last month rather than reworking PDB but I'd rather not spend any more time on that pointless conversation.

1. The PDB side has already been fixed by the commit I pushed yesterday
2. <actype> related issues are non blocking errors, with no demonstrated impact on system performance, care factor NIL
3. What's on my TODO list is the rebuild of the S7 traffic, not some plastering of outdated data, because I consider the longer term and the big picture

Cheers
Ian
Current Projects: AI Traffic, Models & Liveries
User avatar
BecOzIcan
 
Posts: 1302
Joined: Tue Oct 04, 2011 11:43 pm
Location: Sydney, NSW, Australia
Version: 2020.4.0
OS: Win10

Re: ErrorLog cxx:268 exception updating AI model

Postby wkitty42 » Wed May 23, 2018 2:27 pm

BecOzIcan wrote in Wed May 23, 2018 4:52 am:It's not really clear why you didn't follow your own advice and updated <actype> in the S7 traffic file last month rather than reworking PDB

S7?? wotzat?

for your edification, i did both... i updated the pdb as well as the traffic files... i ALSO detailed what i did in each file in the pdb topic so that you and anyone else could do the same... apparently you chose only to include what gooneybird did rather than to also include my work or just take the updated pdb file from the repo i pointed you guys to... i NEVER said i was going to apply them to the official repo NOR did i ever say that i was in a position to do that... you guys are already clearing the field so rather than me starting my own clearing project of the same field, i prefer to tell you about the problem, the fix and maybe even provide you with the patch and let you do the merging since you're already set up for it... if i also started doing it to fix these things, we'd very quickly come to loggerheads with conflicts... i don't want to do that but if that's what you prefer, then maybe i can accommodate you... even this tiff between us right now is drek and i hate it...

BecOzIcan wrote in Wed May 23, 2018 4:52 am:but I'd rather not spend any more time on that pointless conversation.

whatever... i'm just very disappointed that you didn't use the file AND make the traffic file changes i specifically noted in the pdb topic while i was doing that work... if you had wanted me to make those traffic file changes and send them to you, you didn't say that... if you had wanted me to send you the updated pdb file, i could have done that, too... BUT instead i did the repo thing to keep track of the changes to the pdb AND to make it easier for YOU to get to so you would not have to duplicate the work and possibly miss some things... but whatever...

BecOzIcan wrote in Wed May 23, 2018 4:52 am:1. The PDB side has already been fixed by the commit I pushed yesterday

no it hasn't... not completely... it is only partially fixed and only up to the point that it matches with your traffic files and updating schedule, whatever that is... did you even look at the file i posted from the repo?? the majik 8-ball says "unlikely"...

BecOzIcan wrote in Wed May 23, 2018 4:52 am:2. <actype> related issues are non blocking errors, with no demonstrated impact on system performance, care factor NIL

i don't care if they are blocking or not... they are ERRORS and they need to be fixed... sooner rather than later... look, man, i'm not trying to ""take over"" if that's what you think... i don't want to step on your's and others' toes... i just want to contribute and have that contribution accepted and utilized as given... you seem to be the chief indian of the ad hoc "AI Updater Group"... one of my main goals is finding and cleaning up logged errors or at least figuring out why they are happening and letting someone else who can fix them do so... i'm not a C/C++ coder but i can still wade in and find the problems... i can certainly fix the data... old data, new data, it is all the same and the errors should not exist... if that means updating old data files, then so be it... it is not wasted work if the error is removed even if the data file is completely rewritten some weeks from now...

BecOzIcan wrote in Wed May 23, 2018 4:52 am:3. What's on my TODO list is the rebuild of the S7 traffic, not some plastering of outdated data, because I consider the longer term and the big picture

again with the S7... no clue what that is... i'm not looking for ""plastering of outdated data"... i'm looking to fix/remove the error reports now rather than umpteen months down the road... i, too, look at the long term and the big picture... if i didn't i would never have even started trying to give something back to the community... have a day, sir...


PS: it looks like all those old parking.xml files and stuff in the old fgdata/AI/Airports tree are going to be removed... quite possibly that whole Airports tree... apparently there's no code at all in FG any more that reads parking.xml files... the rwyuse.xml files? there's only six of them (EHAM, KLAX, KOAK, KSFO, LEPA, OMDB) and it looks like they are already included in the terrasync/Airports tree along with only 12 others... i don't know if they are the same in both trees or not, though... no real big loss as few as they are...
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 9148
Joined: Fri Feb 20, 2015 4:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 20.04

Re: ErrorLog cxx:268 exception updating AI model

Postby BecOzIcan » Wed May 23, 2018 4:28 pm

Hi wkitty.

Rest assured there is no tiff at play here but simply (In my view at least) some misaligned expectation and -as a result- a lot of frustration on both sides of the Pacific Ocean.

The AI data is mostly maintained by only 3 active contributors and I am sure you can appreciate how much has been achieved in the past three years. Part of the deal (and enjoyment) of contributing to an open source project is the freedom to pick your projects.

In that context a statement like
I don't care if they are blocking or not... they are ERRORS and they need to be fixed... sooner rather than later...

does sound like some aspiring Chief Indian is trying to force his problem onto me (no punt intended). I appreciate your focus is on identifying and resolving the issues behind the FG Log errors but you have to realise it is not mine (unless -again- they are blocking my own simulation or coming from problems in data I contributed) .

For this reason, rebuilding the traffic file for S7 Airlines (legally PJSC Siberia Airlines) ie the one behind the Airbus_310 actype problem (kindly identified by you yesterday) is only on my todo list. A lot of traffic files still need cleaning and we manage our priorities through the "what traffic file are you working on" thread in the AI section of the forum. Post there if you are working on any traffic update.

Apologies on the PDB hickup. I t was my understanding from the pms exchanged before my vacation that you were not actually contributing any data but simply aggregating the information collected by Gooneybird. For that reason (and as indicated in my last pm, 2 weeks ahead of proceeding) I updated the file and commited it myself as I normally do with G's data.

Feel free to pm me if anything remains unclear

Cheers
Ian
Current Projects: AI Traffic, Models & Liveries
User avatar
BecOzIcan
 
Posts: 1302
Joined: Tue Oct 04, 2011 11:43 pm
Location: Sydney, NSW, Australia
Version: 2020.4.0
OS: Win10

Re: ErrorLog cxx:268 exception updating AI model

Postby gooneybird » Thu May 24, 2018 1:23 pm

bugman wrote in Tue May 22, 2018 11:10 pm:How would you design the AI logic in this scenario?

Regards,
Edward


Not sure on that but it would nice if on any errors the AI traffic would taxi directly to the runway. Not perfect I know but it will be better than what we have now. Aircraft moving directly to runways are also more likely to be noticed by people at the airport more than an error report in the console and hopefully encourage them to report the issue.

On airports with no groundnetwork no aircraft will take off apart from a few at FG start up but any incoming traffic will land and taxi directly to the centre of the airport and wait there until it's next departure then taxi directly to the runway for take off.
My Github repository (mostly AI stuff) https://github.com/gooneybird47
User avatar
gooneybird
 
Posts: 3046
Joined: Sat May 31, 2008 2:57 pm

Previous

Return to AI Traffic

Who is online

Users browsing this forum: No registered users and 3 guests