Upcoming Games

No games to display

Add a game

Upcoming Events

01/06/2019 12:00:00
23/03/2019 12:00:00

Odd timetable validation

You are here: Home > Forum > Simulations > Timetables > Exeter > Odd timetable validation

Page 1 of 1

Odd timetable validation 14/10/2018 at 02:45 #112614
AlexRail575
Avatar
25 posts
Trying to create a timetable for Exeter based on the current WTT, there's an odd situation where, for a train headed from Newton Abbot to Paignton - 2T04 (I tried with other trains terminating at Paignton, but only this one seems to have it, trains running from Exeter or all the way from Exmouth do not have this problem), once I validate the timetable, the order of the last two stations (Torquay and Paignton) is reversed...

Has this happened to anybody else?
[attachment=9859]Timetable editor bug 2.png[/attachment]

[attachment=9860]Timetable editor bug 1.png[/attachment]

Post has attachments. Log in to view them.
Log in to reply
Odd timetable validation 14/10/2018 at 06:15 #112615
pedroathome
Avatar
390 posts
Online
AlexRail575 in post 112614 said:
Trying to create a timetable for Exeter based on the current WTT, there's an odd situation where, for a train headed from Newton Abbot to Paignton - 2T04 (I tried with other trains terminating at Paignton, but only this one seems to have it, trains running from Exeter or all the way from Exmouth do not have this problem), once I validate the timetable, the order of the last two stations (Torquay and Paignton) is reversed...

Has this happened to anybody else?
[attachment=9859]Timetable editor bug 2.png[/attachment]

[attachment=9860]Timetable editor bug 1.png[/attachment]

Your Torquay departure time is 1720 not 0720 like I suspct you mean. Thiss is why it woulc have come out of order

James

Log in to reply
Odd timetable validation 14/10/2018 at 23:15 #112633
AlexRail575
Avatar
25 posts
Bad typo on my part - however, I ask: even with the departure times being later than the arrival, would it not be more convenient (easier to fix) if the validator simply pointed this out without changing the order of the involved stations?
Log in to reply
Odd timetable validation 14/10/2018 at 23:32 #112634
postal
Avatar
3482 posts
AlexRail575 in post 112633 said:
easier to fix
For the user or the coder?

"I drink to make other people seem more interesting." - George Jean Nathan, (14/02/1882 - 08/04/1958), American author, editor, and drama critic
Log in to reply
Odd timetable validation 15/10/2018 at 01:56 #112637
AlexRail575
Avatar
25 posts
For the user, since, usually, messing station order up (in my example, selecting "Paignton" instead of "Torquay" ) isn't as easy as accidentally typing (again in my example) "17" instead of "07". Not that it seems like a major change...
Last edited: 15/10/2018 at 01:56 by AlexRail575
Reason: None given

Log in to reply
Odd timetable validation 15/10/2018 at 06:52 #112639
Peter Bennet
Avatar
4422 posts
Online
AlexRail575 in post 112633 said:
Bad typo on my part - however, I ask: even with the departure times being later than the arrival, would it not be more convenient (easier to fix) if the validator simply pointed this out without changing the order of the involved stations?
Good point, every location has a number (in the down direction), which is what keeps them in order (unless times dictate otherwise) so it might be possible to validate the number order between key locations (which are kept in rigid order).

[Actually, I have a vague thought that it already did something along those lines in the TT analyser but I don't have time to test at the moment]

Peter

There is no such thing as public opinion. There is only published opinion. Winston Churchill
Last edited: 15/10/2018 at 06:57 by Peter Bennet
Reason: Additional thought.

Log in to reply
Odd timetable validation 15/10/2018 at 12:52 #112643
clive
Avatar
1764 posts
Online
Okay, let me attempt to explain. A lot of this is historical - the code was designed a long time ago and much has changed since.

There are two kinds of locations in simulations: "key" and "non-key". Very simply, a non-key location is one that trains can stop at or pass but nothing else, while key locations can have actions and other things.

In the simulation data are lists of "paths" which are pairs of key locations that can come after one another; these may also carry information about path, line, and platform numbers. (They also do other things in ARS sims.) So when validation says that there is no path between two locations, that's because that pair aren't in the list. To do this - and I don't know why it was written this way - the core code strips out all the non-key locations before doing the validation. It then puts them back in afterwards and - and this is what's causing the issue - puts them in in time order. I have no idea why it was done like that, even though I've got a vague feeling that I wrote or rewrote some of the code.

Yes, this could be fixed. No, it's not going to be at all simple. I've raised Mantis 21783 so it doesn't get lost.

Peter mentions location numbers. These aren't required to be in the down direction and in fact they aren't checked at all (or weren't last time I looked at the code). They're only there for ordering the spreadsheets that one user facility produces. Before they could be used we'd have to update *every* sim to use them.

Log in to reply
Odd timetable validation 15/10/2018 at 19:28 #112654
Peter Bennet
Avatar
4422 posts
Online
Just checked and the analyser reports if locations are out of order.

Peter

There is no such thing as public opinion. There is only published opinion. Winston Churchill
Log in to reply
Odd timetable validation 15/10/2018 at 19:37 #112655
9pN1SEAp
Avatar
109 posts
Online
clive in post 112643 said:

Yes, this could be fixed. No, it's not going to be at all simple. I've raised Mantis 21783 so it doesn't get lost.
Not just a matter of adding a second pointer in a linked list to next Key Location?

Jamie S (JAMS)
Log in to reply
Odd timetable validation 15/10/2018 at 19:58 #112657
headshot119
Avatar
2867 posts
Online
9pN1SEAp in post 112655 said:
clive in post 112643 said:

Yes, this could be fixed. No, it's not going to be at all simple. I've raised Mantis 21783 so it doesn't get lost.
Not just a matter of adding a second pointer in a linked list to next Key Location?
No because the next key location may not be a key location for all trains.

"James Street, Moorfields, Liverpool Lime Street and Liverpool Central"
Log in to reply