Move to a rail that is not defined by the PlayerPath

[ginga81]

Description

I wrote each rail to come together at 1375m, and switched rails with Playerpath.
However, when I moved to it, I moved to a rail that I have no recollection of defining, and additionally, when I returned to Rail 0, I moved to a location that I had not defined, so I don't understand.
I loaded the three attached files into RouteViewer, but even when searching for the same distance, I could not find the corresponding rail.
I tried to debug it, but determined this to be difficult, so I will submit this as an issue.

using route files
routefiles.zip

fail-playerpath.mp4

Reproduction

I apologize for not being able to submit it as a sample as this is production data, under constructing.

Route

I apologize for not being able to submit it as a sample as this is production data, under constructing.

Train

any

Logs

RouteViewer and OpenBVE not report any bugs.

Related information

My PC's spec:
OS:Ubuntu 22.04 LTS 64bit pro with realtime-kernel
CPU:AMD® Ryzen 7 2700x eight-core processor × 16
Memory:31.3GB
Graphic:AMD Radeon RX 590 Series (polaris10, LLVM 15.0.7, DRM 3.57, 6.8.0-47-generic)

[ginga81]

If we go before 1375m, say 1350m, we can move to any track.

But it fails at the most important 1375m.

So I tried running it with old test data and it worked fine.

When the coordinates go wrong, I thought it would be helpful if the log output showed what information was used to move to which track.

Even if I press the P key to output the track number, no rail is displayed for the mysterious number.

When I searched the file as I did at the beginning, all the coordinates were aligned to 1375m, and there were no other strange coordinates.

Furthermore, if you want to return to Rail 0, if you forcefully change it from 1650m to the front, it should return.
However, it doesn't return and is thrown into the air.
I couldn't figure out which command in these three files was affecting it, and I was at a loss as to what to do, so I proposed this.

There are a lot of rail numbers used, but they're probably unrelated.

[ginga81]

player-path-not-good.zip
I thought it would be difficult because it was production data, but I was able to delete it successfully.
It doesn't jump at 1625m like in the video, but I was able to reproduce the initial behavior when moving from Rail 1 to Rail 2, and the behavior up to 1625m.

[leezer3]

OK, it took me a little while to figure out what's going on there.

The issue is that you're issuing the .RailEnd command in the same block as the switch.
This means that the switch isn't actually being created, and the position is moving on the direction vector of the last valid block.

Move the RailEnd command one block past the switch, and it should work OK.

I'll see if I can make this handle things better, but will be later on in the week before I do so.