NAVAND-1108: test that reproduces the issue

[VysotskiVadim]

No description provided.

[github-actions]

Changelog

Features

• Added RoadComponent.language value. #6833
• ⚠️ Changed EHorizonEdgeMetadata.names class from RoadName to RoadComponent. #6833
  To migrate change your code from:

val shielded = roadName.shielded

into:

val shielded = roadComponent.shield != null

• Added support for continuous EV alternatives in NavigationRouteAlternativesObserver. #6833
• Fixed issues with map-matching to HOV-only roads. #6833
• Set a limit of simultaneously running onboard route requests to avoid too many tasks blocking too much of the device's computing resources. #6833
• Fixed an issue with Road Access Policy ignoring the setting to map-match to closed road sections, when enabled. #6833
• Added MapboxTripStarter to simplify the solution for managing the trip session and replaying routes. This also makes it possible to share the replay state between drop-in-ui and android-auto. #6794

Bug fixes and improvements

• Increased max distance from the user indicator to route line valid to continue vanishing updates from 3m to 10m. #6854
• Fixed an issue where alternative routes might have had an incorrect or incomplete portion of the route refreshed or occasionally fail to refresh. #6848
• Removed NavigationRoute#hasUnexpectedClosures and added RouteProgress#hasUnexpectedUpcomingClosures instead that checks whether route has upcoming unexpected closures (the algorithm does not take into account closures that the puck has already been on) #6841
• Improved NavigationView camera behavior to go back into overview state if routes change during route preview state. #6840
• Improved NavigationView camera behavior to ignore keyboard insets. #6845

Known issues ⚠️

Other changes

Android Auto Changelog

Features

• Added support for Junction Views. #6849

Bug fixes and improvements

• Optimized SpeedLimitWidget memory usage. #6859

[codecov]

Codecov Report

Merging #6869 (7b9a80e) into main (09b5248) will not change coverage.
The diff coverage is n/a.

@@            Coverage Diff            @@
##               main    #6869   +/-   ##
=========================================
  Coverage     72.68%   72.68%           
  Complexity     5572     5572           
=========================================
  Files           782      782           
  Lines         30166    30166           
  Branches       3562     3562           
=========================================
  Hits          21926    21926           
  Misses         6814     6814           
  Partials       1426     1426           

Impacted Files	Coverage Δ
...n/base/trip/model/roadobject/UpcomingRoadObject.kt	21.05% <ø> (ø)

[dzinad]

Why do we expect them to be equal?
The routes are different, do they have the same geometry (at least until the incident)? If yes, it should be specified in the comments. If the geometry is different, distanceToStart may be different (it's distance along the route, not direct distance, right?).

[VysotskiVadim]

Why do we expect them to be equal?

Current location doesn't change and both routers heads to the same direction thought the same incident.

[VysotskiVadim]

If yes, it should be specified in the comments.

Thank you for the feedback. I will try to improve readability of the test

[VysotskiVadim]

@dzinad , I updated the test. Do you understand what and why it checks now?

[dzinad]

My point is that if we have the same origin and the same incident location, when we add a waypoint, distances may still differ. See the drawing.
And it's still not clear that the geometries from the origin until the incident are exactly the same.
Or am I missing something?

[VysotskiVadim]

I see your point, thank you. The routes I test follow the same roads. Like

[dzinad]

So the waypoint is strictly on the route's geometry? Could you mention it somewhere in the comment?