generalizedCost is zero for legs nextLegs

[miklcct]

I am trying to replace the legs in an itinerary with a leg from nextLegs and recalculate the generalizedCost for the new itinerary. However, the generalizedCost for the legs in nextLegs are all zero so it is not possible to get an accurate figure for the new itinerary.

Expected behavior

The generalizedCost of the alternative legs are some sensitive numbers, similar to the original.

Observed behavior

the generalizedCost is 0.

Version of OTP used (exact commit hash or JAR name)

2.6.0

Data sets in use (links to GTFS and OSM PBF files)

Bus Open Data Service

Steps to reproduce the problem

Make a journey plan with nextLegs and generalizedCost fetched for the legs.

[optionsome]

We could discuss this in a dev meeting at some point if you are able to join.