WaitAtBeginningFactor ignored when using optimizeOrReverse method

NOTE: this issue system is intended for reporting bugs and tracking progress in software development. For all other usage and software development questions or discussion, please write to one of the mailing lists (Google groups): https://groups.google.com/forum/#!forum/opentripplanner-dev https://groups.google.com/forum/#!forum/opentripplanner-users

Expected behavior

WaitAtBeginningFactor should be applied so that identical trips that depart later have a higher weight.

Observed behavior

WaitAtBeginningFactor is ignored. However, if optimize is disabled by modifying line 370 in AStar.java, WaitAtBeginningFactor is applied correctly.

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

210b2b38ed579776c04d5d6e70954f122d7a8b5b (latest)

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

https://drive.google.com/file/d/1g8dURCXpMyRUttGfiuxsu4GD26Pq_7yj/view?usp=sharing

(Minimal GTFS set)

Command line used to start OTP

--build --inMemory

Router config and graph build config JSON

None. WaitAtBeginningFactor defaults to 0.4.

Steps to reproduce the problem

Send the following request:

http://localhost:8080/otp/routers/default/plan?fromPlace=60.192955%2C11.097078&toPlace=59.91132021529975%2C10.754928588867188&time=9%3A35am&date=06-01-2017&mode=TRANSIT%2CWALK&maxWalkDistance=804.672&arriveBy=false&wheelchair=false&locale=en

Put a breakpoint here and inspect the last state of the first result:

https://github.com/opentripplanner/OpenTripPlanner/blob/210b2b38ed579776c04d5d6e70954f122d7a8b5b/src/main/java/org/opentripplanner/routing/impl/GraphPathFinder.java#L205

Set optimize to false here and run a new search. The weight should be different, but the itinerary should otherwise be identical.

https://github.com/opentripplanner/OpenTripPlanner/blob/210b2b38ed579776c04d5d6e70954f122d7a8b5b/src/main/java/org/opentripplanner/routing/algorithm/AStar.java#L370

opentripplanner / OpenTripPlanner