burghout / mezzo

Mezzo mesoscopic traffic simulation
2 stars 0 forks source link

Second viable busline eliminated instead of merged in generated passenger path #59

Open Trateotu opened 5 years ago

Trateotu commented 5 years ago

Not sure what goes wrong just yet but pretty sure its a bug somewhere in Network::merge_paths_by_common_lines()

burghout commented 5 years ago

@Trateotu Can you write a test for this in the DRT_test?

Trateotu commented 5 years ago

@burghout Sure I need to prioritize other things at the moment but can write one in the future. Also I think it might be better documentation to add this to the integration test with the SF network, or to a 'CSGM' or 'Network' test group, since I do not think this is dependent on DRT being active.

burghout commented 5 years ago

@Trateotu ok will see what I can do this morning. Will open a bug branch from this from which of your branches?

Trateotu commented 5 years ago

@burghout ok thanks that would be great. I think David-DRT, I added a few updates and merged #57 there recently. Should we close #57 too?

Trateotu commented 5 years ago

@burghout Note: To be clear I do not believe that this occurs for ALL viable buslines that are to be merged into one path. I think that if we have two alternative lines for the same path then the second line is eliminated even when it seems like it should be kept. If, however, we have n (> 2) viable buslines that should be merged into one path then then I believe lines 1, 3, ... n are kept and the second one is elminated even though it should be kept.