ABRP Avoids segment of US-90 in Louisiana

[KSevcik 1]

Description: Was planning a longer route to Morgan City, St Mary, LA, USA, and ABRP routed it strangely.  I started poking and it looks like ABRP doesn't believe a chunk of US 90 exists.

Type: Webpage planned

Link to Plan: https://abetterrouteplanner.com/?plan_uuid=151a8e38-3bdf-4757-be2d-bebca55b01d5

Replication Steps:

1. Click plan link

As you can see on the plan, the route only approaches the waypoints on the road, but never routes on the road.  Here's a link to part of that highway on OSM:

https://www.openstreetmap.org/way/117379972#map=17/29.70142/-91.38017

It's listed as a Trunk Road?  Not sure if that's relevant, but the problematic section is where it changes from "motorway" to "trunk".

[CBenson 1]

This really looks to me to be a coding issue in OSM:

https://www.openstreetmap.org/directions?engine=fossgis_osrm_car&route=29.72338%2C-91.40956%3B29.69412%2C-91.34048#map=10/30.0487/-91.5076

But I can't figure out what's wrong or how to fix it.

[KSevcik 1]

So I've figured out the issue, but I don't know OSM etiquette for fixing it.  The issue is that whole stretch has the tag proposed=motorway.  I assume because someone thinks that's how to tag it as a proposed upgrade to the existing highway?  But this breaks OSRM, per taginfo's notes on the proposed=motorway tag:

I assume the fix is to just delete the proposed=motorway tag, but I've never (yet) edited OSM, and don't know if that info should be shunted to some other tag/feature.

I guess the other fix is for OSRM to switch to whatever OSMaxx is doing and only ignore ways that are highway=proposed.  But I expect that's a heavier lift than just fixing the data in OSM.

[KSevcik 1]

Alright, I created an OSM account and deleted the tag myself.  I'm assuming that will fix it when the change propagates.  Also looks like I've collected a new hobby to amuse myself with at lunch.