I am looking at using the data from OS in graphhopper which is now available in OSM.PBF format and would like to take advantage of the avgspeed data contained within it. My PR for that got rejected as “avgspeed” is not a part of the OSM spec and karussel suggested that I create a specialised parser for the OS data.
I have a couple of questions:
Would creation of an OS specific parser be something that has a chance of being merged into core GH or was the intention that that would stay as a part of a fork?
If it could potentially be merged into core graphhopper, what would be the preferred implementation of that? Is it to create a separate OSCarAverageSpeedParser similar to how the standard one has been modded in the PR and then include that parser instead of the standard one in the ImportRegistry based upon a configuration parameter? Or create a separate ImportRegistry altogether?