I was using osm2pgrouting on few previous days to handle a large dataset (car road network for whole France). At start I tried to insert the whole dataset (OSM XML filtered to contain only roads), but was still too big to work. So I tried using incremental insert, but was struggling with the documentation. From my experiment, I deduced that :
It works only if data split is done by separating geographically data (like four distinct areas with a few hundred meters of overlap). Other kind of split (like inserting first motorways, then primary road, then secondary roads) doesn't work : edges are not split when necessary, making the graph not completely usable.
It works only with --addnodes option (but I'm not sure for this one, saw it on an issue)
So I would like to write a bit of documentation to describe more completely this use case (step by step split with osmium then import with options). Were would be the best place to do so ? In the readme ? In the project wiki ?
Hello,
I was using osm2pgrouting on few previous days to handle a large dataset (car road network for whole France). At start I tried to insert the whole dataset (OSM XML filtered to contain only roads), but was still too big to work. So I tried using incremental insert, but was struggling with the documentation. From my experiment, I deduced that :
--addnodesoption (but I'm not sure for this one, saw it on an issue)So I would like to write a bit of documentation to describe more completely this use case (step by step split with osmium then import with options). Were would be the best place to do so ? In the readme ? In the project wiki ?
Best regards.