[OSM-dev] osm2pgsql, direction of a virtual way (based on a route)

[André Joost]

Am 26.10.11 19:03, schrieb marqqs at gmx.eu:
> Hi Igor,
>
>> are you sure? I thought one of the major goals of the _new_ PT
>> schema was precisely to eliminate this forward/backward mess with
>> the "relation per route variant" concept. Just looked it up [1]:
>
> No, I'm indeed very unsure in aspects of the new schema. If the ways
> don't have roles, the direction of the route will have to be
> determined based alone on the sequence of ways in the route.
>

No, you have to look at the from and to tags of the relation. The 
relation should be ordered in that way, so from the first to the last 
way-member should be the direction.

Except for those cases, whre platforms are ways or areas, not just 
nodes. But those shuold have a platform-tag.

HTH
André Joost