Oneway=no vs. motorway_link routing
Hi, I tried a route along this motorway exit: http://data.giub.uni-bonn.de/openrouteservice/index.php?start=10.0073556,49.... on my etrex and in mapsource. Both avoided the exit. The part of the motorway_link which is common for both directions has "oneway_no". The routing on openroutservice shows this, too. My rule for the line type is: highway=motorway_link {add oneway=yes; add bicycle=no; add foot=no} [0x09 road_class=4 road_speed=3 level 4] The "add oneway=yes" should only make it oneway if it is NOT "oneway=no", right? If I look at the line in GPSMapEdit it says "Polyline has direction". Does this mean that line type 0x09 is always oneway for Garmin or is the tag setting in mkgmap buggy? I used r973. BTW: GPSMapEdit always shows "Polyline has direction" but not "One way" in the "routing" tab, even for residential one-ways. Is that correct? I am sure this type of motorway exit worked before I switched to type 0x09 (the reason was better routing instructions).
On Mon, Mar 23, 2009 at 3:28 PM, Ralf Kleineisel <ralf@kleineisel.de> wrote:
Hi,
I tried a route along this motorway exit:
http://data.giub.uni-bonn.de/openrouteservice/index.php?start=10.0073556,49....
on my etrex and in mapsource. Both avoided the exit.
I have the following entry in my lines style file: highway=motorway_link {add bicycle = no; add foot = no } [0x09 road_class=4 road_speed=3 resolution 17] I was able to correctly route your example in MapSource. This would mean: - 0x09 is NOT inherently one-way. - "oneway=no" does not necessarily override the "add oneway=yes;" directive in mkgmap. The easiest solution would be to also remove "add oneway=yes;" from your lines file. I have so far not encountered a case where I was routed the wrong direction on a motorway_link. Most likely the reason is that the actual motorway element will have a one-way direction, which will be the determining factor in calculating the route. Perhaps you could try this and report on your results? It may be necessary, however, to clarify the precedence of oneway attributes in mkgmap. Hope this helps.
Clinton Gladstone wrote:
This would mean:
- 0x09 is NOT inherently one-way.
- "oneway=no" does not necessarily override the "add oneway=yes;" directive in mkgmap.
The easiest solution would be to also remove "add oneway=yes;" from your lines file. I have so far not encountered a case where I was routed the wrong direction on a motorway_link. Most likely the reason is that the actual motorway element will have a one-way direction, which will be the determining factor in calculating the route.
Perhaps you could try this and report on your results?
You are right. Removing the "add oneway=yes" helped. It seems that mkgmap overwrites existing "oneway=no" tags if I use "add oneway=yes". This is contrary to the documentation in http://wiki.openstreetmap.org/wiki/Mkgmap/help/style_rules#add . Is this a bug or is the documentation wrong?
participants (2)
-
Clinton Gladstone -
Ralf Kleineisel