[mkgmap-dev] access=no//access=private on short lines vs long lines

Thu Aug 30 15:39:44 BST 2012

On Thu, Aug 30, 2012 at 11:03:55AM +0200, WanMil wrote:
>1. Highways are often divided into several parts due to tagging 
>bridges, maxspeed, lanes etc. So in your example all bridges tagged 
>with access=private|no will disappear although they might belong to 
>mucher longer highways. Is that ok for you? Do you have an idea for a 
>solution how to avoid that?

One more reason for splitting roads are route relations.

On a related note, if a hyper-precise junction=roundabout is split 
because of bus route relations, mkgmap will start emitting warnings like 
this:

2012/08/30 09:42:59 WARNING (StyledConverter): 63240002.osm.pbf: 
Roundabout segment 176923123 direction looks wrong (see 
http://www.openstreetmap.org/?mlat=60.37620&mlon=25.27184&zoom=17)
2012/08/30 09:43:00 WARNING (StyledConverter): 63240002.osm.pbf: 
Roundabout segment 176875842 direction looks wrong (see 
http://www.openstreetmap.org/?mlat=60.37657&mlon=25.26866&zoom=17)

Maybe the roundabout code should merge the roundabout before checking 
the segment directions? Mkgmap did not warn before I split these 
roundabouts. I do not remember touching any nodes when splitting them.

>2. You want to introduce some kind of functions in the style system.

This would indeed require some planning. Maybe it would be worthwhile to 
study the theory behind grammar attributes in parser generators, and 
then come up with a design? Perhaps there already are some useful 
implementations that cache generated attribute values? Even if existing 
code cannot be copied, the algorithms could.

	Marko