Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Apollinaris Schoell
This is a roundabout for cars and a cycle path in circular shape around it. I 
would just not tag the cycle way with roundabout and map what is on the ground. 
 Don't tag for Merkaator …
 

On 23 Apr 2011, at 8:41 , Minko wrote:

> a) because it is part of that roundabout. See
> http://maps.google.nl/maps?f=q&source=s_q&hl=nl&geocode=&q=vathorst&aq=&sll=52.187628,5.415144&sspn=0.007656,0.022767&ie=UTF8&hq=&hnear=Vathorst,+Amersfoort,+Utrecht&ll=52.196813,5.406022&spn=0.000432,0.002846&z=19&layer=c&cbll=52.196813,5.406021&panoid=1w29L-X2unA1vfIPXIZPzg&cbp=11,47.28,,0,11.62
>  Even the streetname here is officially called rotonde #1 (rotonde=roundabout)
> b) you have to yield to the cycle traffic on the roundabout
> c) If you work with the merkaartor editor its very convenient to draw 
> roundabout circles ;-)
> ___
> mkgmap-dev mailing list
> mkgmap-dev@lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Dermot McNally
On 23 April 2011 16:41, Minko  wrote:
> a) because it is part of that roundabout. See

Yes, I see now, that alone makes it all make sense. And it also means
that the oneway=no suggestions of others won't work either.

The best I can come up with is to have separate ways for the cycle
path part, either not tagged as roundabout at all or tagged oneway=no.

I'd consider it very unwise to have roundabouts default to 2-way for
cyclists. The ones around here do enough mad stuff without having them
come the wrong way around a roundabout too...

Dermot

-- 
--
Igaühel on siin oma laul
ja ma oma ei leiagi üles
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Minko
a) because it is part of that roundabout. See
http://maps.google.nl/maps?f=q&source=s_q&hl=nl&geocode=&q=vathorst&aq=&sll=52.187628,5.415144&sspn=0.007656,0.022767&ie=UTF8&hq=&hnear=Vathorst,+Amersfoort,+Utrecht&ll=52.196813,5.406022&spn=0.000432,0.002846&z=19&layer=c&cbll=52.196813,5.406021&panoid=1w29L-X2unA1vfIPXIZPzg&cbp=11,47.28,,0,11.62
 Even the streetname here is officially called rotonde #1 (rotonde=roundabout)
b) you have to yield to the cycle traffic on the roundabout
c) If you work with the merkaartor editor its very convenient to draw 
roundabout circles ;-)
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Dermot McNally
I can't quite work out why we would consider a circular bit of road a
roundabout if it isn't one way. The whole point of a roundabout is
that you yield to traffic on the roundabout and go around in only one
direction. What is special about the conventions of such a circular
cycle traffic circle that would make us consider it a roundabout?

Dermot

-- 
--
Igaühel on siin oma laul
ja ma oma ei leiagi üles
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Minko
I agree, it should be done in the first place on osm. I've started to map the 
bike path roundabouts which allow twoway cycle traffic with oneway=no ;-)
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Lambertus
Op 23-04-11 10:42, Henning Scholland schreef:
> Hi Minko
> Might it be an fault in OSM-data? In Germany junction=roundabout imply
> oneway=yes. I don't know exactly how this is handled in other countries,
> but I won't be surprised, if this is globally implied. So if there is an
> roundabout, which isn't an oneway, you should add an oneway=no.
>
> Henning
>
> ___
> mkgmap-dev mailing list
> mkgmap-dev@lists.mkgmap.org.uk
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
I might add that I suspect Ligfietser (=recumbent cyclist) is talking 
about cycleway roundabouts (or cycleways that circle a roundabout). Many 
of these (at least in the Netherlands) allow cyclists to take the 
roundabout in both directions.

So, yes, for cars roundabouts will be oneways mostly everywhere, but not 
necessarily for cyclists or pedestrians. I also note that the wiki 
tagging page suggests that oneway=yes is implied, so it follows that 
every non-oneway roundabout should be tagged oneway=no?
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread harri
On 04/23/2011 10:12 AM, Minko wrote:
> I cannot {add oneway=no} with the style file rules because all roundabouts 
> are hardcoded set to oneway=yes.
> So this means if someone didnt tag a roundabout with a tag oneway=* there 
> will be allways oneway=yes and I cannot see if this yes is hardcoded or not.
> Or can I use something like junction=roundabout&  mkgmap:oneway=yes {set 
> oneway=no} ??
>
> Is it not better to add this rule to the default style file instead of hard 
> coding it?

Junction=roundabout implies oneway=yes. I think the best solution is to 
add oneway=no tag to the mapping data if it is missing.
--
harri
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Henning Scholland
Hi Minko
Might it be an fault in OSM-data? In Germany junction=roundabout imply 
oneway=yes. I don't know exactly how this is handled in other countries, 
but I won't be surprised, if this is globally implied. So if there is an 
roundabout, which isn't an oneway, you should add an oneway=no.

Henning

___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] junction=roundabout

2011-04-23 Thread Minko
Thanks Marko, I didn't notice it until now ;-)

I cannot {add oneway=no} with the style file rules because all roundabouts are 
hardcoded set to oneway=yes.
So this means if someone didnt tag a roundabout with a tag oneway=* there will 
be allways oneway=yes and I cannot see if this yes is hardcoded or not.
Or can I use something like junction=roundabout & mkgmap:oneway=yes {set 
oneway=no} ??

Is it not better to add this rule to the default style file instead of hard 
coding it?

___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


Re: [mkgmap-dev] mkgmap Exception in thread "main" java.lang.AssertionError

2011-04-23 Thread fla...@googlemail.com
I have the same error since 2 week with the germany.pbf-files from
ther geofabrik.

mkgmap svn 1922

java -Xmx4000M -ea -Dlog.config=/osm/garmin/aio/bin/logging.properties
-jar mkgmap-svn.jar
--max-jobs=4
--style-file=basemap_style
--remove-short-arcs=5
--description='aio-germany'
--country-name='germany'
--country-abbr='ger'
--area-name='germany'
--family-id=4
--product-id=45
--series-name='aio-germany'
--family-name=basemap
--latin1
--mapname=70013001 --draw-priority=10
--add-pois-to-areas --make-all-cycleways
--generate-sea=multipolygon,extend-sea-sectors,close-gaps=6000,floodblocker
--link-pois-to-ways
--route --index --gmapsupp
--report-dead-ends=2 --check-roundabouts --check-roundabout-flares
--keep-going --tdbfile --nsis --verbose
germany-raw/*.osm.gz
basemap.TYP

The ouput of gmaptool is :

gmt v0.4.2.1891n   (C) AP
Wrong header (first FAT), file gmapsupp.img


But now the strange thing :


1) running the africa.pbf-dump works well . Gmapsupp ist ok and work
on the garmin-hardware.

2) java -jar /mkgmap.jar --gmapsupp 70013*.img basemap.TYP meake a
working gmapsupp from the not-working data.

3) i tried to go back to svn1906 .. same error .. so i think there
might be a problem with the data ?


Perhaps i could help a little bit.
Greeting Dirk
___
mkgmap-dev mailing list
mkgmap-dev@lists.mkgmap.org.uk
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev