logo separator

[mkgmap-dev] Bug in Road Merging - actually doubling roads.

From Henning Scholland osm at hscholland.de on Thu Mar 29 14:48:16 BST 2018

So far I know there are already worldwide available qs tools for it. So just reporting them seems to be not necessary. After thinking about it, you are right. There will be complicated cases which are not easy to fix by code. 

Henning 

On 29 Mar 2018, 15:39, at 15:39, Gerd Petermann <gpetermann_muenchen at hotmail.com> wrote:
>Hi Henning,
>
>well, I can start to code the test to detect and report overlapping
>highways. Maybe there is a way to fix some of them automatically,
>but I'd really prefer to just report them so that they are fixed in
>OSM. In most cases this should be easy.
>
>Gerd
>
>________________________________________
>Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von
>Henning Scholland <osm at hscholland.de>
>Gesendet: Donnerstag, 29. März 2018 13:48:46
>An: Development list for mkgmap
>Betreff: Re: [mkgmap-dev] Bug in Road Merging - actually doubling
>roads.
>
>Hi Gerd,
>A hopefully easy solution could be to just take the first connection
>between two nodes and if there will be a second one, just copy name-tag
>for house numbers and then skip that second way. Or just report them.
>Can be two options like: - - duplicate-ways=remove/report
>
>Henning
>On 29 Mar 2018, at 09:32, Gerd Petermann
><gpetermann_muenchen at hotmail.com<mailto:gpetermann_muenchen at hotmail.com>>
>wrote:
>
>Hi Felix,
>
>sorry to hear. I am running out of ideas for a simple patch  :-(
>Maybe it is not the number of arcs with the same initial bearing but
>the number of arcs going to the same destination node.
>
>Anyway, seems we really have to remove overlapping parts of roads much
>earlier to make your style work in that situation,
>and I think this would open a new box of worms reg. housenumbers, turn
>restrictions and so on, in other words,
>I fear we might add a lot of complex code that is rather bad for other
>users.
>So, maybe it would be better to just fix those errors in OSM?
>
>Gerd
>
>________________________________
>
>Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von
>Felix Hartmann <extremecarver at gmail.com>
>Gesendet: Mittwoch, 28. März 2018 23:10:16
>An: Development list for mkgmap
>Betreff: Re: [mkgmap-dev] Bug in Road Merging - actually doubling
>roads.
>
>Sorry for taking so long - needed to setup a new desktop computer with
>my old mainboard dead. The new patch does not really seem to improve
>anything. It's not dropping the route in simulation mode anymore - but
>getting stuck and not routing along.
>
>On 23 March 2018 at 09:50, Felix Hartmann
><extremecarver at gmail.com<mailto:extremecarver at gmail.com>> wrote:
>Well all the problems about problematic arcs so far came up from a
>combination of two things:
>a) my style creates two additional ways (bit different
>road_class/road_speed based on input) if the following is met: route
>relation hiking, bicycle or mtb | ( mtb:scale>0 & mtb:scale<5) |
>class:bicycle>1. Additionally if a road is oneway - and it's a low
>level road (so neither motorway,trunk,primary,secondary,tertiary) - and
>oneway, a low class/speed non oneway road is added as well (intention
>to have lower priority against the oneway than driving with the oneway
>direction. Also additionally if bicycle lane against the traffic flow
>is possible, it will create a line against the oneway.
>b) mistakes in OSM data where wrongly several objects lie on top of
>each other.
>
>
>Sorry my computer crashed yesterday evening, and I cannot boot it right
>now. Seems my Mainboard is dead. So I probably won't have possibilities
>for testing before Monday or Tuesday.
>
>On 23 March 2018 at 09:39, Gerd Petermann
><gpetermann_muenchen at hotmail.com<mailto:gpetermann_muenchen at hotmail.com>>
>wrote:
>Hi Henning,
>
>Henning Scholland wrote
>Hi Gerd, I just was thinking about if one road is leaving the
>roundabout
> at two angles it may count on the Garmin as two roads and will lead to
> 'exit on 3rd exit' instead of 'exit on 2nd exit' for example.
>
> But I haven't gave it a try yet.
>>
>ah, okay, got it. I think Garmin should calculate the number for the
>exit
>without "thinking" about the further
>route but who knows.
>I'd like to test it on my own but I found it too difficut to create a
>style
>that tries to do the same as Felix'
>style, and you don't get the same results when you simply add
>overlapping
>highways in JOSM.
>
>I am not aware of another style that adds multiple routable ways for
>one OSM
>way, and Felix doesn't want to share his, so testing is his work. I can
>only
>make educated guesses about the effects,
>and of course I'll test it with other styles before committing it.
>
>Gerd
>
>
>
>
>
>
>
>
>
>--
>Sent from:
>http://gis.19327.n8.nabble.com/Mkgmap-Development-f5324443.html
>________________________________
>
>mkgmap-dev mailing list
>mkgmap-dev at lists.mkgmap.org.uk<mailto:mkgmap-dev at lists.mkgmap.org.uk>
>http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>
>
>
>--
>Felix Hartman - Openmtbmap.org<http://Openmtbmap.org> &
>VeloMap.org<http://VeloMap.org>
>Schusterbergweg 32/8
>6020 Innsbruck
>Austria - Österreich
>
>
>
>--
>Felix Hartman - Openmtbmap.org<http://Openmtbmap.org> &
>VeloMap.org<http://VeloMap.org>
>Schusterbergweg 32/8
>6020 Innsbruck
>Austria - Österreich
>________________________________
>
>mkgmap-dev mailing list
>mkgmap-dev at lists.mkgmap.org.uk
>http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>_______________________________________________
>mkgmap-dev mailing list
>mkgmap-dev at lists.mkgmap.org.uk
>http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20180329/5e902b00/attachment-0001.html>


More information about the mkgmap-dev mailing list