Since trunk mkgmap-r3230 the routing has apperently changed and routing has completely gone wrong on my OFM styles. My last Benelux maps are made with mkgmap-r3193 and there the routing seems ok. Gerd, do you have made any important changes between r3193 and r3230 that could cause this? Which version could I test? I also tried mkgmap-performance2-r3245, time reduction is 12% but the routing is still weird.
Some more observations mkgmap-refactoring-r3220: routing ok mkgmap-r3229 routing: not ok It seems to be style file related with oneway=yes and cycleway=opposite tags In my style file I remove the oneway tag if there is a cycleway=opposite or opposite_lane because this road must be accesible from both sides. And this goes wrong after r3220. The whole road is flagged with access=no for all vehicles. One way and toll flags are not I dont use multiple routable lines in these cases, just one routable way which should be accessible for all vehicles. I also dont use --make-opposite-cycleways because its not relevant for my maps. Any idea what went wrong here?
Gerd, do you have made any important changes between r3193 and r3230 that could cause this? Which version could I test?
I also tried mkgmap-performance2-r3245, time reduction is 12% but the routing is still weird.
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Minko, I have no idea. Can I use the files in http://mijndev.openstreetmap.nl/~ligfietser/openfietsmap/Scripts/ to reproduce the error? Please give me an OSM id of a way that produces wrong results. Gerd ligfietser wrote
Some more observations mkgmap-refactoring-r3220: routing ok mkgmap-r3229 routing: not ok
It seems to be style file related with oneway=yes and cycleway=opposite tags In my style file I remove the oneway tag if there is a cycleway=opposite or opposite_lane because this road must be accesible from both sides. And this goes wrong after r3220. The whole road is flagged with access=no for all vehicles. One way and toll flags are not
I dont use multiple routable lines in these cases, just one routable way which should be accessible for all vehicles. I also dont use --make-opposite-cycleways because its not relevant for my maps. Any idea what went wrong here?
Gerd, do you have made any important changes between r3193 and r3230 that could cause this? Which version could I test?
I also tried mkgmap-performance2-r3245, time reduction is 12% but the routing is still weird.
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804849.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi Gerd Yes you can take those styles. All roads with oneway=yes & cycleway=opposite(_lane) are getting no access for all vehicles. See http://overpass-turbo.eu/s/3dz
I have no idea. Can I use the files in http://mijndev.openstreetmap.nl/~ligfietser/openfietsmap/Scripts/ to reproduce the error?
Please give me an OSM id of a way that produces wrong results.
Gerd
Hi Minko, the problem was introduced with r3227. It's a combination of two errors : I forgot that overlays are still supported, and your style uses them. As a result, all ways with overlay types were no longer recognized as roads. A 2nd error prohibited that the corresponding warning is written :-( The error is fixed with r3246 in trunk and r3247 in the branch. Gerd ligfietser wrote
Hi Gerd Yes you can take those styles. All roads with oneway=yes & cycleway=opposite(_lane) are getting no access for all vehicles. See http://overpass-turbo.eu/s/3dz
I have no idea. Can I use the files in http://mijndev.openstreetmap.nl/~ligfietser/openfietsmap/Scripts/ to reproduce the error?
Please give me an OSM id of a way that produces wrong results.
Gerd
mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804871.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
Thanks Gerd, mkgmap-r3247 is working as expected. FYI: improvements in the performance is less than r3245, 3% vs 12% faster than r3193 (with 1% more data than last week). I don't mind a few minutes more or less, my Benelux map is compiled in 18 minutes or so. ;-)
Hi Minko,
the problem was introduced with r3227. It's a combination of two errors : I forgot that overlays are still supported, and your style uses them. As a result, all ways with overlay types were no longer recognized as roads. A 2nd error prohibited that the corresponding warning is written :-(
The error is fixed with r3246 in trunk and r3247 in the branch.
Gerd
Hi Minko,
Thanks Gerd, mkgmap-r3247 is working as expected.
fine :-)
FYI: improvements in the performance is less than r3245, 3% vs 12% faster than r3193 (with 1% more data than last week).
I don't mind a few minutes more or less, my Benelux map is compiled in 18 minutes or so. ;-)
Yes, sure. It's not a big breakthrough, and I don't expect find one for normal cases. I just like to find faster algos or data structures, so I hope WanMil and Steve are stopping me when I am starting to mess up the code ... At least the error above was not caused by optimization. Gerd
Hi Minko,
the problem was introduced with r3227. It's a combination of two errors : I forgot that overlays are still supported, and your style uses them. As a result, all ways with overlay types were no longer recognized as roads. A 2nd error prohibited that the corresponding warning is written :-(
The error is fixed with r3246 in trunk and r3247 in the branch.
Gerd
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Faster, less memory hungry code is always welcome Gerd! :) On 02/05/2014 11:32, Gerd Petermann wrote:
Hi Minko,
Thanks Gerd, mkgmap-r3247 is working as expected.
fine :-)
FYI: improvements in the performance is less than r3245, 3% vs 12% faster than r3193 (with 1% more data than last week).
I don't mind a few minutes more or less, my Benelux map is compiled
in 18 minutes or so. ;-)
Yes, sure. It's not a big breakthrough, and I don't expect find one for normal cases. I just like to find faster algos or data structures, so I hope WanMil and Steve are stopping me when I am starting to mess up the code ... At least the error above was not caused by optimization.
Gerd
Hi Minko,
the problem was introduced with r3227. It's a combination of two errors : I forgot that overlays are still supported, and your style uses them. As a result, all ways with overlay types were no longer recognized as roads. A 2nd error prohibited that the corresponding warning is written :-(
The error is fixed with r3246 in trunk and r3247 in the branch.
Gerd
_______________________________________________ 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
Hi Minko, forgot to mention that --check-styles complains about line 266 in the lines file: highway=bridleway & bicycle!=yes & bicycle!=designated & bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 road_speed=0 resolution 23] Type 0x2915 is in the overlays file : 0x2915: 0x15, 0x29 Do you think type 0x15 is routable? Gerd
Date: Fri, 2 May 2014 10:59:31 +0200 From: ligfietser@online.nl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing
Thanks Gerd, mkgmap-r3247 is working as expected.
FYI: improvements in the performance is less than r3245, 3% vs 12% faster than r3193 (with 1% more data than last week).
I don't mind a few minutes more or less, my Benelux map is compiled in 18 minutes or so. ;-)
Hi Minko,
the problem was introduced with r3227. It's a combination of two errors : I forgot that overlays are still supported, and your style uses them. As a result, all ways with overlay types were no longer recognized as roads. A 2nd error prohibited that the corresponding warning is written :-(
The error is fixed with r3246 in trunk and r3247 in the branch.
Gerd
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Gerd Yes, seems 0x15 is routable.
forgot to mention that --check-styles complains about line 266 in the lines file: highway=bridleway & bicycle!=yes & bicycle!=designated & bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 road_speed=0 resolution 23]
Type 0x2915 is in the overlays file : 0x2915: 0x15, 0x29
Do you think type 0x15 is routable?
Gerd
Hi Minko, yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 296 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 297 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 298 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 299 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 341 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 342 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345 So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines? Gerd
Date: Fri, 2 May 2014 13:12:15 +0200 From: ligfietser@online.nl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing
Hi Gerd Yes, seems 0x15 is routable.
forgot to mention that --check-styles complains about line 266 in the lines file: highway=bridleway & bicycle!=yes & bicycle!=designated & bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 road_speed=0 resolution 23]
Type 0x2915 is in the overlays file : 0x2915: 0x15, 0x29
Do you think type 0x15 is routable?
Gerd
mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
I know that a long time ago I tried 0x15 - and it would not route on my Vista HCx - but mind - that's a long time ago. Never tried id since... On 02.05.2014 14:54, Gerd Petermann wrote:
Hi Minko,
yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 296 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 297 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 298 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 299 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 341 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 342 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345
So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines?
Gerd
Date: Fri, 2 May 2014 13:12:15 +0200 From: ligfietser@online.nl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing
Hi Gerd Yes, seems 0x15 is routable.
forgot to mention that --check-styles complains about line 266 in the lines file: highway=bridleway & bicycle!=yes & bicycle!=designated & bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 road_speed=0 resolution 23]
Type 0x2915 is in the overlays file : 0x2915: 0x15, 0x29
Do you think type 0x15 is routable?
Gerd
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
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Hi Felix, on the other hand, the test that prints the messages below may also be out-dated. I am no longer sure that the use of routable types for non-routable lines really breaks routing. I'll have to search the archives to find out why I've coded that test. Gerd Date: Fri, 2 May 2014 14:56:10 +0200 From: extremecarver@gmail.com To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing I know that a long time ago I tried 0x15 - and it would not route on my Vista HCx - but mind - that's a long time ago. Never tried id since... On 02.05.2014 14:54, Gerd Petermann wrote: Hi Minko, yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 296 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 297 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 298 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 299 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 341 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 342 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345 So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines? Gerd > Date: Fri, 2 May 2014 13:12:15 +0200 > From: ligfietser@online.nl > To: mkgmap-dev@lists.mkgmap.org.uk > Subject: Re: [mkgmap-dev] strange routing > > Hi Gerd > Yes, seems 0x15 is routable. > > > > forgot to mention that --check-styles complains about line 266 in the > > lines file: > > highway=bridleway & bicycle!=yes & bicycle!=designated & > > bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = > > no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 > > road_speed=0 resolution 23] > > > > Type 0x2915 is in the overlays file : > > 0x2915: 0x15, 0x29 > > > > Do you think type 0x15 is routable? > > > > Gerd > _______________________________________________ > 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 -- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Well, I think using nonroutable lines for routable types broke the address search (you tried to get somewhere that didn't exist for routing or so...). But I'm not sure what it was about exactly. I think it broke something because I remember I used 0x00 and 0x01 unroutable long time ago and stopped using it.. On the other hand, if 0x15 is routable without problems now, could this be true for all types? - Assuming the nonroutable types are actually routable, just mkgmap cannot create them routable??? On 02.05.2014 14:59, Gerd Petermann wrote:
Hi Felix,
on the other hand, the test that prints the messages below may also be out-dated. I am no longer sure that the use of routable types for non-routable lines really breaks routing. I'll have to search the archives to find out why I've coded that test.
Gerd
------------------------------------------------------------------------ Date: Fri, 2 May 2014 14:56:10 +0200 From: extremecarver@gmail.com To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing
I know that a long time ago I tried 0x15 - and it would not route on my Vista HCx - but mind - that's a long time ago. Never tried id since... On 02.05.2014 14:54, Gerd Petermann wrote:
Hi Minko,
yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 296 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 297 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 298 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 299 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 341 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 342 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345
So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines?
Gerd
> Date: Fri, 2 May 2014 13:12:15 +0200 > From: ligfietser@online.nl <mailto:ligfietser@online.nl> > To: mkgmap-dev@lists.mkgmap.org.uk <mailto:mkgmap-dev@lists.mkgmap.org.uk> > Subject: Re: [mkgmap-dev] strange routing > > Hi Gerd > Yes, seems 0x15 is routable. > > > > forgot to mention that --check-styles complains about line 266 in the > > lines file: > > highway=bridleway & bicycle!=yes & bicycle!=designated & > > bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = > > no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 > > road_speed=0 resolution 23] > > > > Type 0x2915 is in the overlays file : > > 0x2915: 0x15, 0x29 > > > > Do you think type 0x15 is routable? > > > > Gerd > _______________________________________________ > mkgmap-dev mailing list > mkgmap-dev@lists.mkgmap.org.uk <mailto: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 <mailto:mkgmap-dev@lists.mkgmap.org.uk> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails
Felix openmtbmap.org &www.velomap.org <http://www.velomap.org>
_______________________________________________ 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
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Hi Felix, yes, that's my concern as well. Maybe the active routing changed something? I'll try to create a small test. I am pretty sure that extended types are not allowed for routing. Also, routable maps from Garmin don't seem to use any type for roads which we call non-routable. Gerd Felix Hartmann-2 wrote
Well, I think using nonroutable lines for routable types broke the address search (you tried to get somewhere that didn't exist for routing or so...). But I'm not sure what it was about exactly. I think it broke something because I remember I used 0x00 and 0x01 unroutable long time ago and stopped using it..
On the other hand, if 0x15 is routable without problems now, could this be true for all types? - Assuming the nonroutable types are actually routable, just mkgmap cannot create them routable??? On 02.05.2014 14:59, Gerd Petermann wrote:
Hi Felix,
on the other hand, the test that prints the messages below may also be out-dated. I am no longer sure that the use of routable types for non-routable lines really breaks routing. I'll have to search the archives to find out why I've coded that test.
Gerd
------------------------------------------------------------------------ Date: Fri, 2 May 2014 14:56:10 +0200 From:
extremecarver@
To:
mkgmap-dev@.org
Subject: Re: [mkgmap-dev] strange routing
I know that a long time ago I tried 0x15 - and it would not route on my Vista HCx - but mind - that's a long time ago. Never tried id since... On 02.05.2014 14:54, Gerd Petermann wrote:
Hi Minko,
yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 296 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 297 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 298 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 299 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 341 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 342 Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345
So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines?
Gerd
> Date: Fri, 2 May 2014 13:12:15 +0200 > From:
ligfietser@
<mailto:
ligfietser@
>
> To:
mkgmap-dev@.org
<mailto:
mkgmap-dev@.org
>
> Subject: Re: [mkgmap-dev] strange routing > > Hi Gerd > Yes, seems 0x15 is routable. > > > > forgot to mention that --check-styles complains about line 266 in the > > lines file: > > highway=bridleway & bicycle!=yes & bicycle!=designated & > > bicycleroute!=yes & mtb_from_relation!=yes & mtb!=yes {add access = > > no; add foot = yes; set mkgmap:unpaved=1} [0x2915 road_class=0 > > road_speed=0 resolution 23] > > > > Type 0x2915 is in the overlays file : > > 0x2915: 0x15, 0x29 > > > > Do you think type 0x15 is routable? > > > > Gerd > _______________________________________________ > mkgmap-dev mailing list >
mkgmap-dev@.org
<mailto:
mkgmap-dev@.org
>
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
<mailto:
mkgmap-dev@.org
>
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails
Felix openmtbmap.org &www.velomap.org <http://www.velomap.org>;
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- keep on biking and discovering new trails
Felix openmtbmap.org & www.velomap.org
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804925.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
I can confirm that 0x15 is routable on Basecamp as well. Interestingly, it is not included in any of the European Garmin topo typ files, unlike 0x16 -- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804961.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
Gerd, It happens when you search for an address that is located at a line which is routable, but made non-routable for some reason (for instance if bicycle=no, someone who makes a bike map wants to remove the road parameters for a line like highway=secondary [0x04 resolution 20] This was done because of bicycle=no was not recognized anymore by Basecamp (until recently). On the osm forum there are some examples. http://forum.openstreetmap.org/viewtopic.php?pid=208081#p208081 http://forum.openstreetmap.org/viewtopic.php?id=13884&p=5 I dont know it is still the case with the latest releases.
Onderwerp: Re: [mkgmap-dev] strange routing Hi Felix,
on the other hand, the test that prints the messages below may also be out-dated. I am no longer sure that the use of routable types for non-routable lines really breaks routing. I'll have to search the archives to find out why I've coded that test.
Gerd
Hi Minko, okay, I have to try that. My understanding is that address search knows only routable ways (those that appear in NOD). And to NOD we write all lines with road_class/road_speed attribute and a non-extended type. Maybe Garmin software expects (or expected) to find NOD data for all lines with routable types. Gerd
Date: Fri, 2 May 2014 15:11:57 +0200 From: ligfietser@online.nl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] strange routing
Gerd, It happens when you search for an address that is located at a line which is routable, but made non-routable for some reason (for instance if bicycle=no, someone who makes a bike map wants to remove the road parameters for a line like highway=secondary [0x04 resolution 20] This was done because of bicycle=no was not recognized anymore by Basecamp (until recently). On the osm forum there are some examples.
http://forum.openstreetmap.org/viewtopic.php?pid=208081#p208081 http://forum.openstreetmap.org/viewtopic.php?id=13884&p=5
I dont know it is still the case with the latest releases.
Onderwerp: Re: [mkgmap-dev] strange routing Hi Felix,
on the other hand, the test that prints the messages below may also be out-dated. I am no longer sure that the use of routable types for non-routable lines really breaks routing. I'll have to search the archives to find out why I've coded that test.
Gerd
_______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Gerd, I have checked if those routing errors still exist by making a street non routable, eg highway=* & name=x [0x05 resolution 20] The addresses at street x are not in the index anymore and therefore not findable. A few years ago those were still in the index and caused problems. But now this bug seems eliminated, those warning messages are also not needed anymore? Now I've tried what happens if I make this street x with type 0x15 routable. highway=* & name=x [0x15 road_class=2 road_speed=2 resolution 20] 0x14, 0x15, 0x19 even 0x32 are routable, type 0x10014 not So I guess there are more routable types? Which types appear in NOD?
Hi Minko,
okay, I have to try that. My understanding is that address search knows only routable ways (those that appear in NOD). And to NOD we write all lines with road_class/road_speed attribute and a non-extended type. Maybe Garmin software expects (or expected) to find NOD data for all lines with routable types.
Gerd
On 02.05.2014 15:53, Minko wrote:
Hi Gerd, I have checked if those routing errors still exist by making a street non routable, eg highway=* & name=x [0x05 resolution 20]
The addresses at street x are not in the index anymore and therefore not findable. A few years ago those were still in the index and caused problems. But now this bug seems eliminated, those warning messages are also not needed anymore?
Now I've tried what happens if I make this street x with type 0x15 routable. highway=* & name=x [0x15 road_class=2 road_speed=2 resolution 20]
0x14, 0x15, 0x19 even 0x32 are routable, type 0x10014 not So I guess there are more routable types? Which types appear in NOD? Are they also shown and routbale in: Mapsource 6.13 Mapsource 6.16 Basecamp (don't mind version 1 or 2, they had too many problems and are not related much to GPS) - so just latest version new generation GPS like Oregon old generation like Vista HCx
and the jackpot - Nuvi?? Nuvis are somehower superspecial, and really behave different. On many Nuvis lots of extended types that show on all other units, do not show...
Hi Minko,
okay, I have to try that. My understanding is that address search knows only routable ways (those that appear in NOD). And to NOD we write all lines with road_class/road_speed attribute and a non-extended type. Maybe Garmin software expects (or expected) to find NOD data for all lines with routable types.
Gerd
mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Felix wrote
Are they also shown and routbale in: Mapsource 6.13 Mapsource 6.16 Basecamp (don't mind version 1 or 2, they had too many problems and are not related much to GPS) - so just latest version new generation GPS like Oregon old generation like Vista HCx
I did a few tests, it works on Basecamp 4.2.5 and Mapsource 6.16.3 Dakota 20, Nüvi 310 Maybe we need to create a big demo test map with all lines available?
and the jackpot - Nuvi?? Nuvis are somehower superspecial, and really behave different. On many Nuvis lots of extended types that show on all other units, do not show...
Yeah I know, that will be a problem. I also experienced this on the latest Monterra. The map display on that device is totally crap. :-/
Hi Minko, ligfietser wrote
0x14, 0x15, 0x19 even 0x32 are routable, type 0x10014 not So I guess there are more routable types? Which types appear in NOD?
All non-extended valid line types (0x00 - 0x3f) with road_speed/road_class. Gerd -- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804936.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
but what about extended types? I do suspect that garmin uses some extended types for their activity routing... Would it be possible to test how they behave - I assume right now mkgmap simply cannot write them routable... On 02.05.2014 16:15, GerdP wrote:
Hi Minko,
ligfietser wrote
0x14, 0x15, 0x19 even 0x32 are routable, type 0x10014 not So I guess there are more routable types? Which types appear in NOD? All non-extended valid line types (0x00 - 0x3f) with road_speed/road_class.
Gerd
-- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804936.html Sent from the Mkgmap Development mailing list archive at Nabble.com. _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-- keep on biking and discovering new trails Felix openmtbmap.org & www.velomap.org
Good question, wish I knew it ;-) BTW I noticed that some types are invisible in Mapsource, like 0x3c, even with a typ file. I also dont see any name nor address on that road. But they are routable. Maybe an interesting feature, if you dont use any typ file and still want two routable lines on top of each other (like opposite cycleways?).
but what about extended types? I do suspect that garmin uses some extended types for their activity routing... Would it be possible to test how they behave - I assume right now mkgmap simply cannot write them routable...
All active route lines have the same routable type numbers , ie 0x2 ,0x5 . Although the line plotting is done in the same RGN area as 'normal' highways, active routes can be distinguished from the rest by pointers to a unique block in a typ file which also contains its activity, hiking,cycling etc -- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804964.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
btw 0x2d and 0x2e are routable as well. -- View this message in context: http://gis.19327.n5.nabble.com/strange-routing-tp5804835p5804965.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
I've no problem with those warnings Gerd, in my styles I have a few more of these. Also a warning that 0x15 might be non-routable line is ok, so it's up to you which warning you prefer.
yes, MapSource uses type 0x15 for routing. When I add it to the list of routable types --check-styles complains about other lines in your style: Warning: routable type 0x15 is used for non-routable line with level 0. This may break routing. Style file lines, line 345
So I am not sure if I should simply commit that. Does anybody else use type 0x15 for lines?
Gerd
participants (6)
-
Felix Hartmann -
Gerd Petermann -
GerdP
-
Lambertus -
Minko -
nwillink