Regression in splitter r277 compared to r270? Area too small ...
Hi, with splitter r270, all of my maps are free of SEVERE errors. But with splitter r277, I get a lot of: 2013/01/14 06:33:46 SEVERE (MapSplitter): 71200148.osm.pbf: Area too small to split at http://www.openstreetmap.org/?mlat=54.45305&mlon=13.44031&zoom=17 (reduce the density of points, length of lines, etc.) Where there any changes which let expect such a result? mkgmap, the planet file and the style are the same, I really only exchanged splitter from r270 to r277 and reverted it afterwards to verify it is really this update. Splitter version unknown compiled 2013-01-11T10:25:35+0000 cache= description=TK-DACH-Tile geonames-file=osmmaps/scripts/cities/DACH.txt keep-complete=true mapid=71200001 max-areas=1024 max-nodes=1000000 max-threads=8 (auto) mixed=false no-trim=false output=pbf output-dir=build/DACH/tiles overlap=0 polygon-file= precomp-sea= problem-file= problem-report= resolution=13 split-file= status-freq=120 stop-after=dist write-kml= Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Thorsten Kukuk wrote
Where there any changes which let expect such a result?
Well, most changes were optimizations, but some also change the output, r277 compared to r270: - keep-complete is the default - Changes suggested by WanMil to reduce the number of problem candidates - different handling of polygon files Do you see different areas.list files or just different output files? Please send me the logs of both versions for detailed analyses. Ciao, Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Regression-in-splitter-r277-compared-to-r270-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi, On Mon, Jan 14, GerdP wrote:
Thorsten Kukuk wrote
Where there any changes which let expect such a result?
Well, most changes were optimizations, but some also change the output, r277 compared to r270: - keep-complete is the default - Changes suggested by WanMil to reduce the number of problem candidates - different handling of polygon files
Do you see different areas.list files or just different output files? Please send me the logs of both versions for detailed analyses.
No difference in areas.list or densities-out.txt, but the output files are much smaller. I will send the output later. Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Hi Thorsten, I think we can skip the logs, the different file sizes are intended. (as long as you see smaller files with r277) See also http://gis.19327.n5.nabble.com/splitter-relations-to-be-checked-with-keep-co... regarding the changes suggested by WanMil. I don't know why this results in more mkgmap errors. Please send me a link to the two different 71200148.osm.pbf files and let me know the parms for mkgmap and the version of mkgmap you are using. Gerd
Date: Mon, 14 Jan 2013 14:25:32 +0100 From: kukuk@suse.de To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Regression in splitter r277 compared to r270? Area too small ...
Hi,
On Mon, Jan 14, GerdP wrote:
Thorsten Kukuk wrote
Where there any changes which let expect such a result?
Well, most changes were optimizations, but some also change the output, r277 compared to r270: - keep-complete is the default - Changes suggested by WanMil to reduce the number of problem candidates - different handling of polygon files
Do you see different areas.list files or just different output files? Please send me the logs of both versions for detailed analyses.
No difference in areas.list or densities-out.txt, but the output files are much smaller.
I will send the output later.
Thorsten
-- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg) _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
On Mon, Jan 14, Gerd Petermann wrote:
Hi Thorsten,
I think we can skip the logs, the different file sizes are intended. (as long as you see smaller files with r277)
See also http://gis.19327.n5.nabble.com/splitter-relations-to-be-checked-with-keep-co... regarding the changes suggested by WanMil.
I don't know why this results in more mkgmap errors. Please send me a link to the two different 71200148.osm.pbf files and let me know the parms for mkgmap and the version of mkgmap you are using.
Ok, splitter is r271, mkgmap is r2443. Commandline for splitter is: java -Xmx6000M -jar /usr/share/java/splitter.jar --mapid=71200001 --max-nodes=1000000 --overlap=0 --keep-complete=true --max-areas=1024 --geonames-file=osmmaps/scripts/cities/DACH.txt --description=TK-DACH-Tile --output=pbf --output-dir=build/DACH/tiles data/osm/DACH.osm.pbf mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp -c maps.cfg --description=TK-DACH-Basemap TK_DACH.TYP The osm.pbf files: http://osm.thkukuk.de/tmp/r270-71200148.osm.pbf http://osm.thkukuk.de/tmp/r271-71200148.osm.pbf Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Hi Thorsten, Thorsten Kukuk wrote
mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp -c maps.cfg --description=TK-DACH-Basemap TK_DACH.TYP
The osm.pbf files:
http://osm.thkukuk.de/tmp/r270-71200148.osm.pbf http://osm.thkukuk.de/tmp/r271-71200148.osm.pbf
I wasn't yet able to reproduce the problem. Please, what is the content of mkgmap.cfg and maps.cfg ? I downloaded http://osm.thkukuk.de/tk-osm.tar.bz2, but can't find them. Is it mkgmap r2443 without any patches? Ciao, Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Regression-in-splitter-r277-compared-to-r270-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
On Mon, Jan 14, GerdP wrote:
Hi Thorsten,
Thorsten Kukuk wrote
mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp -c maps.cfg --description=TK-DACH-Basemap TK_DACH.TYP
The osm.pbf files:
http://osm.thkukuk.de/tmp/r270-71200148.osm.pbf http://osm.thkukuk.de/tmp/r271-71200148.osm.pbf
I wasn't yet able to reproduce the problem. Please, what is the content of mkgmap.cfg and maps.cfg ? I downloaded http://osm.thkukuk.de/tk-osm.tar.bz2, but can't find them.
mkgmap.cfg is basemap/mkgmap.DACH.cfg: family-id: 7120 product-id: 1 draw-priority: 10 mapname: 71201001 maps.cfg contains descriptions for each input file: description: DE-Rostock input-file: 71200148.osm.pbf
Is it mkgmap r2443 without any patches?
It contains only patches to build without network, but no code changes. Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Thorsten Kukuk wrote
mkgmap.cfg is basemap/mkgmap.DACH.cfg: family-id: 7120 product-id: 1 draw-priority: 10 mapname: 71201001
maps.cfg contains descriptions for each input file: description: DE-Rostock input-file: 71200148.osm.pbf
Is it mkgmap r2443 without any patches?
It contains only patches to build without network, but no code changes.
OK, that's what I thought. Are you able to reproduce the error when you replace -c maps.cfg by r271-71200148.osm.pbf ? Gerd -- View this message in context: http://gis.19327.n5.nabble.com/Regression-in-splitter-r277-compared-to-r270-... Sent from the Mkgmap Development mailing list archive at Nabble.com.
On Mon, Jan 14, GerdP wrote:
OK, that's what I thought. Are you able to reproduce the error when you replace -c maps.cfg by r271-71200148.osm.pbf ?
No, I'm not :( But I'm able to reproduce it with http://osm.thkukuk.de/tmp/r281-71200148.osm.pbf and this script: #!/bin/bash MKGMAP_LOG=../../../tk-osm/scripts/logging.properties export MKGMAP_LOG bash -x mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp --description="DE-Rostock" --input-file=r281-71200148.osm.pbf --description=TK-DACH-Basemap TK_DACH.TYP Maybe I made a mistake when uploading r271-*, sorry. Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Hi Thorsten, ok, now I see the error messages. Interesting is that they disappear when I replace --generate-sea=extend-sea-sectors by --precomp-sea=..\osm\sea --generate-sea (..\osm\sea contains the precompiled sea data from WanMil) So, it seems that the change in splitter influences coastline data. I'll have a closer look now. Gerd
Date: Mon, 14 Jan 2013 17:40:20 +0100 From: kukuk@suse.de To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Regression in splitter r277 compared to r270? Area too small ...
On Mon, Jan 14, GerdP wrote:
OK, that's what I thought. Are you able to reproduce the error when you replace -c maps.cfg by r271-71200148.osm.pbf ?
No, I'm not :( But I'm able to reproduce it with http://osm.thkukuk.de/tmp/r281-71200148.osm.pbf
and this script: #!/bin/bash MKGMAP_LOG=../../../tk-osm/scripts/logging.properties export MKGMAP_LOG bash -x mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp --description="DE-Rostock" --input-file=r281-71200148.osm.pbf --description=TK-DACH-Basemap TK_DACH.TYP
Maybe I made a mistake when uploading r271-*, sorry.
Thorsten
-- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg) _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Thorsten, a bit off topic: When I run mkgmap with -ea I see an error caused by this statement in basemap\style\points: natural=peak & (name=* | ele=* ) {name '${name|} (${ele}m)' | '${ele}m' | '${name}' } natural=peak & (name=* | ele=* ) {name '${name} (${ele}m)' | '${ele}m' | '${name}' } (removed a vertical bar) Gerd
Date: Mon, 14 Jan 2013 17:40:20 +0100 From: kukuk@suse.de To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Regression in splitter r277 compared to r270? Area too small ...
On Mon, Jan 14, GerdP wrote:
OK, that's what I thought. Are you able to reproduce the error when you replace -c maps.cfg by r271-71200148.osm.pbf ?
No, I'm not :( But I'm able to reproduce it with http://osm.thkukuk.de/tmp/r281-71200148.osm.pbf
and this script: #!/bin/bash MKGMAP_LOG=../../../tk-osm/scripts/logging.properties export MKGMAP_LOG bash -x mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp --description="DE-Rostock" --input-file=r281-71200148.osm.pbf --description=TK-DACH-Basemap TK_DACH.TYP
Maybe I made a mistake when uploading r271-*, sorry.
Thorsten
-- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg) _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Gerd, On Tue, Jan 15, Gerd Petermann wrote:
a bit off topic: When I run mkgmap with -ea I see an error caused by this statement in basemap\style\points: natural=peak & (name=* | ele=* ) {name '${name|} (${ele}m)' | '${ele}m' | '${name}' }
natural=peak & (name=* | ele=* ) {name '${name} (${ele}m)' | '${ele}m' | '${name}' } (removed a vertical bar)
Thanks, fixed in my style. Thorsten -- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg)
Hi all, it's hard to say why mkgmap produces more error messages with the output of r281, but I think the reason is that the SeaGenerater functions have more luck with the data from r270. Both r270 and r281 give incomplete coastline data to mkgmap, but r270 contains a few more ways (because they belong to some relations), and therefore the SeaGenerator creates different areas. I think the precomp-sea option is the only good solution for this issue. Gerd
Date: Mon, 14 Jan 2013 17:40:20 +0100 From: kukuk@suse.de To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Regression in splitter r277 compared to r270? Area too small ...
On Mon, Jan 14, GerdP wrote:
OK, that's what I thought. Are you able to reproduce the error when you replace -c maps.cfg by r271-71200148.osm.pbf ?
No, I'm not :( But I'm able to reproduce it with http://osm.thkukuk.de/tmp/r281-71200148.osm.pbf
and this script: #!/bin/bash MKGMAP_LOG=../../../tk-osm/scripts/logging.properties export MKGMAP_LOG bash -x mkgmap --style-file=style --family-name=TK-DACH-Basemap --country-name=DACH --country-abbr=DACH --area-name=DACH --latin1 --license-file=TK-DACH-Basemap_license.txt '--copyright-message=OpenStreetMap contributors, ODbL. See: http://www.openstreetmap.org/copyright. TK-DACH-Basemap based on data from 2013-01-13.' --series-name=TK-DACH-Basemap --bounds=bounds --location-autofill=bounds,is_in,nearest --process-destination --add-pois-to-areas '--pois-to-areas-placement=entrance=main;entrance=yes;building=entrance;barrier=entrance' --reduce-point-density=4 --reduce-point-density-polygon=8 --min-size-polygon=8 --make-opposite-cycleways --adjust-turn-headings --name-tag-list=name,place_name --link-pois-to-ways --route --net --generate-sea=extend-sea-sectors --index -c mkgmap.cfg --gmapsupp --description="DE-Rostock" --input-file=r281-71200148.osm.pbf --description=TK-DACH-Basemap TK_DACH.TYP
Maybe I made a mistake when uploading r271-*, sorry.
Thorsten
-- Thorsten Kukuk, Project Manager/Release Manager SLES SUSE LINUX Products GmbH, Maxfeldstr. 5, D-90409 Nuernberg GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer, HRB 16746 (AG Nürnberg) _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://lists.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
participants (3)
-
Gerd Petermann -
GerdP
-
Thorsten Kukuk