Hi all, I think r3466 is now stable enough for a first public test. It is fully compatible with trunk r3449. Improvements compared to trunk: 1) Number nodes are now supported. A number node is a special node, house number intervals start and end with a number node. In trunk, only route nodes (road junctions and end points) are used for this, in the branch we can add many more number nodes. The advantage is that we can create more (and smaller) intervals when needed. 2) The number nodes are used to support "random" house numbers (numbers with no particular order). The program simulates an address search for any of the house numbers found in OSM data. If a number is found in multiple places, the corresponding intervals are split so that wrong overlaps are removed. This should lead to good results for correct OSM data and better results for wrong data like duplicate numbers with large distances and other typical copy+paste errors. 3) An additional step is performed to improve the results of the address search. As a result, an address search should point you to a point on the road that is not more than 40m away from the closest point. The trunk version doesn't care about this and results maybe > 200m away. Performance: The img size increases by a a few KB, compile time is a bit higher, routing should work as with trunk. Use logging property uk.me.parabola.mkgmap.osmstyle.housenumber.level=INFO to get details. Todo: Support for addr:place , code review and unit tests Any feedback is welcome. Gerd
Hi Gerd, Thank's a lot ! The compilation and some routing tests are OK. Test for address searching coming soon... Steph Le 18/02/2015 08:39, Gerd Petermann a écrit :
Hi all,
I think r3466 is now stable enough for a first public test. It is fully compatible with trunk r3449.
Improvements compared to trunk: 1) Number nodes are now supported. A number node is a special node, house number intervals start and end with a number node. In trunk, only route nodes (road junctions and end points) are used for this, in the branch we can add many more number nodes. The advantage is that we can create more (and smaller) intervals when needed.
2) The number nodes are used to support "random" house numbers (numbers with no particular order). The program simulates an address search for any of the house numbers found in OSM data. If a number is found in multiple places, the corresponding intervals are split so that wrong overlaps are removed. This should lead to good results for correct OSM data and better results for wrong data like duplicate numbers with large distances and other typical copy+paste errors.
3) An additional step is performed to improve the results of the address search. As a result, an address search should point you to a point on the road that is not more than 40m away from the closest point. The trunk version doesn't care about this and results maybe > 200m away.
Performance: The img size increases by a a few KB, compile time is a bit higher, routing should work as with trunk. Use logging property uk.me.parabola.mkgmap.osmstyle.housenumber.level=INFO to get details.
Todo: Support for addr:place , code review and unit tests
Any feedback is welcome.
Gerd
Hi Gerd, Thank's a lot ! The compilation and some routing tests are OK. Tests for address searching coming soon... Steph Le 18/02/2015 08:39, Gerd Petermann a écrit :
Hi all,
I think r3466 is now stable enough for a first public test. It is fully compatible with trunk r3449.
Improvements compared to trunk: 1) Number nodes are now supported. A number node is a special node, house number intervals start and end with a number node. In trunk, only route nodes (road junctions and end points) are used for this, in the branch we can add many more number nodes. The advantage is that we can create more (and smaller) intervals when needed.
2) The number nodes are used to support "random" house numbers (numbers with no particular order). The program simulates an address search for any of the house numbers found in OSM data. If a number is found in multiple places, the corresponding intervals are split so that wrong overlaps are removed. This should lead to good results for correct OSM data and better results for wrong data like duplicate numbers with large distances and other typical copy+paste errors.
3) An additional step is performed to improve the results of the address search. As a result, an address search should point you to a point on the road that is not more than 40m away from the closest point. The trunk version doesn't care about this and results maybe > 200m away.
Performance: The img size increases by a a few KB, compile time is a bit higher, routing should work as with trunk. Use logging property uk.me.parabola.mkgmap.osmstyle.housenumber.level=INFO to get details.
Todo: Support for addr:place , code review and unit tests
Any feedback is welcome.
Gerd
Hi, On Wed, Feb 18, Gerd Petermann wrote:
Hi all,
I think r3466 is now stable enough for a first public test. It is fully compatible with trunk r3449.
For, it looks like it's ending in an endless loop. The input file: http://osm.thkukuk.de/tmp/71200082.osm.pbf The last 2000 lines of the log file: http://osm.thkukuk.de/tmp/mkgmap.log It's my basemap style. But the default style shows the same problem. Thorsten -- Thorsten Kukuk, Senior Architect SLES & Common Code Base SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendörffer, Jane Smithard, Jennifer Guild, Dilip Upmanyu, Graham Norton, HRB 21284 (AG Nürnberg)
Am Donnerstag, 19. Februar 2015, 11:02:04 schrieb Thorsten Kukuk:
For, it looks like it's ending in an endless loop. Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull Bernd
Am Donnerstag, 19. Februar 2015, 12:14:59 schrieb Bernd Weigelt:
Am Donnerstag, 19. Februar 2015, 11:02:04 schrieb Thorsten Kukuk:
For, it looks like it's ending in an endless loop.
Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull
Bernd
Made another test with my sri lankian extract, has one only tile, the map is build was successful. Maybe something is wrong while merging the images? Bernd
Hi Bernd, logging option: uk.me.parabola.mkgmap.osmstyle.housenumber.level=DEBUG I never tested with --gmapsupp, maybe one of the readers fails to read the new number nodes. Will look at this later. Gerd Bernd Weigelt wrote
Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull
Bernd
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/housenumber2-branch-tp5834070p5834236.html Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi, Error compiling Germany with r3466 : java.lang.AssertionError: internal error id=24813533, Süderende (n19),N,0,0,E,192,156 [] [192(19), 156(20)] at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.verify(ExtNumbers.java:318) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.getNumbers(ExtNumbers.java:100) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.dupNode(ExtNumbers.java:617) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.tryAddNumberNode(ExtNumbers.java:368) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.checkChainPlausibility(ExtNumbers.java:892) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberRoad.checkIntervals(HousenumberRoad.java:150) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.buildNumberIntervals(HousenumberGenerator.java:844) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.assignHouseNumbersToRoads(HousenumberGenerator.java:660) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.matchCluster(HousenumberGenerator.java:546) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.generate(HousenumberGenerator.java:493) at uk.me.parabola.mkgmap.osmstyle.StyledConverter.end(StyledConverter.java:599) at uk.me.parabola.mkgmap.reader.osm.ElementSaver.convert(ElementSaver.java:250) at uk.me.parabola.mkgmap.reader.osm.bin.OsmBinMapDataSource.load(OsmBinMapDataSource.java:67) at uk.me.parabola.mkgmap.reader.osm.OsmMapDataSource.load(OsmMapDataSource.java:130) at uk.me.parabola.mkgmap.main.MapMaker.loadFromFile(MapMaker.java:167) at uk.me.parabola.mkgmap.main.MapMaker.makeMap(MapMaker.java:63) at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:253) at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:249) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:745) Need more ? Steph Le 19/02/2015 09:42, GerdP a écrit :
Hi Bernd,
logging option: uk.me.parabola.mkgmap.osmstyle.housenumber.level=DEBUG
I never tested with --gmapsupp, maybe one of the readers fails to read the new number nodes. Will look at this later.
Gerd
Bernd Weigelt wrote
Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull
Bernd
Hi Steph, I can reproduce the problem with a download around the way 24813533. It also happens with r3467. Looking at it now. Gerd
Date: Thu, 19 Feb 2015 10:26:34 -0300 From: st3ph.martin@laposte.net To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] housenumber2 branch
Hi,
Error compiling Germany with r3466 :
java.lang.AssertionError: internal error id=24813533, Süderende (n19),N,0,0,E,192,156 [] [192(19), 156(20)] at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.verify(ExtNumbers.java:318) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.getNumbers(ExtNumbers.java:100) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.dupNode(ExtNumbers.java:617) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.tryAddNumberNode(ExtNumbers.java:368) at uk.me.parabola.mkgmap.osmstyle.housenumber.ExtNumbers.checkChainPlausibility(ExtNumbers.java:892) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberRoad.checkIntervals(HousenumberRoad.java:150) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.buildNumberIntervals(HousenumberGenerator.java:844) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.assignHouseNumbersToRoads(HousenumberGenerator.java:660) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.matchCluster(HousenumberGenerator.java:546) at uk.me.parabola.mkgmap.osmstyle.housenumber.HousenumberGenerator.generate(HousenumberGenerator.java:493) at uk.me.parabola.mkgmap.osmstyle.StyledConverter.end(StyledConverter.java:599) at uk.me.parabola.mkgmap.reader.osm.ElementSaver.convert(ElementSaver.java:250) at uk.me.parabola.mkgmap.reader.osm.bin.OsmBinMapDataSource.load(OsmBinMapDataSource.java:67) at uk.me.parabola.mkgmap.reader.osm.OsmMapDataSource.load(OsmMapDataSource.java:130) at uk.me.parabola.mkgmap.main.MapMaker.loadFromFile(MapMaker.java:167) at uk.me.parabola.mkgmap.main.MapMaker.makeMap(MapMaker.java:63) at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:253) at uk.me.parabola.mkgmap.main.Main$1.call(Main.java:249) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang.Thread.run(Thread.java:745)
Need more ?
Steph
Le 19/02/2015 09:42, GerdP a écrit :
Hi Bernd,
logging option: uk.me.parabola.mkgmap.osmstyle.housenumber.level=DEBUG
I never tested with --gmapsupp, maybe one of the readers fails to read the new number nodes. Will look at this later.
Gerd
Bernd Weigelt wrote
Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull
Bernd
mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Bernd, sorry, option is uk.me.parabola.mkgmap.osmstyle.housenumber.level=FINE I think you saw the same endless loop as Thorsten. I found no problem using the --gmapsupp option. Please try again with r3467 Gerd
Date: Thu, 19 Feb 2015 05:42:43 -0700 From: gpetermann_muenchen@hotmail.com To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] housenumber2 branch
Hi Bernd,
logging option: uk.me.parabola.mkgmap.osmstyle.housenumber.level=DEBUG
I never tested with --gmapsupp, maybe one of the readers fails to read the new number nodes. Will look at this later.
Gerd
Bernd Weigelt wrote
Maybe the same problem here, but it's seems to happen after the last tile. The build porcess hangs now over 15 minutes
Please, give give me a hind which logging options are usefull
Bernd
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/housenumber2-branch-tp5834070p5834236.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
Hi Thorsten, thanks for reporting. I fixed the endless loop with r3467. Gerd
Date: Thu, 19 Feb 2015 11:02:04 +0100 From: kukuk@suse.de To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] housenumber2 branch
Hi,
On Wed, Feb 18, Gerd Petermann wrote:
Hi all,
I think r3466 is now stable enough for a first public test. It is fully compatible with trunk r3449.
For, it looks like it's ending in an endless loop.
The input file: http://osm.thkukuk.de/tmp/71200082.osm.pbf
The last 2000 lines of the log file: http://osm.thkukuk.de/tmp/mkgmap.log
It's my basemap style. But the default style shows the same problem.
Thorsten
-- Thorsten Kukuk, Senior Architect SLES & Common Code Base SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nuernberg, Germany GF: Felix Imendörffer, Jane Smithard, Jennifer Guild, Dilip Upmanyu, Graham Norton, HRB 21284 (AG Nürnberg) _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
participants (5)
-
Bernd Weigelt -
Gerd Petermann -
GerdP
-
Stéphane MARTIN -
Thorsten Kukuk