Hi, I would like to split Germany data into 5 tiles. I have added contour lines and sea area to OSM data but this probably is not important. My file has about 200M of nodes. If I split with max-nodes 50M, then I get about 50 tiles, which is wrong. If I split with num-tiles=5 then bounding polygon is ignored and I get tiles with boxes like: 63240002: 2176512,-935424 to 2796288,355584 # : 46.702881,-20.072021 to 60.001831,7.630005 This is useless for me, since I need true bounding boxes for tiles. I have used splitter r412. Any solution for this problem? Temporary I'm going to clip data with osmosis. -- Best regards, Andrzej
El 11/10/14 22:54, Andrzej Popowski escribió:
Hi,
I would like to split Germany data into 5 tiles. I have added contour lines and sea area to OSM data but this probably is not important.
My file has about 200M of nodes. If I split with max-nodes 50M, then I get about 50 tiles, which is wrong.
If I split with num-tiles=5 then bounding polygon is ignored and I get tiles with boxes like:
63240002: 2176512,-935424 to 2796288,355584 # : 46.702881,-20.072021 to 60.001831,7.630005
This is useless for me, since I need true bounding boxes for tiles.
I have used splitter r412. Any solution for this problem? Temporary I'm going to clip data with osmosis.
Have you tried increasing max-nodes value?. If 50M gives you 50 tiles, then 500M should will give around 5 tiles.
Hi Carlos, this was my first idea, I thought that I had missed a "0'. With 500M max-nodes I get 46 tiles. I tried max-tiles=4 and splitter failed, no solution. -- Best regards, Andrzej
Hi Andrzej, now that the cycling season is over (for me) I'll continue the development of splitter, hope I can fix this problem. Gerd
Date: Sun, 12 Oct 2014 11:24:39 +0200 From: popej@poczta.onet.pl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Can't split into big tiles
Hi Carlos,
this was my first idea, I thought that I had missed a "0'. With 500M max-nodes I get 46 tiles.
I tried max-tiles=4 and splitter failed, no solution.
-- Best regards, Andrzej _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Gerd, maybe it was my fault, I can't repeat the problem with split into 50M nodes. Now I get 5 tiles which is ok. Probably my previous input data contained duplicated ID numbers. As for split with --num-tiles=5, I get info: "Polygons are ignored because parameter --num-tiles is used" I guess it is about --polygon-file parameter. If bounding poly is not used, then tiles can be bigger. I think you could convert polygon into bounding box to limit area. Without bounding poly, calculated aspect ratio is of no use for tiles at boundary. I'm attaching logs for splitting, there was no densities-out.txt. -- Best regards, Andrzej
Hi Andrzej, okay, now I see what happens. You feed a map containing a big part of Europe to splitter, and you want to cut out a rectangular area covering Germany, and this part should be devided into 5 tiles using the --num-tiles=5 option. Correct? So, this is a special case as your bounding polygon is already a rectangle. I'll see if I can add code to handle this case. Gerd Date: Tue, 14 Oct 2014 16:12:44 +0200 From: popej@poczta.onet.pl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Can't split into big tiles Hi Gerd, maybe it was my fault, I can't repeat the problem with split into 50M nodes. Now I get 5 tiles which is ok. Probably my previous input data contained duplicated ID numbers. As for split with --num-tiles=5, I get info: "Polygons are ignored because parameter --num-tiles is used" I guess it is about --polygon-file parameter. If bounding poly is not used, then tiles can be bigger. I think you could convert polygon into bounding box to limit area. Without bounding poly, calculated aspect ratio is of no use for tiles at boundary. I'm attaching logs for splitting, there was no densities-out.txt. -- Best regards, Andrzej _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Gerd, splitter is a good tool and I'm trying to use it for other purposes than preparing data for mkgmap. In this case I was trying to compile maps for Locus. That's why I find unusual problems ;) It's low priority, I think that using some kind of bounding box or poly for --num-tiles could be usable. -- Best regards, Andrzej
Hi Andrzej, attached is a patch which implements this. Please try if it works for you. Gerd useBBoxOfPolygons-v1.patch <http://gis.19327.n5.nabble.com/file/n5822800/useBBoxOfPolygons-v1.patch> popej wrote
Hi Gerd,
splitter is a good tool and I'm trying to use it for other purposes than preparing data for mkgmap. In this case I was trying to compile maps for Locus. That's why I find unusual problems ;)
It's low priority, I think that using some kind of bounding box or poly for --num-tiles could be usable.
-- Best regards, Andrzej _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
-- View this message in context: http://gis.19327.n5.nabble.com/Can-t-split-into-big-tiles-tp5820060p5822800.... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi Gerd, I have applied your patch against r414, I hope correctly. I can see this message: Exact map coverage read from input file(s) is (46.7038893699646,-20.071227550506592) to (60.539002418518066,28.109700679779053) Exact map coverage after applying bounding box of polygon-file is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) Exact map coverage is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) But calculated areas are exactly the same as in original version: Area 63240012 covers (46.669921875,-20.0830078125) to (60.029296875,7.646484375) and contains 41151962 nodes (84 %) I have attached logs from both versions. -- Best regards, Andrzej
Hi Andrzej, yes, seems my patch is not working. It changes the bbox that is used to display the vlaues, but not the one that is used for the tile calculation :-( I'll look at it tomorrow. Gerd Date: Mon, 3 Nov 2014 21:09:26 +0100 From: popej@poczta.onet.pl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Can't split into big tiles Hi Gerd, I have applied your patch against r414, I hope correctly. I can see this message: Exact map coverage read from input file(s) is (46.7038893699646,-20.071227550506592) to (60.539002418518066,28.109700679779053) Exact map coverage after applying bounding box of polygon-file is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) Exact map coverage is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) But calculated areas are exactly the same as in original version: Area 63240012 covers (46.669921875,-20.0830078125) to (60.029296875,7.646484375) and contains 41151962 nodes (84 %) I have attached logs from both versions. -- Best regards, Andrzej _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Hi Andrzej, sorry, don't know what I tested yesterday, here is the new patch. Gerd From: gpetermann_muenchen@hotmail.com To: mkgmap-dev@lists.mkgmap.org.uk Date: Mon, 3 Nov 2014 21:25:22 +0100 Subject: Re: [mkgmap-dev] Can't split into big tiles Hi Andrzej, yes, seems my patch is not working. It changes the bbox that is used to display the vlaues, but not the one that is used for the tile calculation :-( I'll look at it tomorrow. Gerd Date: Mon, 3 Nov 2014 21:09:26 +0100 From: popej@poczta.onet.pl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Can't split into big tiles Hi Gerd, I have applied your patch against r414, I hope correctly. I can see this message: Exact map coverage read from input file(s) is (46.7038893699646,-20.071227550506592) to (60.539002418518066,28.109700679779053) Exact map coverage after applying bounding box of polygon-file is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) Exact map coverage is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) But calculated areas are exactly the same as in original version: Area 63240012 covers (46.669921875,-20.0830078125) to (60.029296875,7.646484375) and contains 41151962 nodes (84 %) I have attached logs from both versions. -- Best regards, Andrzej _______________________________________________ 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 all, I just noticed that I did not yet commit patch useBBoxOfPolygons-v2.patch Did anybody try it? Gerd GerdP wrote
Hi Andrzej,
sorry, don't know what I tested yesterday, here is the new patch.
Gerd
From:
gpetermann_muenchen@
To:
mkgmap-dev@.org
Date: Mon, 3 Nov 2014 21:25:22 +0100 Subject: Re: [mkgmap-dev] Can't split into big tiles
Hi Andrzej,
yes, seems my patch is not working. It changes the bbox that is used to display the vlaues, but not the one that is used for the tile calculation :-( I'll look at it tomorrow.
Gerd
Date: Mon, 3 Nov 2014 21:09:26 +0100 From:
popej@.onet
To:
mkgmap-dev@.org
Subject: Re: [mkgmap-dev] Can't split into big tiles
Hi Gerd,
I have applied your patch against r414, I hope correctly. I can see this message: Exact map coverage read from input file(s) is (46.7038893699646,-20.071227550506592) to (60.539002418518066,28.109700679779053) Exact map coverage after applying bounding box of polygon-file is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126) Exact map coverage is (47.26541519165039,6.1527299880981445) to (55.147762298583984,15.05077600479126)
But calculated areas are exactly the same as in original version: Area 63240012 covers (46.669921875,-20.0830078125) to (60.029296875,7.646484375) and contains 41151962 nodes (84 %)
I have attached logs from both versions.
-- Best regards, Andrzej
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev _______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
useBBoxOfPolygons-v2.patch (6K) <http://gis.19327.n5.nabble.com/attachment/5822978/0/useBBoxOfPolygons-v2.pat...;
-- View this message in context: http://gis.19327.n5.nabble.com/Can-t-split-into-big-tiles-tp5820060p5825911.... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi Gerd, sorry, somehow I forgot about this patch. I have tested it now and it works correctly. See attached picture with tiles without patch - big area - and with patch - correct area around bounding poly. -- Best regards, Andrzej
Hi Andrzej, okay, I've committed that patch now. Gerd popej wrote
Hi Gerd,
sorry, somehow I forgot about this patch. I have tested it now and it works correctly. See attached picture with tiles without patch - big area - and with patch - correct area around bounding poly.
-- Best regards, Andrzej
_______________________________________________ mkgmap-dev mailing list
mkgmap-dev@.org
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
bpoly2.png (9K) <http://gis.19327.n5.nabble.com/attachment/5825953/0/bpoly2.png>;
-- View this message in context: http://gis.19327.n5.nabble.com/Can-t-split-into-big-tiles-tp5820060p5825982.... Sent from the Mkgmap Development mailing list archive at Nabble.com.
Hi Andrzej, the combination of a polygon file and the --num-tiles parameter is likely to create complex cases, as splitter can only write rectangular tiles. Please provide your poly file, the densities-out.txt and the splitter.log for further analyses. Gerd
Date: Sun, 12 Oct 2014 11:24:39 +0200 From: popej@poczta.onet.pl To: mkgmap-dev@lists.mkgmap.org.uk Subject: Re: [mkgmap-dev] Can't split into big tiles
Hi Carlos,
this was my first idea, I thought that I had missed a "0'. With 500M max-nodes I get 46 tiles.
I tried max-tiles=4 and splitter failed, no solution.
-- Best regards, Andrzej _______________________________________________ mkgmap-dev mailing list mkgmap-dev@lists.mkgmap.org.uk http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
participants (4)
-
Andrzej Popowski -
Carlos Dávila -
Gerd Petermann -
GerdP