[mkgmap-dev] Internal Error Failed to render Multipolygon

Tue Dec 24 10:00:14 GMT 2024

Hi Ticker,

sorry for the late response. I think the main problem is the value for niceSplitShift. I see much better results when I change that to 0 so that partitions are divided into equal halves.
I've also learned that the reported internal error is more often caused by invalid multipolygons where a ring with role inner is outside of the bbox of the outer ring.
Try e.g. https://files.mkgmap.org.uk/detail/566

I've attached a patch to address these two problems. One other problem that I noticed is in method IsInUtil.checkAllOn()
Coord pTest = p1.destOnRhumbLine(EPS_OFF, p1.bearingTo(p2));

I noticed cases where this code calculates a point which is not on the line between p1 and p2, e.g. if the distance between these two points is smaller than EPS_OFF.
I introduced this line with r4499 which refers to your patch isPointInShape_v3.patch. No idea why, I can't find the changes in that patch. Maybe I look at wrong file.

The partitioning is really a performance boost with very complex MPs like that for lake Huron (relation 1205151). I see 10 secs instead of +130 secs with partitioning.
On the other hand it is more likely to hit edge cases where small fragments of touching inner rings are reported as possibly wrong nested inners.

Gerd

________________________________________
Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Ticker Berkin <rwb-mkgmap at jagit.co.uk>
Gesendet: Samstag, 7. September 2024 20:32
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] Internal Error Failed to render Multipolygon

Hi Gerd

My version is the one before:
  <relation id='1754729' timestamp='2024-09-01T17:50:47Z' uid='20024786'
user='Cmapuk-noxa6bl4' visible='true' version='1231' changeset='156066008'>

I was thinking something similar, but, instead of changing niceSplitShift,
changing the fixed depth limit of 10, to be based on the size of the area, so
that it stops a couple splits before the split would fail. In this example I
found changing it to
   if (depth >= 9 || ...
it worked and took the same time as the failing value of 10.

Regards
Ticker

On Sat, 2024-09-07 at 04:33 +0000, Gerd Petermann wrote:
> Hi Ticker,
>
> I cannot reproduce the "Failed to render ... " message.
> Maybe I am looking at a different version of the MP. Mine is v1232.
> I see
> INFO: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation  f:\osm\lake.osm:
> Processing multipolygon http://www.openstreetmap.org/relation/1754729
> WARN: uk.me.parabola.imgfmt.app.Area  f:\osm\lake.osm: Area.split 2852864
> 1400832 2854912 1402880 res 11 can't 2 1
> WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation  f:\osm\lake.osm:
> Multipolygon http://www.openstreetmap.org/relation/1754729 [ele=4.84,
> name=Ладожское озеро, natural=water, type=multipolygon, water=lake] contains
> errors.
> WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation  f:\osm\lake.osm:
> Polygon 4611686018427400531(4P)(28079958[166P]) carries role inner but is not
> inside any other polygon. Potentially it does not belong to this multipolygon.
> WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation  f:\osm\lake.osm:
> Multipolygon http://www.openstreetmap.org/relation/1754729 [ele=4.84,
> name=Ладожское озеро, natural=water, type=multipolygon, water=lake] contains
> errors.
> WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation  f:\osm\lake.osm:
> Polygon 4611686018427403162(4P)(259078070[118P]) carries role inner but is not
> inside any other polygon. Potentially it does not belong to this multipolygon.
>
> These messages go away when I change line 1985 in  MultiPolygonRelation from
> final int niceSplitShift = 11;
> to
> final int niceSplitShift = 10;
>
> I can't remember why the value 11 was used.
> Does that help?
>
> Gerd
>
> ________________________________________
> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Gerd
> Petermann <gpetermann_muenchen at hotmail.com>
> Gesendet: Freitag, 6. September 2024 18:04
> An: Development list for mkgmap
> Betreff: Re: [mkgmap-dev] Internal Error Failed to render Multipolygon
>
> Hi all,
>
> forget my previous post. The way https://www.osm.org/way/339720813 is OK.
>
> Gerd
>
> ________________________________________
> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Gerd
> Petermann <gpetermann_muenchen at hotmail.com>
> Gesendet: Freitag, 6. September 2024 18:02
> An: Development list for mkgmap
> Betreff: Re: [mkgmap-dev] Internal Error Failed to render Multipolygon
>
> Hi Ticker,
>
> Maybe the JOSM test doesn't work correctly. I think e.g. this way
> https://www.osm.org/way/339720813
> looks wrong. It is an inner inside an inner area and AFAIR this is not
> allowed.
>
> Gerd
>
> ________________________________________
> Von: mkgmap-dev <mkgmap-dev-bounces at lists.mkgmap.org.uk> im Auftrag von Ticker
> Berkin <rwb-mkgmap at jagit.co.uk>
> Gesendet: Donnerstag, 5. September 2024 12:01
> An: Development list for mkgmap
> Betreff: Re: [mkgmap-dev] Internal Error Failed to render Multipolygon
>
> Hi Gerd
>
> I can make the Area.split and the other error go away by setting
> noPartitioning() true in MultiPolygonRelation.java. The total time goes up
> from
> 17 to 39 seconds.
>
> I can't see any problem in the logic when Area.split is unable to divide the
> largest polygon in two.
>
> The relation might be faulty with intersections of the various components and
> partitioning with ShapeSplitter can expose these problems. Does the JOSM
> verification detect this.
>
> Otherwise this is beyond me.
>
> Regards
> Ticker
>
> On Tue, 2024-09-03 at 17:09 +0100, Ticker Berkin wrote:
> > > Hi Thomas
> > >
> > > I've just loaded the relation 1754729 with JOSM.
> > > JOSM validation reports no errors, but mkgmap (direct, no splitting) gives
> > > errors:
> > >
> > > WARN: uk.me.parabola.imgfmt.app.Area  lakeLadoga.osm:
> > >  Area.split 2852864 1400832 2854912 1402880 res 11 can't 2 1
> > >
> > > WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation
> > > lakeLadoga.osm:
> > >  Multipolygon http://www.openstreetmap.org/relation/1754729 [ele=4.84,
> > > int_name=Ladoga lake, name:en=Lake Ladoga, name=Ладожское озеро,
> > > natural=water,
> > > type=multipolygon, water=lake] contains errors.
> > >
> > > WARN: uk.me.parabola.mkgmap.reader.osm.MultiPolygonRelation
> > > lakeLadoga.osm:
> > >  Polygon 4611686018427403156(4P)(259078070[118P]) carries role inner but
> > > is
> > > not
> > > inside any other polygon. Potentially it does not belong to this
> > > multipolygon
> > >
> > > The area.split WARN happens when there are too many object
> > > (points/lines/polygons) in 1 resolution unit.
> > >
> > > The other error is as described.
> > >
> > > I notice that this relation was edited a couple of days ago so maybe
> > > someone
> > > is
> > > fixing or creating problems.
> > >
> > > I'll try again in a few days and investigate more deeply.
> > >
> > > Regards
> > > Ticker
> > >
> > > On Mon, 2024-09-02 at 19:44 +0200, Thomas Morgenstern wrote:
> > > > >  Hi Ticker, I found : osmium-tool and splitter works proper. I can the
> > > > > splitted <12345678>.osm.pbf open and render in JOSM. JOSM shows the
> > > > > part
> > > > > of
> > > > > the multipolygon 1754729, which is inside the boundary of the
> > > > > polygonfile
> > > > > for
> > > > > splitter. I assume, mkgmap has the problem. But sorry, I am not a good
> > > > > programmer and can not do deeper investigation or find a solution. If
> > > > > you
> > > > > interested, I can give you my polygon and the <maptile-
> > > > > number>.osm.pbf.
> > > > >
> > > > >  Thomas
> > > > >
> > > > >
> > > > > Am 02.09.2024 um 15:04 schrieb Ticker Berkin:
> > > > >
> > > > >
> > > > > > >
> > > > > > > Hi Thomas
> > > > > > >
> > > > > > > I get similar problems just using
> > > > > > >
> > > > > > >  https://download.geofabrik.de/europe/britain-and-ireland.html
> > > > > > >
> > > > > > > directly. The relations that cause the problem are not within the
> > > > > > > above
> > > > > > > combined
> > > > > > > territories but excluding them would required a much more
> > > > > > > complicated
> > > > > > > shape
> > > > > > > to
> > > > > > > define the download area.
> > > > > > >
> > > > > > > Ticker
> > > > > > >
> > > > > > > On Mon, 2024-09-02 at 12:42 +0200, Thomas Morgenstern wrote:
> > > > > > >
> > > > > > > > >
> > > > > > > > >  Hi Ticker, i made further investigation. You are right.: the
> > > > > > > > > MultiPolygonRelation 1754729 ('Ladoga sea') was cut by ectract
> > > > > > > > > Scandinavia
> > > > > > > > > from europe-latest.osm.pbf using osmium-tool .
> > > > > > > > >  I am not sure, but means this, that osmium can not write a
> > > > > > > > > proper
> > > > > > > > > ectract,
> > > > > > > > > if such Relation must bee split ?
> > > > > > > > >  This means also, osmium has a bug ? I must check my osmium
> > > > > > > > > version...
> > > > > > > > >  regards Thomas
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Am 02.09.2024 um 10:30 schrieb Ticker Berkin:
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Hi Thomas
> > > > > > > > > > >
> > > > > > > > > > > It isn't related to AngleChecker.
> > > > > > > > > > >
> > > > > > > > > > > I notice that both these relations have been changed in
> > > > > > > > > > > the
> > > > > > > > > > > last
> > > > > > > > > > > couple
> > > > > > > > > > > of
> > > > > > > > > > > days
> > > > > > > > > > > so maybe they were faulty.
> > > > > > > > > > >
> > > > > > > > > > > The large one is on the boarder with Scandinavia and it is
> > > > > > > > > > > likely that
> > > > > > > > > > > the
> > > > > > > > > > > map
> > > > > > > > > > > extraction download only had part of it and this gives
> > > > > > > > > > > rise
> > > > > > > > > > > to this
> > > > > > > > > > > type
> > > > > > > > > > > of
> > > > > > > > > > > error.
> > > > > > > > > > >
> > > > > > > > > > > Regards
> > > > > > > > > > > Ticker
> > > > > > > > > > >
> > > > > > > > > > > On Mon, 2024-09-02 at 09:43 +0200, Thomas Morgenstern
> > > > > > > > > > > wrote:
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > > > >  Hello , I use mkgmap-4921 with patch from Ticker
> > > > > > > > > > > > > Berkin
> > > > > > > > > > > > > -
> > > > > > > > > > > > > Angelchecker-V6.
> > > > > > > > > > > > > With that i get in Scandinavia following error :
> > > > > > > > > > > > > Internal Error
> > > > > > > > > > > > > Failed
> > > > > > > > > > > > > to
> > > > > > > > > > > > > render MultiPolygonRelation 1754729 and 304507. Rel.
> > > > > > > > > > > > > 1754729 is
> > > > > > > > > > > > > 'Ladoga'-
> > > > > > > > > > > > > sea
> > > > > > > > > > > > > with 5 226 Elements. Rel.  304507 is
> > > > > > > > > > > > > Multipolygon=water,
> > > > > > > > > > > > > 171
> > > > > > > > > > > > > Elements.
> > > > > > > > > > > > >  I assume the error ist not specific to the
> > > > > > > > > > > > > patch...and
> > > > > > > > > > > > > more
> > > > > > > > > > > > > common..
> > > > > > > > > > > > >  How to avoid this ?
> > > > > > > > > > > > >
> > > > > > > > > > > > >  regards Thomas
> > > > > > > > > > > > >
> > > > > > > > > > > > > _______________________________________________
> > > > > > > > > > > > > mkgmap-dev mailing list
> > > > > > > > > > > > > mkgmap-dev at lists.mkgmap.org.uk
> > > > > > > > > > > > > https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> > > > > > > > > > > > >
> > > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > _______________________________________________
> > > > > > > > > > > mkgmap-dev mailing list
> > > > > > > > > > > mkgmap-dev at lists.mkgmap.org.uk
> > > > > > > > > > > https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > >
> > > > >
> > >
> > >
> > > _______________________________________________
> > > mkgmap-dev mailing list
> > > mkgmap-dev at lists.mkgmap.org.uk
> > > https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
> _______________________________________________
> mkgmap-dev mailing list
> mkgmap-dev at lists.mkgmap.org.uk
> https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
mkgmap-dev at lists.mkgmap.org.uk
https://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
-------------- next part --------------
A non-text attachment was scrubbed...
Name: failed-render.patch
Type: application/octet-stream
Size: 2418 bytes
Desc: failed-render.patch
URL: <https://www.mkgmap.org.uk/pipermail/mkgmap-dev/attachments/20241224/f31856df/attachment.obj>