Land rendered as water - Please advise

P Wat

unread,

Dec 15, 2018, 7:38:31 PM12/15/18

to Osmand

Hi folks.

Some areas of land (not all) are being rendered as water. - See attached graphics.

Note that the land to the north (mainland England) is rendering OK when Isle of Wight (south section of map) is not.

You'll note the problem exists in all types of map configuration, ie World, Car, Bike, Foot and Boat.

It occurs when the scale is from 0.5 mile to 2 miles.

The same problem exists on two devices; one is running Android 7.0 and Osmand+ 3.2.7, the other is running Android 4.4.4 and Osmand Osmand+3.2.6

Has anyone else had this problem and resolved it?

Does anyone have any suggestions?

Thanks

Paul W

Screenshot_2018-12-13-19-03-25.png

Screenshot_2018-12-13-19-04-13.png

Screenshot_2018-12-13-19-04-50.png

Screenshot_2018-12-13-15-07-20.png

Screenshot_2018-12-13-19-03-01.png

Harry van der Wolf

unread,

Dec 16, 2018, 7:55:04 AM12/16/18

to osmand

This has happened multiple times in the past with multi-polygons, especially on borders (boundaries), be it country, region, county, "regierungsbezirk", "bundesland", "province", "provincie", etc.

The fact that you mention that it happens between England and Wright confirms this.

It is not necessarily an OsmAnd bug (cutting polyogons on boundaries), but can also be due to errors in the (multi)polygons itself because they are broken, interrupted, corrupt, or else.

Harry

Op zo 16 dec. 2018 om 01:38 schreef 'P Wat' via Osmand <osm...@googlegroups.com>:

--
You received this message because you are subscribed to the Google Groups "Osmand" group.
To unsubscribe from this group and stop receiving emails from it, send an email to osmand+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

P Wat

unread,

Dec 16, 2018, 4:27:47 PM12/16/18

to Osmand

Hello Harry.

Thank you for your feedback.

Does this mean that it is caused only by OSM (map) issues needing correction?
Is there anything to be done in OsmAnd by the user (ie me)?
Is there anything to be done in OsmAnd by the authors?

If (1), is it realistic for me to attempt correcting OSM?

If (2), what is required?

If (3), what is the proccedure for submitting a request?

Paul W

Harry van der Wolf

unread,

Dec 16, 2018, 5:07:00 PM12/16/18

to osmand

Op zo 16 dec. 2018 om 22:27 schreef 'P Wat' via Osmand <osm...@googlegroups.com>:

Hello Harry.
Thank you for your feedback.
Does this mean that it is caused only by OSM (map) issues needing correction?
Is there anything to be done in OsmAnd by the user (ie me)?
Is there anything to be done in OsmAnd by the authors?
If (1), is it realistic for me to attempt correcting OSM?
If (2), what is required?
If (3), what is the proccedure for submitting a request?

Paul W

1. Correcting (multi-)polygons is really work for experienced OMers. If you are one, you can do the job. However, you asking this might mean that you never did this. There can, and most of the time will be lot of relations coupled to (parts of) the polygon(s). That makes it real hard. I did it once in the south of the Netherlands. It was really a horrible job.

2. The only thing you can do is make an OsmAnd issue for it. We had the same some years ago with the Bodensee (between Germany, Switzerland and Austria). Finally the country polygons for OsmAnd were adapted by Victor as it couldn't be solved from Openstreetmap itself. The same might be necessary for this issue, but it will take quite some time. First an analysis of the OSM data needs to be done and when possible repaired. Finding the last few editors need to be done to see what they did and whether they can correct their own mistakes. And when it can't be corrected after all, it might need an OsmAnd correction.

3. See 2 as well.

Harry

P Wat

unread,

Dec 16, 2018, 5:46:55 PM12/16/18

to Osmand

Hi Harry

Thank you for your fast and useful reply.

I am definitely not an experienced OMer, far from it, up to now I've only done simple corrections and additions. It's probably best if I stay clear and leave it to experts.

So, I add a few notes on the OSM map adjacent to the problem to alert people who know what they are doing, then stand back for a few weeks to see if anything is done.

Regards

Paul W.

Martin Trautmann

unread,

Dec 17, 2018, 1:43:41 AM12/17/18

to osm...@googlegroups.com

On 18-12-16 13:54, Harry van der Wolf wrote:
> This has happened multiple times in the past with multi-polygons,
> especially on borders (boundaries), be it country, region, county,
> "regierungsbezirk", "bundesland", "province", "provincie", etc.
> The fact that you mention that it happens between England and Wright
> confirms this.
> It is not necessarily an OsmAnd bug (cutting polyogons on boundaries),
> but can also be due to errors in the (multi)polygons itself because they
> are broken, interrupted, corrupt, or else.

When I did observe this last time, the effect happened, depending on the
zoom level. If you zoomed out, water changed to land again. Thus I
believe, it's not a broken polygon, but an osmand bug, while using
polygons which do span more than the calculated area.

- Martin

signature.asc

P Wat

unread,

Dec 17, 2018, 2:42:05 AM12/17/18

to Osmand

I can confirm that the problem is not present at all zoom levels.

In my initial post I recorded "It occurs when the scale is from 0.5 mile to 2 miles."

ie outside that range the problem is not visible.

PW

Poutnik the Wanderer

unread,

Dec 17, 2018, 3:47:36 AM12/17/18

to osm...@googlegroups.com, Martin Trautmann

Good verification step is also testing the area in an independent
application, based on OSM as well.

e.g. OSMAnd versus LocusMap, using MapsForge compatible OSM based maps,
supposing both are using recent maps.

----------
Dne 17. prosince 2018 7:43:40 Martin Trautmann <tr...@gmx.de> napsal:

P Wat

unread,

Dec 20, 2018, 12:57:34 PM12/20/18

to Osmand

The comment by Martin Trautmann above describes my own experience.
Notably he states; "...the effect happened, depending on the zoom level. If you zoomed out, water changed to land again."

From which he deduces; "... I believe, it's not a broken polygon, but an osmand bug, while using polygons which do span more than the calculated area."

Seems sane logic to me. ie It looks a lot like an Osmand bug.

Any thoughts?

PaulW

Stephan Behrendt

unread,

Jan 5, 2019, 11:00:28 AM1/5/19

to Osmand

In the moment,I have the same problem on the river Rhine north of Rüdesheim in scale between 2m and 2 km, then everything is pink. This appears only on the tablet with Android 7. On the Smartfon with Android 9 the map works well.

Stephan

P Wat

unread,

Jan 22, 2019, 6:04:24 AM1/22/19

to Osmand

One month later. The problem persists. We agree there is a problem. It seems to not be caused by the source maps. Ie it seems to be caused inside OsmAnd. Is there an Osmand-based solution?

Paul W

P Wat

unread,

Mar 19, 2019, 6:45:39 PM3/19/19

to Osmand

2 months later. - All looking good now around The Isle Of Wight (off south coast of England, in case you were wondering).

The areas which had been showing as flooded are now properly rendered. Maybe a combination of OSM map updates and OsmAnd updates(?).