Style for OMT 3.11 - boundary, water brunnel

[eva-j]

• do not display parts of water bodies that are under surface (you can check in Paris: 15/48.85613/2.37204)
• do not display disputed boundary relations in zooms 0-4 (there is Natural Earth data source)

[pathmapper]

Hi @eva-j,

thanks for your PR.

I think the brunnel filter should also be applied to the waterway layers (waterway_river and waterway_other) because there is still water under the surface visible (Paris area you mentioned). What do you think?

Regarding the boundaries:
I don't see any boundaries with a claimed_by field in the current tiles. Can you point me to boundaries where this is the case or is this PR for a future change in the tiles?
According to the OMT changelog, relations with claimed_by should be already included in OMT v3.11.

[eva-j]

Hi @pathmapper, thanks for having a look!

I've just added waterway_tunnel layer, hopefully it looks better.
Anyway, the source

    "openmaptiles": {
      "type": "vector",
      "url": "https://api.maptiler.com/tiles/v3/tiles.json?key={key}"

will be updated to v3.11 during the weekend, so the changes will become visible.

[pathmapper]

Suggested change

"layout": {"visibility": "visible"},

Could you please remove this line, it's not needed and we recently cleaned up the style in #61 😄

[eva-j]

done ;)

[pathmapper]

@eva-j thanks for adding the waterway_tunnel layer. Makes sense because we are showing currently also the names for subsurface waterways.

The issue I meant is related to this riverbank
https://www.openstreetmap.org/way/236735249#map=18/48.85285/2.36905

I think there is currently no option to hide it, therefore I opened:
openmaptiles/openmaptiles#764

Left one comment regarding a small change, otherwise this PR is good to merge.

[pathmapper]

updated to v3.11 during the weekend, so the changes will become visible.

This fixed the riverbank issue.

[boldtrn]

Is maxzoom: 5 correct here?

[eva-j]

Yes, there is different source of boundary data in z0-4 (Natural Earth) and then starting z5 (OSM).

[boldtrn]

Ah I see, thanks for the explanation.