Use alternate source for world_boundaries-spherical.tgz

[pnorman]

We should use an alternate source for this file. I think it's vmap0 but I'm not sure

[pnorman]

No good alternate source, closing. See also #67 for moving those files to planet.osm.org.

[matthijsmelissen]

Re-opening as well. Would osm-data be disastrous (due to the coastline paradox)?

[pnorman]

We've changed enough stuff around since we did this, so it's worth rechecking.

[jeisenbe]

How is this file currently used?

1. I don't see where any of the components are used in the current project.mml code. The strings "world_bnd_m" and "world_boundaries_m" are only found in install.md in the instruction about downloading shapefiles, and in get-shapefiles.py. The string "builtup_area" is only found in the downloaded data files themselves.

2. The data files have not been modified since 2007 (world_bnd_m.) and 2008 (world_boundaries_m.), it appears.

3. I've deleted the whole /world_boundaries directory and found that the map renders without problems down to z2, [but I had issues at z0 and z1 EDIT: see next comment]

[jeisenbe]

Actually, it renders fine at z0 and z1 now that I've change my localconfig.json file back to metatile: 2
I couldn't render the current master branch at z0 or z1 with metatile: 8 even with the /world_boundaries directory in the correct place, so that's probably not related?

[jeisenbe]

@pnorman - is this file still needed? It doesn't appear to be necessary for rendering

[pnorman]

I haven't touched that part of the code in ages so don't know.

[sommerluk]

The last usage of world_boundaries-spherical.tgz had been removed in 5327dbb (buildup areas). Closing.