kylebarron / dem-tiler Goto Github PK

https://github.com/kylebarron/dem-mosaic-tiler/blob/f2a6afc75fec6cf5d6d0d6f2d2f36880dbd3df41/dem_mosaic_tiler/handlers/app.py#L281-L286

This issue was generated by todo based on a TODO comment in f2a6afc. It's been assigned to @kylebarron because they committed the code.

Find dem tile data normally, then pass to pymartini, then pass to quantized-mesh-tile, then return

Especially when exporting to quantized mesh, it might be ideal to use the existing AWS terrain tiles.

Also look into using the GeoTIFF terrain tiles, which might be a little simpler to work with, because you don't have to encode/decode from a PNG.

It would be helpful to support exporting 260px and 516px tiles: e.g. 256px and 512px tiles with an extra 1-pixel border on each side.

These would allow for client-side hillshade/slope calculations without needing to find nearby tiles to backfill the edges

E.g. Set to None by default, and then derive from the z using the same as I do in JS?

function getMeshMaxError(z) {
  return (77067.34 / (1 << z)).toFixed(2);
}

attempt with rasterio.features:

floored = np.floor(tile / 10).astype(np.int16)

shapes = list(features.shapes(floored, transform=transform.from_bounds(*mercantile.bounds(x, y, z), 256, 256)))

fs = []
for geom, value in shapes:
    fs.append(Feature(geometry=mapping(LineString(shape(geom).exterior)), properties={'ele': value}))
fc = FeatureCollection(fs)

https://github.com/kylebarron/dem-mosaic-tiler/blob/a97dea0d4fdb21e651cfb937f9e092bd4133d1d6/dem_mosaic_tiler/reader.py#L74-L79

This issue was generated by todo based on a TODO comment in a97dea0. It's been assigned to @kylebarron because they committed the code.

Contours are in the contour layer