Skip to content

Developer Guide: Working on maps

Pierre Kil edited this page Mar 29, 2021 · 14 revisions

Vector maps (Mapbox GL)

The manager has built in support for Mapbox GL and can serve vector tile data. For this to work please ensure that you have the following files available in the deployment folder:

  • map/mapsettings.json - Contains settings related to the map tiles source data and also UI rendering settings (e.g. center point of map when first loaded)
  • map/mapdata.mbtiles - Contains the vector map data (see Creating a custom project](./Developer-Guide:-Creating-a-custom-project) for more details)

If you have the map data and/or settings in a different location then please ensure that the manager environment variables are set also:

MAP_TILES_PATH=../deployment/map/mapdata.mbtiles
MAP_SETTINGS_PATH=../deployment/map/mapsettings.json

Raster maps (Mapbox JS)

If you are working on raster maps (Mapbox JS) then you will need to have the map docker container running, this container serves the raster map tiles from the vector map data.

The container can be started by using the dev-map.yml profile (see here) or you can add a map service to an existing custom project profile (copy the dev-map.yml as a template).

The manager acts as a reverse proxy for the map service and in order to configure the manager to serve raster tiles you need to set the following environment variables:

MAP_TILESERVER_HOST=localhost
MAP_TILESERVER_PORT=8082
MAP_TILESERVER_REQUEST_TIMEOUT=10000

NOTE: By default MAP_TILESERVER_HOST is null which means the reverse proxy is disabled

Downloading maps and extracting smaller tilesets

We currently do not have our own pipeline for extracting/converting OSM data into vector tilesets but depend on the extracts offered on openmaptiles.org. You can download the vector mbtiles file that contains the bounding box of interest.

From that mbtiles file you can extract smaller tilesets with the following procedure:

  1. Rename the source tileset file input.mbtiles
  2. Create docker container with node and python with dir containing input.mbtiles volume mapped: docker run -it --rm -v PATH_TO_INPUT_MBTILES_DIR:/mapdata nikolaik/python-nodejs:python3.8-nodejs12 bash
  3. Install tilelive converter: npm install -g --unsafe @mapbox/mbtiles @mapbox/tilelive
  4. Select and copy boundary box coordinates of desired region (these are the BOUNDS used in the next command; they must be in format [W,S,E,N] e.g. 4.91, 51.27, 5.84, 51.77): http://tools.geofabrik.de/calc/#tab=1 (use the coordinates in 'Simple Copy' and add the commas)
  5. Extract the desired region (adjust the zoom levels as required and optionally add --timeout=Nms if the task times out): tilelive-copy --minzoom=0 --maxzoom=14 --bounds="FILL_BOUNDS_HERE" /mapdata/input.mbtiles /mapdata/output.mbtiles
  6. Rename output.mbtiles into mapdata.mbtiles
Clone this wiki locally