Skip to content

Latest commit

 

History

History
69 lines (44 loc) · 2.39 KB

INSTALL.md

File metadata and controls

69 lines (44 loc) · 2.39 KB

Installation

This document explain how to configure your system for running OpenStreetMap hiking.

Data configuration

OpenStreetMap data

Please as first step follow OpenStreetMap Carto instructions since for the base data is sharing the same data structure.

The next step is to create the two tables to render corretly the routes. It use the flex output, it allows for a flexible configuration that tells osm2pgsql what OSM data to store in your database and exactly where and how.

osm2pgsql -d gis --output=flex -S scripts/route.lua ~/path/to/data.osm.pbf

Elevation data

This style use several several layers derived from a digital terrain model to correctly represent the shape of the terrain. Starting from a DTM, three raster layers are created to be used as background: a shaded relief map, a slope map and a colour relief map.

You can create these layers using GDAL's tool gdaldem. For each raster we had the overview to speed up raster reading.

First we create the shaded relief

gdaldem hillshade -co compress=lzw -co predictor=2 input_dtm.tif layers/hillshade.tif

gdaladdo --config COMPRESS_OVERVIEW JPEG layers/hillshade.tif

After the colour relief map

gdaldem color-relief -co compress=lzw -co predictor=2 input_dtm.tif scripts/color-relief.txt layers/relief.tif

gdaladdo --config COMPRESS_OVERVIEW JPEG layers/relief.tif

Finally we create the slope map, it requires two steps, first create the slope map and then colour it.

gdaldem slope input_dtm.tif output_slope.tif

gdaldem color-relief -co compress=lzw -co predictor=2 output_slope.tif script/slope-relief.txt layers/slope.tif
gdaladdo --config COMPRESS_OVERVIEW JPEG layers/slope.tif
rm -f output_slope.tif

The last layer required to represent the terrain model is the contour layer, which can be created with the gdal_contour tool. Two values must be set, the name of the column that will contain the altitude values and the interval between the lines, in our case the value is 100 metres

gdal_contour -i 100 -a elev input_dtm.tif contour.shp

Now we need to import the output shapefile into PostgreSQL database

shp2pgsql -g wkb_geometry contour.shp contour | psql gis

Custom index

Custom indexes are required for rendering performance and are essential for each databases.

psql -d gis -f scripts/indexes.sql