Plunder loads its data from static js files.

Those files are generated from the plunder database using the sql files in the geosketch/sql folder.

psql -t -d voyc -U jhagstrand <deserts.sql >deserts.js
psql -t -d voyc -U jhagstrand <mountains.sql >mountains.js
psql -t -d voyc -U jhagstrand <plateaux.sql >plateaux.js
psql -t -d voyc -U jhagstrand <plains.sql >plains.js
psql -t -d voyc -U jhagstrand <swamps.sql >swamps.js
psql -t -d voyc -U jhagstrand <tundras.sql >tundras.js
psql -t -d voyc -U jhagstrand <foothills.sql >foothills.js
psql -t -d voyc -U jhagstrand <valleys.sql >valleys.js
psql -t -d voyccom_plunder -U voyccom_jhagstrand <rivers.sql >../../html/data/rivers.js
psql -t -d voyccom_plunder -U voyccom_jhagstrand <lakes.sql >../../html/data/lakes.js
psql -t -d voyc -U jhagstrand <empire.sql >empire.js
psql -t -d voyc -U jhagstrand <treasure.sql >treasure.js

Originally mountains were broken into three js files by scalerank.

• high scalerank 1,2
• medium scalerank 3,4
• low scalerank 4,5

And rivers were broken into six js files by scalerank, 1 thru 6.

OGR can be used to generate geojson files directly from shapefiles, but we found it more convenient to use a sql database as an interim.

Create one large geojson file from the shapefile.

ogr2ogr -f GeoJSON georegions.js 10m_geography_regions_polys.shp

Never figured out how to create multiple geojson files, one for each featurecla.

ogr2ogr -f GeoJSON -clipsrcwhere "FeatureCla='foothills'" foothills.geojson 10m_geography_regions_polys.shp