Database guides
GeoJSON files
Open a GeoJSON file and its features land on an interactive map — with the properties queryable as a normal table.
GeoJSON is the web’s standard for geographic features, and AddisDB treats it as both: the properties become table columns you can query with SQL, and the geometry renders on the Map view.
Who it is for
GeoJSON is plain text, so it travels well — it is what mapping APIs return, what open-data portals publish, and what most web mapping libraries consume directly. Every feature carries a geometry and a bag of properties.
The specification is unusually strict in one helpful way: coordinates are always longitude then latitude, in WGS 84. Most "my points are in the ocean off West Africa" bugs are that pair swapped, and most of the rest are a file that quietly used a projected coordinate system anyway.
Use it for boundary and administrative data, points of interest, route and trace data, and anything you have downloaded from a public data portal and want to look at before writing code around it.
Getting a file
- Open-data portals — city, state and national — publish GeoJSON directly. So do Natural Earth and OpenStreetMap extract services.
- A FeatureCollection is the usual shape; a bare Feature or geometry works too.
- GeoJSON is defined in WGS 84 (longitude/latitude), which is what the map expects.
- Large files are verbose — a GeoPackage or shapefile of the same data will be considerably smaller.
- Line-delimited GeoJSON (one feature per line) is a different format; convert it first if that is what you have.
Open it in AddisDB
- New Connection → Files → Map → GeoJSON.
- Choose the .geojson (or .json) file, name the connection, and Save.
Seeing it on the map
Run a query that returns the geometry column and a Map tab appears in the results toolbar. Click it and the features are drawn on an interactive map. The basemap is switchable live, so you can move between a street map, satellite imagery and a plain canvas depending on what you are trying to see.

Because the map plots the query result, filtering is how you make a dense file readable — and how you check a hypothesis. Selecting only the features whose properties look wrong puts exactly those on the map and nothing else.
SELECT name, population, geometry
FROM districts
WHERE population > 50000
ORDER BY population DESC;
What else AddisDB gives you
- SQL over the feature properties — filter to the features you care about and the map follows your query.
- The Chart view for aggregating properties, and the JSON tab for the raw feature objects.
- Notebooks and AI chat grounded in the file’s real columns.
- ⌘K search across every connection, and a source file that is never modified.