View, validate, and convert GeoJSON in seconds.
Thirty-two tools for working with GeoJSON: a viewer, a validator, a formatter, a simplifier, a bounding-box clipper, round-trip converters for CSV, KML, GPX and WKT, geodesic area and length, centroids, attribute filtering, a CSV attribute join, and export to PostGIS SQL, SVG, and TopoJSON out and back in. Everything runs in your browser — nothing uploaded.
Moving data between formats
To get your spatial data into a modern stack, the
KML → GeoJSON converter,
GPX → GeoJSON converter, and
CSV → GeoJSON converter turn legacy XML, GPS, and spreadsheet files into clean FeatureCollection objects you can immediately use in your apps. Going the other way, the
GeoJSON → KML converter generates KML 2.2 files perfect for Google Earth, the
GeoJSON → GPX converter emits GPX 1.1 for Garmin / Strava / Komoot devices, and the
GeoJSON → CSV tool flattens nested properties into a simple table with lng, lat, and WKT columns for easy spreadsheet analysis.
Inspecting and reshaping GeoJSON
If you just want to see your data, the GeoJSON viewer renders your features on a map instantly without a single server round-trip. When your files are too heavy for mobile web maps, the Simplify GeoJSON tool uses Douglas-Peucker logic to strip out redundant vertices while maintaining the integrity of your shapes. To clean up messy files, the GeoJSON formatter can pretty-print your JSON with a chosen indent or minify it to a single line to save on transfer costs.
Checking for errors
Before shipping a file to a production API, run it through the GeoJSON validator to catch syntax errors or RFC 7946 spec violations. It provides line-accurate reports on everything from geometry types and coordinate ranges to polygon ring closure, saving you from hours of debugging silently failing map renders.
Measuring and summarising
Two questions come up on every file. How big is this? — the area and length tool measures polygons on the sphere in km², hectares, acres or mi² and lines in km, miles or nautical miles, with holes subtracted, so you get a correct answer without projecting anything first. And what is in this? — the stats tool counts features by geometry type, totals the vertices, names the heaviest single geometry, and profiles every property key with its fill rate, which is how you find the attribute that only 40% of features actually carry. For a single coordinate per shape, the centroid tool offers the area-weighted centre, the mean of the vertices, or the bounding-box centre — three answers because "centre" means three different things.
Reshaping a collection
Real-world files rarely arrive in the shape you need.
Merge combines a folder of per-region files into one
layer (open them all at once, and optionally tag each feature with its source);
filter pulls out the forty features you want from a
national download by property value, geometry type, or a regular expression; and
explode splits a MultiPolygon of islands
into one feature per island so each can be styled, labelled, or measured on its own.
When files are too heavy for a web map, coordinate
precision rounds away the nanometre digits that most exports carry — six decimals is
eleven centimetres — and reports the byte saving, while
simplify removes vertices the zoom level cannot show.
Fixing files that other tools reject
Two problems account for most rejections. Winding order:
rewind makes exterior rings counter-clockwise and holes
clockwise per RFC 7946 §3.1.6, which is what MongoDB's 2dsphere index requires — a
clockwise exterior ring there means "everything except this shape". And coordinate systems:
a file whose numbers run into the millions is Web Mercator, not WGS 84, and
reproject converts between EPSG:3857 metres and
EPSG:4326 degrees in either direction.
Getting GeoJSON into other systems
WKT is the text form spatial SQL parses, and
the reverse turns an ST_AsText() column
back into something you can map. For a database load,
PostGIS SQL writes the CREATE TABLE with
inferred column types, a GIST index, and one INSERT per feature. For a figure or a slide,
SVG projects the layer into a standalone vector graphic
with a live preview. For data pipelines,
NDJSON writes one feature per line for BigQuery,
DuckDB and jq, and NDJSON to GeoJSON
puts the collection back together. And for routing APIs, the
encoded polyline converter decodes Google, OSRM and
Valhalla polylines (precision 5 or 6 — the setting everyone gets wrong once).
What they have in common
Everything runs in your browser. The coordinates and private data you drop or paste never leave your machine — there is no backend server processing your files. I can’t see what you’re working on, which means no size limits, no accounts, and no risk of data leaks.
Sensible defaults, no surprises. These tools follow the GeoJSON spec strictly but stay out of your way. Your properties are preserved exactly as they are, and I don't touch your coordinate precision unless you specifically ask me to.
Free to use, with no warranty. This is a collection of tools I built because I was tired of uploading my private data to random sites to do simple tasks. If you find a bug or have a suggestion, feel free to drop me a line at [email protected].
— S., [email protected]