geodeploy

Command-line client and Python API for GeoDeploy — the self-hosted spatial data platform and geoportal builder.

Upload data, style it, build a portal and publish it, without opening a browser:

pip install geodeploy

geodeploy login https://geodeploy.example.org --token gdp_…
geodeploy upload roads.gpkg sites.csv dem.tif --wait
geodeploy portals create "Field sites 2026" --publish

No dependencies. Python 3.9+. Every request goes through the standard library, so installing this pulls in nothing else — and a QGIS plugin can vendor the same client without asking anyone to pip-install into QGIS.

If your shell cannot find geodeploy after installing, pip's scripts directory is not on your PATH — python -m geodeploy works regardless, and pipx install geodeploy or a virtual environment avoids it.

What it covers

• Uploads — shapefile, GeoPackage, GeoJSON, CSV (X/Y or WKT), GeoParquet, GeoTIFF. Many files in one command; the route is chosen per file, and anything over 48 MB goes direct to object storage in parallel presigned parts, so a multi-gigabyte upload survives a proxy that caps request bodies.
• Layers — list, inspect, rename, share (STAC / OGC API - Features opt-in), share links per tool, download, delete, restart a stalled ingest.
• Symbology — colour, opacity, outlines, dashes, markers, raster colormaps and stretches, plus data-driven styling: --color-field pop --classify jenks --classes 6 --ramp magma, proportional size, and attribute-driven 3D. Classification is computed by the instance, with the same code the portal editor uses, so the CLI and the editor can never disagree about a class.
• Portals — create (web map, story map or catalog), arrange and style layers, folders, About page, assets, access tiers, publish, and a whole-configuration JSON round trip for version control.
• Everything else — external WMS/XYZ/WFS services, registering data already in PostGIS or the bucket, the public STAC and OGC API - Features catalog, ingest jobs, users, and instance administration (health, services, updates, backups, activity log).

Every command takes --json, where stdout is exactly one JSON document, and exit codes distinguish authentication (3) from network (4) from server (5) so a scheduled job can alert on the right thing.

As a library

from geodeploy import Client

gd = Client("https://geodeploy.example.org", token="gdp_…")
result = gd.uploads.upload("roads.gpkg", wait=True)
portal = gd.portals.create("Roads")
gd.portals.add_layer(portal["id"], result.layer_id, "vector", {"color": "#e11d48"})
gd.portals.publish(portal["id"])

Typed exceptions (AuthError, PermissionError_, NotFoundError, ValidationError, ServerError, TransportError, JobFailed), progress callbacks and cancellation on uploads, and a swappable transport so a desktop application can route requests through its own network stack.

Documentation

Full guide: https://docs-geodeploy.kndev.org/cli/

Source and issues: https://github.com/bravemaster3/GeoDeploy

Licensed under the Apache License 2.0.