sedonadb-geopandas

A GeoPandas-compatible API on top of SedonaDB.

The goal is to let existing GeoPandas code run against SedonaDB's relational engine with minimal changes, by providing GeoDataFrame / GeoSeries wrappers whose methods mirror GeoPandas but delegate to SedonaDB expressions.

import geopandas
import sedonadb_geopandas as sgpd

gdf = sgpd.from_geopandas(geopandas.read_file("cities.geojson"))
big = gdf[gdf["pop"] > 1_000_000]          # boolean-mask filter
buffered = gdf.geometry.buffer(0.5)        # element-wise .geo operation
web = gdf.to_crs("EPSG:3857")              # reproject (CRS tracked through)
result = web.to_geopandas()                # back to a real GeoDataFrame

Intentional differences from GeoPandas

This is a compatibility layer over a lazy, relational engine, so it is deliberately not identical to GeoPandas:

• Lazy, not eager: operations build a query; data materializes on to_geopandas() / to_pandas() / display.
• No row index / alignment: there is no pandas Index; joins and filters are positional/relational, not index-aligned.
• Immutable under the hood: "in-place" style operations return a new frame.
• Plotting and arbitrary apply: use the to_geopandas() escape hatch and operate on the materialized result.

See the SedonaDB "Migrating from GeoPandas" guide for the relational model that underlies each method.