GeoSeries.invalid_coverage_edges

Return edges causing invalid polygonal coverage for each geometry. The result preserves every input row and index level. Polygon candidate neighbours are discovered with a native spatial join over envelopes expanded by gap_width and aggregated on Spark executors; geometry rows are not collected to the driver.

Signature

invalid_coverage_edges(*, gap_width=0.0) -> GeoSeries

Parameters

float

default:"0.0"

Finite, non-negative maximum width of gaps to report as coverage invalidities.

Returns

GeoSeries

Invalid LineString or MultiLineString edges for each input geometry. Rows without invalid edges or polygonal components contain an empty line, missing geometries remain missing, and the returned GeoSeries is unnamed.

Executor memory for an individual source row scales with the number of polygon candidates whose envelopes fall within gap_width.

See also

GeoSeries.is_valid_coverage: test whether the full geometry column is
a coverage

Examples

>>> from shapely.geometry import box
>>> from sedona.spark.geopandas import GeoSeries
>>> overlapping = GeoSeries(
...     [box(0, 0, 1, 1), box(0.5, 0, 1.5, 1)]
... )
>>> edges = overlapping.invalid_coverage_edges()
>>> edges.to_geopandas().is_empty.tolist()
[False, False]

​Signature

​Parameters

​Returns

​Examples

Signature

Parameters

Returns

Examples