Skip to content

Chunked Geometry Arrays

Chunked arrays of geospatial geometries, each of the same type.

geoarrow.rust.core

ChunkedPointArray

An immutable chunked array of Point geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> Point

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chunks method descriptor 

chunks() -> List[PointArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> PointArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

length method descriptor 

length(*, method: LengthMethod | LengthMethodT = 'euclidean') -> ChunkedFloat64Array

Calculation of the length of a Line

Other Parameters:

  • method (LengthMethod | LengthMethodT) –

    The method to use for length calculation. One of "Ellipsoidal", "Euclidean", "Haversine", or "Vincenty". Refer to the documentation on LengthMethod for more information.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedLineStringArray

An immutable chunked array of LineString geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> LineString

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chaikin_smoothing method descriptor 

chaikin_smoothing(n_iterations: int) -> Self

Smoothen LineString, Polygon, MultiLineString and MultiPolygon using Chaikins algorithm.

Chaikins smoothing algorithm

Each iteration of the smoothing doubles the number of vertices of the geometry, so in some cases it may make sense to apply a simplification afterwards to remove insignificant coordinates.

This implementation preserves the start and end vertices of an open linestring and smoothes the corner between start and end of a closed linestring.

Parameters:

  • n_iterations (int) –

    Number of iterations to use for smoothing.

Returns:

  • Self –

    Smoothed geometry array.

chunks method descriptor 

chunks() -> List[LineStringArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> LineStringArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

densify method descriptor 

densify(max_distance: float) -> Self

Return a new linear geometry containing both existing and new interpolated coordinates with a maximum distance of max_distance between them.

Note: max_distance must be greater than 0.

Parameters:

  • max_distance (float) –

    maximum distance between coordinates

Returns:

  • Self –

    Densified geometry array

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

frechet_distance method descriptor 

frechet_distance(other: BroadcastGeometry) -> ChunkedFloat64Array

Determine the similarity between two arrays of LineStrings using the Frechet distance.

The Fréchet distance is a measure of similarity: it is the greatest distance between any point in A and the closest point in B. The discrete distance is an approximation of this metric: only vertices are considered. The parameter ‘densify’ makes this approximation less coarse by splitting the line segments between vertices before computing the distance.

Fréchet distance sweep continuously along their respective curves and the direction of curves is significant. This makes it a better measure of similarity than Hausdorff distance for curve or surface matching.

This implementation is based on Computing Discrete Frechet Distance by T. Eiter and H. Mannila.

Parameters:

  • other (BroadcastGeometry) –

    the geometry or geometry array to compare against. This must contain geometries of `LineString`` type. A variety of inputs are accepted:

Returns:

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

length method descriptor 

length(*, method: LengthMethod | LengthMethodT = 'euclidean') -> ChunkedFloat64Array

Calculation of the length of a Line

Other Parameters:

  • method (LengthMethod | LengthMethodT) –

    The method to use for length calculation. One of "Ellipsoidal", "Euclidean", "Haversine", or "Vincenty". Refer to the documentation on LengthMethod for more information.

Returns:

line_interpolate_point method descriptor 

line_interpolate_point(fraction: float | int | ArrowStreamExportable) -> ChunkedPointArray

Returns a point interpolated at given distance on a line.

This is intended to be equivalent to shapely.line_interpolate_point when normalized=True.

If the given fraction is * less than zero (including negative infinity): returns the starting point * greater than one (including infinity): returns the ending point * If either the fraction is NaN, or any coordinates of the line are not finite, returns Point EMPTY.

Parameters:

  • input –

    input geometry array or chunked geometry array

  • fraction (float | int | ArrowStreamExportable) –

    the fractional distance along the line. A variety of inputs are accepted:

    • A Python float or int
    • A numpy ndarray with float64 data type.
    • An Arrow array or chunked array with float64 data type.

Returns:

line_locate_point method descriptor 

line_locate_point(point: GeoInterfaceProtocol | ArrowStreamExportable) -> ChunkedFloat64Array

Returns a fraction of the line's total length representing the location of the closest point on the line to the given point.

This is intended to be equivalent to shapely.line_locate_point when normalized=True.

If the line has zero length the fraction returned is zero.

If either the point's coordinates or any coordinates of the line are not finite, returns NaN.

Parameters:

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedPolygonArray

An immutable chunked array of Polygon geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> Polygon

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chaikin_smoothing method descriptor 

chaikin_smoothing(n_iterations: int) -> Self

Smoothen LineString, Polygon, MultiLineString and MultiPolygon using Chaikins algorithm.

Chaikins smoothing algorithm

Each iteration of the smoothing doubles the number of vertices of the geometry, so in some cases it may make sense to apply a simplification afterwards to remove insignificant coordinates.

This implementation preserves the start and end vertices of an open linestring and smoothes the corner between start and end of a closed linestring.

Parameters:

  • n_iterations (int) –

    Number of iterations to use for smoothing.

Returns:

  • Self –

    Smoothed geometry array.

chunks method descriptor 

chunks() -> List[PolygonArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> PolygonArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

densify method descriptor 

densify(max_distance: float) -> Self

Return a new linear geometry containing both existing and new interpolated coordinates with a maximum distance of max_distance between them.

Note: max_distance must be greater than 0.

Parameters:

  • max_distance (float) –

    maximum distance between coordinates

Returns:

  • Self –

    Densified geometry array

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

polylabel method descriptor 

polylabel(tolerance: float) -> ChunkedPointArray

Calculate a Polygon's ideal label position by calculating its pole of inaccessibility.

The pole of inaccessibility is the most distant internal point from the polygon outline (not to be confused with centroid), and is useful for optimal placement of a text label on a polygon.

The calculation uses an iterative grid-based algorithm, ported from the original JavaScript implementation.

Parameters:

Returns:

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedMultiPointArray

An immutable chunked array of MultiPoint geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> MultiPoint

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chunks method descriptor 

chunks() -> List[MultiPointArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> MultiPointArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

length method descriptor 

length(*, method: LengthMethod | LengthMethodT = 'euclidean') -> ChunkedFloat64Array

Calculation of the length of a Line

Other Parameters:

  • method (LengthMethod | LengthMethodT) –

    The method to use for length calculation. One of "Ellipsoidal", "Euclidean", "Haversine", or "Vincenty". Refer to the documentation on LengthMethod for more information.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedMultiLineStringArray

An immutable chunked array of MultiLineString geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> MultiLineString

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chaikin_smoothing method descriptor 

chaikin_smoothing(n_iterations: int) -> Self

Smoothen LineString, Polygon, MultiLineString and MultiPolygon using Chaikins algorithm.

Chaikins smoothing algorithm

Each iteration of the smoothing doubles the number of vertices of the geometry, so in some cases it may make sense to apply a simplification afterwards to remove insignificant coordinates.

This implementation preserves the start and end vertices of an open linestring and smoothes the corner between start and end of a closed linestring.

Parameters:

  • n_iterations (int) –

    Number of iterations to use for smoothing.

Returns:

  • Self –

    Smoothed geometry array.

chunks method descriptor 

chunks() -> List[MultiLineStringArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> MultiLineStringArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

densify method descriptor 

densify(max_distance: float) -> Self

Return a new linear geometry containing both existing and new interpolated coordinates with a maximum distance of max_distance between them.

Note: max_distance must be greater than 0.

Parameters:

  • max_distance (float) –

    maximum distance between coordinates

Returns:

  • Self –

    Densified geometry array

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

length method descriptor 

length(*, method: LengthMethod | LengthMethodT = 'euclidean') -> ChunkedFloat64Array

Calculation of the length of a Line

Other Parameters:

  • method (LengthMethod | LengthMethodT) –

    The method to use for length calculation. One of "Ellipsoidal", "Euclidean", "Haversine", or "Vincenty". Refer to the documentation on LengthMethod for more information.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedMultiPolygonArray

An immutable chunked array of MultiPolygon geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> MultiPolygon

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chaikin_smoothing method descriptor 

chaikin_smoothing(n_iterations: int) -> Self

Smoothen LineString, Polygon, MultiLineString and MultiPolygon using Chaikins algorithm.

Chaikins smoothing algorithm

Each iteration of the smoothing doubles the number of vertices of the geometry, so in some cases it may make sense to apply a simplification afterwards to remove insignificant coordinates.

This implementation preserves the start and end vertices of an open linestring and smoothes the corner between start and end of a closed linestring.

Parameters:

  • n_iterations (int) –

    Number of iterations to use for smoothing.

Returns:

  • Self –

    Smoothed geometry array.

chunks method descriptor 

chunks() -> List[MultiPolygonArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> MultiPolygonArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

densify method descriptor 

densify(max_distance: float) -> Self

Return a new linear geometry containing both existing and new interpolated coordinates with a maximum distance of max_distance between them.

Note: max_distance must be greater than 0.

Parameters:

  • max_distance (float) –

    maximum distance between coordinates

Returns:

  • Self –

    Densified geometry array

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

simplify method descriptor 

simplify(epsilon: float, *, method: SimplifyMethod | SimplifyMethodT = 'rdp') -> Self

Simplifies a geometry.

Parameters:

  • epsilon (float) –

    tolerance for simplification. An epsilon less than or equal to zero will return an unaltered version of the geometry.

Other Parameters:

Returns:

  • Self –

    Simplified geometry array.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedMixedGeometryArray

An immutable chunked array of Geometry geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> Geometry

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chunks method descriptor 

chunks() -> List[MixedGeometryArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> MixedGeometryArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedGeometryCollectionArray

An immutable chunked array of GeometryCollection geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> GeometryCollection

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

affine_transform method descriptor 

affine_transform(transform)

Apply an affine transformation to geometries.

This is intended to be equivalent to shapely.affinity.affine_transform for 2D transforms.

Parameters:

  • other –

    an affine transformation to apply to all geometries.

    This integrates with the affine Python library, and most users should use that integration, though it allows any input that is a tuple with 6 or 9 float values.

Returns:

  • –

    New GeoArrow array or chunked array with the same type as input and with

  • –

    transformed coordinates.

area method descriptor 

area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Unsigned area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

center method descriptor 

center() -> ChunkedPointArray

Compute the center of geometries

This first computes the axis-aligned bounding rectangle, then takes the center of that box

Returns:

centroid method descriptor 

centroid() -> ChunkedPointArray

Calculation of the centroid.

The centroid is the arithmetic mean position of all points in the shape. Informally, it is the point at which a cutout of the shape could be perfectly balanced on the tip of a pin.

The geometric centroid of a convex object always lies in the object. A non-convex object might have a centroid that is outside the object itself.

Returns:

chunks method descriptor 

chunks() -> List[GeometryCollectionArray]

Convert to a list of single-chunked arrays.

concatenate method descriptor 

concatenate() -> GeometryCollectionArray

Concatenate a chunked array into a contiguous array.

convex_hull method descriptor 

convex_hull() -> ChunkedPolygonArray

Returns the convex hull of a Polygon. The hull is always oriented counter-clockwise.

This implementation uses the QuickHull algorithm, based on Barber, C. Bradford; Dobkin, David P.; Huhdanpaa, Hannu (1 December 1996) Original paper here: www.cs.princeton.edu/~dpd/Papers/BarberDobkinHuhdanpaa.pdf

Returns:

envelope method descriptor 

envelope()

Computes the minimum axis-aligned bounding box that encloses an input geometry

Returns:

  • –

    Array with axis-aligned bounding boxes.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

geodesic_perimeter method descriptor 

geodesic_perimeter() -> ChunkedFloat64Array

Determine the perimeter of a geometry on an ellipsoidal model of the earth.

This uses the geodesic measurement methods given by Karney (2013).

For a polygon this returns the sum of the perimeter of the exterior ring and interior rings. To get the perimeter of just the exterior ring of a polygon, do polygon.exterior().geodesic_length().

Units
  • return value: meter

Returns:

is_empty method descriptor 

is_empty() -> BooleanArray

Returns True if a geometry is an empty point, polygon, etc.

Returns:

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

signed_area method descriptor 

signed_area(*, method: AreaMethod | AreaMethodT = 'euclidean') -> ChunkedFloat64Array

Signed area of a geometry array

Parameters:

  • method (AreaMethod | AreaMethodT, default: 'euclidean' ) –

    The method to use for area calculation. One of "Ellipsoidal", "Euclidean", or "Spherical". Refer to the documentation on AreaMethod for more information.

Returns:

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedWKBArray

An immutable chunked array of WKB-encoded geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__getitem__ method descriptor 

__getitem__(key: int) -> WKB

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

chunks method descriptor 

chunks() -> List[WKBArray]

Convert to a list of single-chunked arrays.

from_arrow_arrays builtin 

from_arrow_arrays(input: Sequence[ArrowArrayExportable]) -> Self

Construct this chunked array from existing Arrow data

This is a temporary workaround for this pyarrow issue, where it's currently impossible to read a pyarrow ChunkedArray directly without adding a direct dependency on pyarrow.

Parameters:

Returns:

from_shapely builtin 

from_shapely(input, *, chunk_size: int = 65536) -> Self

Create this array from a shapely array

Args:

input: Any array object accepted by shapely.to_ragged_array, including numpy object arrays and geopandas.GeoSeries

Other args:

chunk_size: Maximum number of items per chunk.

Returns:

A new chunked array.

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

to_shapely method descriptor 

to_shapely() -> NDArray[np.object_]

Convert this array to a shapely array

Returns:

A shapely array.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns:

ChunkedRectArray

An immutable chunked array of Rect geometries using GeoArrow's in-memory representation.

__arrow_c_stream__ method descriptor 

__arrow_c_stream__(requested_schema: object | None = None) -> object

An implementation of the Arrow PyCapsule Interface. This dunder method should not be called directly, but enables zero-copy data transfer to other Python libraries that understand Arrow memory.

For example (as of the upcoming pyarrow v16), you can call pyarrow.chunked_array() to convert this array into a pyarrow array, without copying memory.

__eq__ method descriptor 

__eq__(value) -> bool

Return self==value.

__getitem__ method descriptor 

__getitem__(key: int) -> Rect

Return self[key].

__repr__ method descriptor 

__repr__() -> str

Return repr(self).

chunks method descriptor 

chunks() -> List[RectArray]

Convert to a list of single-chunked arrays.

num_chunks method descriptor 

num_chunks() -> int

Number of underlying chunks.

total_bounds method descriptor 

total_bounds() -> Tuple[float, float, float, float]

Computes the total bounds (extent) of the geometry.

Returns: