H3¶

Utility H3 related functions.

Functions¶

srai.h3.h3_to_shapely_geometry ¶

h3_to_shapely_geometry(
    h3_index: Union[int, str, Iterable[Union[int, str]]],
) -> Union[Polygon, list[Polygon]]

Convert H3 index to Shapely polygon.

PARAMETER	DESCRIPTION
`h3_index`	H3 index (or list of indexes) to be converted. TYPE: `Union[int, str, Iterable[Union[int, str]]]`

RETURNS	DESCRIPTION
`Union[Polygon, list[Polygon]]`	Union[Polygon, List[Polygon]]: Converted polygon (or list of polygons).

Source code in srai/h3.py

def h3_to_shapely_geometry(
    h3_index: Union[int, str, Iterable[Union[int, str]]],
) -> Union[Polygon, list[Polygon]]:
    """
    Convert H3 index to Shapely polygon.

    Args:
        h3_index (Union[int, str, Iterable[Union[int, str]]]): H3 index (or list of indexes)
            to be converted.

    Returns:
        Union[Polygon, List[Polygon]]: Converted polygon (or list of polygons).
    """
    if isinstance(h3_index, (str, int)):
        coords = h3.cell_to_boundary(h3_index, geo_json=True)
        return Polygon(coords)
    return h3_to_geoseries(h3_index).values.tolist()

srai.h3.get_local_ij_index ¶

get_local_ij_index(
    origin_index: str,
    h3_index: Union[str, list[str]],
    return_as_numpy: bool = False,
) -> Union[tuple[int, int], list[tuple[int, int]], npt.NDArray[np.int8]]

Calculate the local H3 ij index based on provided origin index.

Wraps H3's cell_to_local_ij function and centers returned coordinates around provided origin cell.

PARAMETER	DESCRIPTION
`origin_index`	H3 index of the origin region. TYPE: `str`
`h3_index`	H3 index of the second region or list of regions. TYPE: `Union[str, List[str]]`
`return_as_numpy`	Flag whether to return calculated indexes as a Numpy array or a list of tuples. TYPE: `bool` DEFAULT: `False`

RETURNS	DESCRIPTION
`Union[tuple[int, int], list[tuple[int, int]], NDArray[int8]]`	Union[Tuple[int, int], List[Tuple[int, int]], npt.NDArray[np.int8]]: The local ij index of the second region (or regions) with respect to the first one.

Source code in srai/h3.py

def get_local_ij_index(
    origin_index: str, h3_index: Union[str, list[str]], return_as_numpy: bool = False
) -> Union[tuple[int, int], list[tuple[int, int]], npt.NDArray[np.int8]]:
    """
    Calculate the local H3 ij index based on provided origin index.

    Wraps H3's cell_to_local_ij function and centers returned coordinates
    around provided origin cell.

    Args:
        origin_index (str): H3 index of the origin region.
        h3_index (Union[str, List[str]]): H3 index of the second region or list of regions.
        return_as_numpy (bool, optional): Flag whether to return calculated indexes as a Numpy array
            or a list of tuples.

    Returns:
        Union[Tuple[int, int], List[Tuple[int, int]], npt.NDArray[np.int8]]: The local ij index of
            the second region (or regions) with respect to the first one.
    """
    origin_coords = h3.cell_to_local_ij(origin_index, origin_index)
    if isinstance(h3_index, str):
        ijs = h3.cell_to_local_ij(origin_index, h3_index)
        return (origin_coords[0] - ijs[0], origin_coords[1] - ijs[1])
    ijs = np.array([h3.cell_to_local_ij(origin_index, h3_cell) for h3_cell in h3_index])
    local_ijs = np.array(origin_coords) - ijs

    if not return_as_numpy:
        local_ijs = [(coords[0], coords[1]) for coords in local_ijs]

    return local_ijs

srai.h3.ring_buffer_geometry ¶

ring_buffer_geometry(
    geometry: Union[
        BaseGeometry, Iterable[BaseGeometry], gpd.GeoSeries, gpd.GeoDataFrame
    ],
    h3_resolution: int,
    distance: int,
) -> Union[gpd.GeoSeries, BaseGeometry]

Buffer a Shapely geometry with H3 cells, and return the bounding geometry.

If a GeoDataFrame is passed, the geometry column will be used and the return will be a GeoSeries

PARAMETER	DESCRIPTION
`geometry`	The geometry to buffer. TYPE: `Union[BaseGeometry, Iterable[BaseGeometry], GeoSeries, GeoDataFrame]`
`h3_resolution`	The H3 resolution to use. TYPE: `int`
`distance`	The k-ring buffer distance in H3 cells. TYPE: `int`

RETURNS	DESCRIPTION
`Union[GeoSeries, BaseGeometry]`	Union[gpd.GeoSeries, BaseGeometry]: The buffered geometry.

Source code in srai/h3.py

def ring_buffer_geometry(
    geometry: Union[BaseGeometry, Iterable[BaseGeometry], gpd.GeoSeries, gpd.GeoDataFrame],
    h3_resolution: int,
    distance: int,
) -> Union[gpd.GeoSeries, BaseGeometry]:
    """
    Buffer a Shapely geometry with H3 cells, and return the bounding geometry.

    If a GeoDataFrame is passed, the geometry column will be used and the return will be a GeoSeries

    Args:
        geometry (Union[BaseGeometry, Iterable[BaseGeometry], gpd.GeoSeries, gpd.GeoDataFrame]):
            The geometry to buffer.
        h3_resolution (int): The H3 resolution to use.
        distance (int): The k-ring buffer distance in H3 cells.

    Returns:
        Union[gpd.GeoSeries, BaseGeometry]: The buffered geometry.
    """
    if isinstance(geometry, gpd.GeoDataFrame):
        geometry = geometry[GEOMETRY_COLUMN]
        return geometry.apply(lambda x: ring_buffer_geometry(x, h3_resolution, distance))

    if isinstance(geometry, gpd.GeoSeries):
        return geometry.apply(lambda x: ring_buffer_geometry(x, h3_resolution, distance))

    if isinstance(geometry, Iterable):
        return gpd.GeoSeries([ring_buffer_geometry(x, h3_resolution, distance) for x in geometry])

    assert isinstance(geometry, BaseGeometry)
    h3s = shapely_geometry_to_h3(geometry, h3_resolution, buffer=True)
    # buffer all the h3
    buffered_h3s = ring_buffer_h3_indexes(h3s, distance=distance)
    # get the bounding geometry
    return h3_to_geoseries(buffered_h3s).union_all()

srai.h3.h3_to_geoseries ¶

h3_to_geoseries(
    h3_index: Union[int, str, Iterable[Union[int, str]]],
) -> gpd.GeoSeries

Convert H3 index to GeoPandas GeoSeries.

PARAMETER	DESCRIPTION
`h3_index`	H3 index (or list of indexes) to be converted. TYPE: `Union[int, str, Iterable[Union[int, str]]]`

RETURNS	DESCRIPTION
`GeoSeries`	Geometries as GeoSeries with default CRS applied. TYPE: `GeoSeries`

Source code in srai/h3.py

def h3_to_geoseries(h3_index: Union[int, str, Iterable[Union[int, str]]]) -> gpd.GeoSeries:
    """
    Convert H3 index to GeoPandas GeoSeries.

    Args:
        h3_index (Union[int, str, Iterable[Union[int, str]]]): H3 index (or list of indexes)
            to be converted.

    Returns:
        GeoSeries: Geometries as GeoSeries with default CRS applied.
    """
    if isinstance(h3_index, (str, int)):
        return h3_to_geoseries([h3_index])
    else:
        h3_int_indexes = list(
            h3_cell if isinstance(h3_cell, int) else h3.str_to_int(h3_cell) for h3_cell in h3_index
        )
        return gpd.GeoSeries.from_wkb(cells_to_wkb_polygons(h3_int_indexes), crs=WGS84_CRS)

srai.h3.ring_buffer_h3_regions_gdf ¶

ring_buffer_h3_regions_gdf(
    regions_gdf: gpd.GeoDataFrame, distance: int
) -> gpd.GeoDataFrame

Buffer H3 indexes by a given number of k-rings.

List of provided H3 indexes will be buffered by a given distance.

PARAMETER	DESCRIPTION
`regions_gdf`	GeoDataFrame with H3 regions from H3Regionalizer. TYPE: `GeoDataFrame`
`distance`	The k-ring buffer distance in H3 cells. TYPE: `int`

RETURNS	DESCRIPTION
`GeoDataFrame`	gpd.GeoDataFrame: Buffered regions_gdf with new H3 cells added.

Source code in srai/h3.py

def ring_buffer_h3_regions_gdf(regions_gdf: gpd.GeoDataFrame, distance: int) -> gpd.GeoDataFrame:
    """
    Buffer H3 indexes by a given number of k-rings.

    List of provided H3 indexes will be buffered by a given distance.

    Args:
        regions_gdf (gpd.GeoDataFrame): GeoDataFrame with H3 regions from H3Regionalizer.
        distance (int): The k-ring buffer distance in H3 cells.

    Returns:
        gpd.GeoDataFrame: Buffered regions_gdf with new H3 cells added.
    """
    buffered_h3_indexes = ring_buffer_h3_indexes(h3_indexes=regions_gdf.index, distance=distance)
    buffered_gdf_h3 = gpd.GeoDataFrame(
        data={REGIONS_INDEX: buffered_h3_indexes},
        geometry=h3_to_geoseries(buffered_h3_indexes),
        crs=WGS84_CRS,
    ).set_index(REGIONS_INDEX)
    return buffered_gdf_h3

srai.h3.ring_buffer_h3_indexes ¶

ring_buffer_h3_indexes(
    h3_indexes: Iterable[Union[int, str]], distance: int
) -> list[str]

Buffer H3 indexes by a given number of k-rings.

List of provided H3 indexes will be buffered by a given distance.

PARAMETER	DESCRIPTION
`h3_indexes`	List of H3 indexes to be buffered. TYPE: `Iterable[Union[int, str]]`
`distance`	The k-ring buffer distance in H3 cells. TYPE: `int`

RETURNS	DESCRIPTION
`list[str]`	List[str]: Buffered list of H3 cells containing both original and new cells.

Source code in srai/h3.py

def ring_buffer_h3_indexes(h3_indexes: Iterable[Union[int, str]], distance: int) -> list[str]:
    """
    Buffer H3 indexes by a given number of k-rings.

    List of provided H3 indexes will be buffered by a given distance.

    Args:
        h3_indexes (Iterable[Union[int, str]]): List of H3 indexes to be buffered.
        distance (int): The k-ring buffer distance in H3 cells.

    Returns:
        List[str]: Buffered list of H3 cells containing both original and new cells.
    """
    assert all(
        h3.is_valid_cell(h3_cell) for h3_cell in h3_indexes
    ), "Not all values in h3_indexes are valid H3 cells."

    h3_int_indexes = list(
        h3_cell if isinstance(h3_cell, int) else h3.str_to_int(h3_cell) for h3_cell in h3_indexes
    )

    if is_new_h3ronpy_api:
        buffered_h3s = np.unique(
            cells_to_string(grid_disk(h3_int_indexes, distance, flatten=True)).to_numpy()
        )
    else:
        buffered_h3s = set(
            cells_to_string(grid_disk(h3_int_indexes, distance, flatten=True)).tolist()
        )

    return list(buffered_h3s)

srai.h3.shapely_geometry_to_h3 ¶

shapely_geometry_to_h3(
    geometry: Union[
        BaseGeometry, Iterable[BaseGeometry], gpd.GeoSeries, gpd.GeoDataFrame
    ],
    h3_resolution: int,
    buffer: bool = True,
) -> list[str]

Convert Shapely geometry to H3 indexes.

PARAMETER	DESCRIPTION
`geometry`	Shapely geometry to be converted. TYPE: `Union[BaseGeometry, Iterable[BaseGeometry], GeoSeries, GeoDataFrame]`
`h3_resolution`	H3 resolution of the cells. See [1] for a full comparison. TYPE: `int`
`buffer`	Whether to fully cover the geometries with H3 Cells (visible on the borders). Defaults to True. TYPE: `bool` DEFAULT: `True`

RETURNS	DESCRIPTION
`list[str]`	List[str]: List of H3 indexes that cover a given geometry.

RAISES	DESCRIPTION
`ValueError`	If resolution is not between 0 and 15.

References

https://h3geo.org/docs/core-library/restable/

Source code in srai/h3.py

def shapely_geometry_to_h3(
    geometry: Union[BaseGeometry, Iterable[BaseGeometry], gpd.GeoSeries, gpd.GeoDataFrame],
    h3_resolution: int,
    buffer: bool = True,
) -> list[str]:
    """
    Convert Shapely geometry to H3 indexes.

    Args:
        geometry (Union[BaseGeometry, Iterable[BaseGeometry], GeoSeries, GeoDataFrame]):
            Shapely geometry to be converted.
        h3_resolution (int): H3 resolution of the cells. See [1] for a full comparison.
        buffer (bool, optional): Whether to fully cover the geometries with
            H3 Cells (visible on the borders). Defaults to True.

    Returns:
        List[str]: List of H3 indexes that cover a given geometry.

    Raises:
        ValueError: If resolution is not between 0 and 15.

    References:
        1. https://h3geo.org/docs/core-library/restable/
    """
    if not (0 <= h3_resolution <= 15):
        raise ValueError(f"Resolution {h3_resolution} is not between 0 and 15.")

    wkb = []
    if isinstance(geometry, gpd.GeoSeries):
        wkb = geometry.to_wkb()
    elif isinstance(geometry, gpd.GeoDataFrame):
        wkb = geometry[GEOMETRY_COLUMN].to_wkb()
    elif isinstance(geometry, Iterable):
        wkb = [sub_geometry.wkb for sub_geometry in geometry]
    else:
        wkb = [geometry.wkb]

    containment_mode = ContainmentMode.Covers if buffer else ContainmentMode.ContainsCentroid
    h3_indexes = wkb_to_cells(
        wkb, resolution=h3_resolution, containment_mode=containment_mode, flatten=True
    )

    if is_new_h3ronpy_api:
        h3_indexes = np.unique(h3_indexes.to_numpy())
    else:
        h3_indexes = h3_indexes.unique()

    return [h3.int_to_str(h3_index) for h3_index in h3_indexes.tolist()]