routing.core.functions#

Base file for storing all useful functions.

class routing.core.functions.CompressedArray(data: bytes, shape: tuple, dtype: dtype)[source]#

Bases: object

A compressed numpy array using zlib.

data: bytes#

dtype: dtype#

property nbytes: int#: Return compressed size in bytes.

shape: tuple#

routing.core.functions.box_window(bounds: dict[str, tuple[int, int]]) → tuple[slice, slice, slice][source]#: Index slices for the inclusive box bounds (per-axis (lowest, highest)), to be applied to any grid-shaped array.

routing.core.functions.build_frame_from_vector(node: ndarray, vector: ndarray) → ndarray[source]#: Build a standard default frame from origin node and main axis vector.

routing.core.functions.build_frame_vectorized(node: ndarray, next_node: ndarray) → ndarray[source]#: Build a standard default frame from two points.

routing.core.functions.clipped_grid_window(lower: ndarray, upper: ndarray, grid_shape: ndarray) → tuple[slice, slice, slice][source]#: Index slices for [lower, upper) trimmed to grid_shape; empty slices when the box misses the grid.

routing.core.functions.cluster_points(points: list[ndarray], tolerance: float) → list[ndarray][source]#: Greedily cluster points lying within tolerance of an existing cluster member; one centroid per cluster.

routing.core.functions.compress_array(array: ndarray | CompressedArray | None) → CompressedArray | None[source]#: Compress a numpy array using zlib compression.

routing.core.functions.csv_to_dicts(path: str) → list[dict][source]#: Read a CSV file into a list of dicts, parsing numeric values and skipping empty cells.

routing.core.functions.cylinder_bounding_box(start: ndarray, end: ndarray, radius: float) → ndarray[source]#: Get the bounding box of a cylinder.

routing.core.functions.from_bounding_rectangles(bounding_rectangles: list[BoundingRectangle], name: str = '') → BoundingRectangle[source]#: Bounding rectangle enclosing all bounding_rectangles.

routing.core.functions.get_bounding_rectangle(center: tuple[float, float], radius: float) → BoundingRectangle[source]#: Bounding rectangle of the circle of given center and radius.

routing.core.functions.get_crossed_voxels(weights: ndarray, node1: tuple[int, int, int], node2: tuple[int, int, int]) → ndarray[source]#: Voxels crossed by the node1-node2 segment, node1 first.

routing.core.functions.get_direction(start: ndarray, end: ndarray) → tuple[ndarray, float, ndarray][source]#: Get vector, length and unit_vector of a vector defined with start and end points.

routing.core.functions.grid_index_bounds(origin: tuple[float, float, float], shape: tuple[int, int, int], voxel_size: float, parent_origin: tuple[float, float, float], parent_voxel_size: float) → dict[str, tuple[int, int]][source]#

Inclusive index box a grid spans, expressed in a coarser parent grid’s indices.

Both extreme voxel centres are rounded onto the parent grid, so the two voxel sizes need not divide one another — unlike RoutingVoxelization.nesting_in(), which requires a whole block.

routing.core.functions.grid_window(lower: ndarray, upper: ndarray) → tuple[slice, slice, slice][source]#: Index slices for the half-open box [lower, upper), to be applied to any grid-shaped array.

routing.core.functions.h_coord(coords: ndarray) → ndarray[source]#: Convert coordinates to homogeneous coordinates by adding a row of ones.

routing.core.functions.index_sphere(radius: float) → ndarray[source]#

Boolean cube marking the index offsets no further than radius voxels from its centre.

The lattice ball a halo grows by: radius 0 is the centre alone, 1 the six face neighbours (the diagonals are at sqrt(2)), 2 reaches the second cardinal ring.

routing.core.functions.invert_frame(frame: ndarray) → ndarray[source]#: Invert a HTM frame with inverse movement instead of inverse matrix.

routing.core.functions.ordinal_suffix(number: int) → str[source]#

routing.core.functions.parse_csv_value(value: str) → int | float | str[source]#: Parse a CSV string value to int, float, or str.

routing.core.functions.points_to_homogeneous_matrix(points: list) → ndarray[source]#: Convert a list of points to a homogeneous transformation matrix.

routing.core.functions.redefines(method: Callable, base_function: Callable) → bool[source]#: Whether the bound method runs another function than base_function: a subclass or the instance redefined it.

routing.core.functions.set_primitives_name(primitives: list[Primitive3D], name: str) → None[source]#: Add an indexed name to primitives when a name is defined as argument.

routing.core.functions.sheet_to_dicts(sheet: openpyxl.worksheet.worksheet.Worksheet) → list[dict][source]#

Convert an openpyxl worksheet to a list of dicts using the first row as headers.

Empty cells are excluded from each dict so that callers can rely on dict.get(key, default).

routing.core.functions.unflat_coords(flatten_coords: ndarray) → ndarray[source]#: Reshape a flat (3 * n_points,) coordinate vector into (n_points, 3).