API

gtravyl.euclidean_dist(ind1: tuple[int, int], ind2: tuple[int, int], value1: Any, value2: Any)

Neighbor indices are computed under the standard Euclidean distance.

Parameters:

  • ind1 – Location in Euclidean space for the first point.

  • ind2 – Location in Euclidean space for the second point.

  • value1 – Needed for type signature, essentially ignored.

  • value2 – Needed for type signature, essentially ignored.

gtravyl.in_bounds(pt: tuple[int, int], dim: tuple[int, int]) → bool

Checks if the candidate point actually exists in the grid.

Parameters:

  • pt – The point to check if it is in bounds.

  • dim – Dimensions (row, column) of the grid.

gtravyl.moore_neighbors(ind: tuple[int, int], grid: ~numpy.ndarray[tuple[~typing.Any, ...], ~numpy.dtype[~numpy._typing._array_like._ScalarT]], wrap=<function no_wrap>)

Return the Moore neighborhood of a particular cell in the grid.

Parameters:

  • ind – Index to compute Moore neighborhood of.

  • grid – The grid ind belongs to.

gtravyl.no_wrap(ind: tuple[int, int], dim: tuple[int, int]) → tuple[int, int]

Does not wrap indices. Essentially does nothing in this case.

Parameters:

  • ind – Needed for type signature. This function is equivalent to: id(ind) -> ind

  • dim – Needed for type signature, it is otherwise ignored.

gtravyl.not_one(value: Any)

Checks if a value is not 1. You can think of 1 as though there is a wall and you are not allowed to traverse there. Any other value is fine. (This is the default function for the keyword argument allowed of shortest_path.

Parameters:

value – Value to check.

gtravyl.shortest_path(grid: ~numpy.ndarray[tuple[~typing.Any, ...], ~numpy.dtype[~numpy._typing._array_like._ScalarT]], si: tuple[int, int] | None = None, ti: tuple[int, int] | None = None, sv: ~typing.Any | None = None, tv: ~typing.Any | None = None, neighbors=<function vn_neighbors>, allowed=<function not_one>, wrap=<function no_wrap>, dist=<function unit_dist>, heuristic=<function no_heuristic>) → list[tuple[int, int]]

Find shortest path from s to t in a given grid.

Parameters:

  • grid – grid (2d array) representation of the world to traverse.

  • si – The “source” index, i.e. where the path search starts.

  • ti – The “destination” index, i.e. where the path should end.

  • sv – The “source” value, i.e. the value of the start index.

  • tv – The “destination” value, i.e. the value of the end index.

  • neighbors – Computes the neighborhood of any choice of index in the grid.

  • wrap – Does the grid wrap around. Defaults to no wrap.

  • dist – The distance function used. Defaults to unit distance.

gtravyl.unit_dist(ind1: tuple[int, int], ind2: tuple[int, int], value1: Any, value2: Any)

Every neighbor is 1 unit distant from the other.

gtravyl.vn_neighbors(ind: tuple[int, int], grid: ~numpy.ndarray[tuple[~typing.Any, ...], ~numpy.dtype[~numpy._typing._array_like._ScalarT]], wrap=<function no_wrap>)

Return the von Neumann neighborhood of a particular cell in the grid.

Parameters:

  • ind – Index to compute von Neumann neighborhood of.

  • grid – The grid ind belongs to.

gtravyl.wrap(ind: tuple[int, int], dim: tuple[int, int]) → tuple[int, int]

Wraps the cell values around. So, if you start at (0, 0) and go left you end up at (x dimension, 0) :param ind: The index to potentially modify (that is, wrap it if

applicable).

Parameters:

dim – Dimensions of the grid.