from functools import partial import numpy as np import shapely __all__ = ["assert_geometries_equal"] def _equals_exact_with_ndim(x, y, tolerance): dimension_equals = shapely.get_coordinate_dimension( x ) == shapely.get_coordinate_dimension(y) with np.errstate(invalid="ignore"): # Suppress 'invalid value encountered in equals_exact' with nan coordinates geometry_equals = shapely.equals_exact(x, y, tolerance=tolerance) return dimension_equals & geometry_equals def _replace_nan(arr): return np.where(np.isnan(arr), 0.0, arr) def _assert_nan_coords_same(x, y, tolerance, err_msg, verbose): x, y = np.broadcast_arrays(x, y) x_coords = shapely.get_coordinates(x, include_z=True) y_coords = shapely.get_coordinates(y, include_z=True) # Check the shapes (condition is copied from numpy test_array_equal) if x_coords.shape != y_coords.shape: return False # Check NaN positional equality x_id = np.isnan(x_coords) y_id = np.isnan(y_coords) if not (x_id == y_id).all(): msg = build_err_msg( [x, y], err_msg + "\nx and y nan coordinate location mismatch:", verbose=verbose, ) raise AssertionError(msg) # If this passed, replace NaN with a number to be able to use equals_exact x_no_nan = shapely.transform(x, _replace_nan, include_z=True) y_no_nan = shapely.transform(y, _replace_nan, include_z=True) return _equals_exact_with_ndim(x_no_nan, y_no_nan, tolerance=tolerance) def _assert_none_same(x, y, err_msg, verbose): x_id = shapely.is_missing(x) y_id = shapely.is_missing(y) if not (x_id == y_id).all(): msg = build_err_msg( [x, y], err_msg + "\nx and y None location mismatch:", verbose=verbose, ) raise AssertionError(msg) # If there is a scalar, then here we know the array has the same # flag as it everywhere, so we should return the scalar flag. if x.ndim == 0: return bool(x_id) elif y.ndim == 0: return bool(y_id) else: return y_id def assert_geometries_equal( x, y, tolerance=1e-7, equal_none=True, equal_nan=True, normalize=False, err_msg="", verbose=True, ): """Raises an AssertionError if two geometry array_like objects are not equal. Given two array_like objects, check that the shape is equal and all elements of these objects are equal. An exception is raised at shape mismatch or conflicting values. In contrast to the standard usage in shapely, no assertion is raised if both objects have NaNs/Nones in the same positions. Parameters ---------- x : Geometry or array_like y : Geometry or array_like equal_none : bool, default True Whether to consider None elements equal to other None elements. equal_nan : bool, default True Whether to consider nan coordinates as equal to other nan coordinates. normalize : bool, default False Whether to normalize geometries prior to comparison. err_msg : str, optional The error message to be printed in case of failure. verbose : bool, optional If True, the conflicting values are appended to the error message. """ __tracebackhide__ = True # Hide traceback for py.test if normalize: x = shapely.normalize(x) y = shapely.normalize(y) x = np.array(x, copy=False) y = np.array(y, copy=False) is_scalar = x.ndim == 0 or y.ndim == 0 # Check the shapes (condition is copied from numpy test_array_equal) if not (is_scalar or x.shape == y.shape): msg = build_err_msg( [x, y], err_msg + f"\n(shapes {x.shape}, {y.shape} mismatch)", verbose=verbose, ) raise AssertionError(msg) flagged = False if equal_none: flagged = _assert_none_same(x, y, err_msg, verbose) if not np.isscalar(flagged): x, y = x[~flagged], y[~flagged] # Only do the comparison if actual values are left if x.size == 0: return elif flagged: # no sense doing comparison if everything is flagged. return is_equal = _equals_exact_with_ndim(x, y, tolerance=tolerance) if is_scalar and not np.isscalar(is_equal): is_equal = bool(is_equal[0]) if np.all(is_equal): return elif not equal_nan: msg = build_err_msg( [x, y], err_msg + f"\nNot equal to tolerance {tolerance:g}", verbose=verbose, ) raise AssertionError(msg) # Optionally refine failing elements if NaN should be considered equal if not np.isscalar(is_equal): x, y = x[~is_equal], y[~is_equal] # Only do the NaN check if actual values are left if x.size == 0: return elif is_equal: # no sense in checking for NaN if everything is equal. return is_equal = _assert_nan_coords_same(x, y, tolerance, err_msg, verbose) if not np.all(is_equal): msg = build_err_msg( [x, y], err_msg + f"\nNot equal to tolerance {tolerance:g}", verbose=verbose, ) raise AssertionError(msg) ## BELOW A COPY FROM numpy.testing._private.utils (numpy version 1.20.2) def build_err_msg( arrays, err_msg, header="Geometries are not equal:", verbose=True, names=("x", "y"), precision=8, ): msg = ["\n" + header] if err_msg: if err_msg.find("\n") == -1 and len(err_msg) < 79 - len(header): msg = [msg[0] + " " + err_msg] else: msg.append(err_msg) if verbose: for i, a in enumerate(arrays): if isinstance(a, np.ndarray): # precision argument is only needed if the objects are ndarrays r_func = partial(np.array_repr, precision=precision) else: r_func = repr try: r = r_func(a) except Exception as exc: r = f"[repr failed for <{type(a).__name__}>: {exc}]" if r.count("\n") > 3: r = "\n".join(r.splitlines()[:3]) r += "..." msg.append(f" {names[i]}: {r}") return "\n".join(msg)