import math import geopandas as gpd import numpy as np import pytest from shapely.geometry import MultiLineString, MultiPolygon, Point, Polygon import momepy as mm from momepy.shape import _circle_area class TestShape: def setup_method(self): test_file_path = mm.datasets.get_path("bubenec") self.df_buildings = gpd.read_file(test_file_path, layer="buildings") self.df_streets = gpd.read_file(test_file_path, layer="streets") self.df_tessellation = gpd.read_file(test_file_path, layer="tessellation") self.df_buildings["height"] = np.linspace(10.0, 30.0, 144) self.df_buildings["volume"] = mm.Volume(self.df_buildings, "height").series def test_FormFactor(self): self.df_buildings["ff"] = mm.FormFactor( self.df_buildings, "volume", heights="height" ).series check = 5.4486362624193 assert self.df_buildings["ff"].mean() == pytest.approx(check) self.df_buildings["ff"] = mm.FormFactor( self.df_buildings, mm.Volume(self.df_buildings, "height").series, areas=self.df_buildings.geometry.area, heights=self.df_buildings["height"], ).series assert self.df_buildings["ff"].mean() == pytest.approx(check) def test_FractalDimension(self): self.df_buildings["fd"] = mm.FractalDimension(self.df_buildings).series check = ( 2 * math.log(self.df_buildings.geometry[0].length / 4) / math.log(self.df_buildings.geometry[0].area) ) assert self.df_buildings["fd"][0] == check self.df_buildings["fd2"] = mm.FractalDimension( self.df_buildings, areas=self.df_buildings.geometry.area, perimeters=self.df_buildings.geometry.length, ).series assert self.df_buildings["fd2"][0] == check def test_VolumeFacadeRatio(self): self.df_buildings["peri"] = self.df_buildings.geometry.length self.df_buildings["vfr"] = mm.VolumeFacadeRatio( self.df_buildings, "height", "volume", "peri" ).series check = self.df_buildings.volume[0] / ( self.df_buildings.peri[0] * self.df_buildings.height[0] ) assert self.df_buildings["vfr"][0] == check peri = self.df_buildings.geometry.length volume = mm.Volume(self.df_buildings, "height").series self.df_buildings["vfr2"] = mm.VolumeFacadeRatio( self.df_buildings, "height", volume, peri ).series assert self.df_buildings["vfr2"][0] == check self.df_buildings["peri"] = self.df_buildings.geometry.length self.df_buildings["vfr3"] = mm.VolumeFacadeRatio( self.df_buildings, "height" ).series assert self.df_buildings["vfr3"][0] == check def test_CircularCompactness(self): self.df_buildings["area"] = self.df_buildings.geometry.area self.df_buildings["circom"] = mm.CircularCompactness( self.df_buildings, "area" ).series check = self.df_buildings.area[0] / ( _circle_area( list(self.df_buildings.geometry[0].convex_hull.exterior.coords) ) ) assert self.df_buildings["circom"][0] == check area = self.df_buildings.geometry.area self.df_buildings["circom2"] = mm.CircularCompactness( self.df_buildings, area ).series assert self.df_buildings["circom2"][0] == check self.df_buildings["circom3"] = mm.CircularCompactness(self.df_buildings).series assert self.df_buildings["circom3"][0] == check def test_SquareCompactness(self): self.df_buildings["sqcom"] = mm.SquareCompactness(self.df_buildings).series check = ( (4 * math.sqrt(self.df_buildings.geometry.area[0])) / (self.df_buildings.geometry.length[0]) ) ** 2 assert self.df_buildings["sqcom"][0] == check self.df_buildings["sqcom2"] = mm.SquareCompactness( self.df_buildings, areas=self.df_buildings.geometry.area, perimeters=self.df_buildings.geometry.length, ).series assert self.df_buildings["sqcom2"][0] == check def test_Convexity(self): self.df_buildings["conv"] = mm.Convexity(self.df_buildings).series check = ( self.df_buildings.geometry.area[0] / self.df_buildings.geometry.convex_hull.area[0] ) assert self.df_buildings["conv"][0] == check self.df_buildings["conv2"] = mm.Convexity( self.df_buildings, areas=self.df_buildings.geometry.area ).series assert self.df_buildings["conv2"][0] == check def test_CourtyardIndex(self): cas = self.df_buildings["cas"] = mm.CourtyardArea(self.df_buildings).series self.df_buildings["cix"] = mm.CourtyardIndex(self.df_buildings, "cas").series self.df_buildings["cix_array"] = mm.CourtyardIndex( self.df_buildings, cas, self.df_buildings.geometry.area ).series check = self.df_buildings.cas[80] / self.df_buildings.geometry.area[80] assert self.df_buildings["cix"][80] == check assert self.df_buildings["cix_array"][80] == check def test_Rectangularity(self): self.df_buildings["rect"] = mm.Rectangularity(self.df_buildings).series self.df_buildings["rect_array"] = mm.Rectangularity( self.df_buildings, self.df_buildings.geometry.area ).series check = ( self.df_buildings.geometry[0].area / self.df_buildings.geometry[0].minimum_rotated_rectangle.area ) assert self.df_buildings["rect"][0] == check assert self.df_buildings["rect_array"][0] == check def test_ShapeIndex(self): la = self.df_buildings["la"] = mm.LongestAxisLength(self.df_buildings).series self.df_buildings["shape_index"] = mm.ShapeIndex(self.df_buildings, "la").series self.df_buildings["shape_index_array"] = mm.ShapeIndex( self.df_buildings, la, self.df_buildings.geometry.area ).series check = math.sqrt(self.df_buildings.area[0] / math.pi) / ( 0.5 * self.df_buildings.la[0] ) assert self.df_buildings["shape_index"][0] == check assert self.df_buildings["shape_index_array"][0] == check def test_Corners(self): self.df_buildings["corners"] = mm.Corners(self.df_buildings).series check = 24 assert self.df_buildings["corners"][0] == check def test_Squareness(self): self.df_buildings["squ"] = mm.Squareness(self.df_buildings).series check = pytest.approx(3.707, rel=1e-3) assert self.df_buildings["squ"][0] == check self.df_buildings["squ"] = mm.Squareness(self.df_buildings.exterior).series assert self.df_buildings["squ"].isna().all() df_buildings_multi = self.df_buildings.copy() df_buildings_multi["geometry"] = df_buildings_multi["geometry"].apply( lambda geom: MultiPolygon([geom]) ) self.df_buildings["squm"] = mm.Squareness(df_buildings_multi).series assert self.df_buildings["squm"][0] == check def test_EquivalentRectangularIndex(self): self.df_buildings["eri"] = mm.EquivalentRectangularIndex( self.df_buildings ).series self.df_buildings["eri_array"] = mm.EquivalentRectangularIndex( self.df_buildings, areas=self.df_buildings.geometry.area, perimeters=self.df_buildings.geometry.length, ).series check = pytest.approx(0.7879, rel=1e-3) assert self.df_buildings["eri"][0] == check assert self.df_buildings["eri_array"][0] == check def test_Elongation(self): self.df_buildings["elo"] = mm.Elongation(self.df_buildings).series check = pytest.approx(0.908, rel=1e-3) assert self.df_buildings["elo"][0] == check def test_CentroidCorners(self): self.df_buildings.loc[144] = [145, Point(0, 0).buffer(10), 0, 0] self.df_buildings.loc[145] = [ 145, Polygon([s + (0,) for s in Point(0, 0).buffer(10).exterior.coords]), 0, 0, ] check = pytest.approx(15.961, rel=1e-3) check_devs = pytest.approx(3.081, rel=1e-3) cc = mm.CentroidCorners(self.df_buildings) self.df_buildings["ccd"] = cc.mean self.df_buildings["ccddev"] = cc.std assert self.df_buildings["ccd"][0] == check assert self.df_buildings["ccddev"][0] == check_devs df_buildings_multi = self.df_buildings.copy() df_buildings_multi["geometry"] = df_buildings_multi["geometry"].apply( lambda geom: MultiPolygon([geom]) ) cc = mm.CentroidCorners(df_buildings_multi) df_buildings_multi["ccd"] = cc.mean df_buildings_multi["ccddev"] = cc.std assert df_buildings_multi["ccd"][0] == check assert df_buildings_multi["ccddev"][0] == check_devs def test_Linearity(self): self.df_streets["lin"] = mm.Linearity(self.df_streets).series euclidean = Point(self.df_streets.geometry[0].coords[0]).distance( Point(self.df_streets.geometry[0].coords[-1]) ) check = euclidean / self.df_streets.geometry[0].length assert self.df_streets["lin"][0] == pytest.approx(check, rel=1e-6) self.df_streets.loc[len(self.df_streets)] = MultiLineString( [[(0, 0), (-1, 1)], [(10, 10), (11, 11)]] ) def test_CompactnessWeightedAxis(self): self.df_buildings["cwa"] = mm.CompactnessWeightedAxis(self.df_buildings).series self.df_buildings["cwa_array"] = mm.CompactnessWeightedAxis( self.df_buildings, areas=self.df_buildings.geometry.area, perimeters=self.df_buildings.geometry.length, longest_axis=mm.LongestAxisLength(self.df_buildings).series, ).series check = pytest.approx(26.327, rel=1e-3) assert self.df_buildings["cwa"][0] == check assert self.df_buildings["cwa_array"][0] == check def test__circle_area(self): poly = Polygon([(0, 1, 0), (1, 1, 0), (2, 4, 0)]) check = _circle_area(poly.exterior.coords) assert check == pytest.approx(10.210, rel=1e-3)