Skip to content

Commit

Permalink
Merge pull request #57 from IMMM-SFA/test/nodes
Browse files Browse the repository at this point in the history 
Test/nodes
  • Loading branch information
@levisweetbreu
levisweetbreu authored Jul 14, 2023
2 parents dc5f7a1 + 8bbdbc5 commit 8a300b5
Showing 1 changed file with 313 additions and 2 deletions.
315 changes: 313 additions & 2 deletions tests/test_nodes.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
import numpy as np
import geopandas as gpd
import pandas as pd
from shapely.geometry import Point, Polygon
from shapely.geometry import Point, Polygon, JOIN_STYLE
from typing import List


Expand Down Expand Up @@ -319,13 +319,324 @@ def test_average_distance_between_buildings(self):
default_street_width = Settings.DEFAULT_STREET_WIDTH

expected = pd.DataFrame(
{"id": ids, "distance": pd.Series([1.5, 2.0, 1.0, 3.0, default_street_width])}
{
Settings.id_field: ids,
Settings.average_distance_between_buildings: pd.Series(
[1.5, 2.0, 1.0, 3.0, default_street_width]
),
}
)

actual = nodes.average_distance_between_buildings(ids, distance)

pd.testing.assert_frame_equal(expected, actual)

def test_buildings_intersecting_plan_area(self):
"""Test that the function buildings_intersecting_plan_area returns the correct intersecting buildings."""

polygon1 = Polygon([[0, 0], [0, 1], [1, 1], [1, 0]])
polygon2 = Polygon([[3, 3], [3, 4], [4, 4], [4, 3]])
building_id = pd.Series([0, 1])
building_height = pd.Series([5, 10])
building_geometry = pd.Series([polygon1, polygon2])
building_area = pd.Series([polygon1.area, polygon2.area])
crs = "epsg:3857"

total_plan_area_geometry_no_overlap = pd.Series([polygon1.buffer(1), polygon2.buffer(1)])
total_plan_area_geometry_some_overlap = pd.Series(
[polygon1.buffer(3.5), polygon2.buffer(3.5)]
)
total_plan_area_geometry_total_overlap = pd.Series([polygon1.buffer(5), polygon2.buffer(5)])

no_overlap_output_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": building_id,
"building_id_target": building_id,
"building_height_target": building_height,
"building_area_target": building_area,
"building_geometry": building_geometry,
"building_buffered_target": gpd.GeoSeries(total_plan_area_geometry_no_overlap),
"index_neighbor": building_id,
"building_height_neighbor": building_height,
"building_area_neighbor": building_area,
"building_buffered_neighbor": total_plan_area_geometry_no_overlap,
"building_geometry_neighbor": gpd.GeoSeries(building_geometry),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

some_overlap_output_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": pd.Series([0, 0, 1, 1]),
"building_id_target": pd.Series([0, 1, 0, 1]),
"building_height_target": pd.Series([5, 10, 5, 10]),
"building_area_target": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_geometry": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[1],
building_geometry[0],
building_geometry[1],
]
),
"building_buffered_target": gpd.GeoSeries(
[
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
]
),
"index_neighbor": pd.Series([0, 0, 1, 1]),
"building_height_neighbor": pd.Series([5, 5, 10, 10]),
"building_area_neighbor": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_buffered_neighbor": pd.Series(
[
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
total_plan_area_geometry_some_overlap[1],
]
),
"building_geometry_neighbor": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[0],
building_geometry[1],
building_geometry[1],
]
),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

total_overlap_output_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": pd.Series([0, 0, 1, 1]),
"building_id_target": pd.Series([0, 1, 0, 1]),
"building_height_target": pd.Series([5, 10, 5, 10]),
"building_area_target": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_geometry": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[1],
building_geometry[0],
building_geometry[1],
]
),
"building_buffered_target": gpd.GeoSeries(
[
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
]
),
"index_neighbor": pd.Series([0, 0, 1, 1]),
"building_height_neighbor": pd.Series([5, 5, 10, 10]),
"building_area_neighbor": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_buffered_neighbor": pd.Series(
[
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
total_plan_area_geometry_total_overlap[1],
]
),
"building_geometry_neighbor": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[0],
building_geometry[1],
building_geometry[1],
]
),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

@dataclass
class TestCase:
name: str
input: pd.Series
expected: gpd.GeoDataFrame

testcases = [
TestCase(
name="no overlap",
input=total_plan_area_geometry_no_overlap,
expected=no_overlap_output_gdf,
),
TestCase(
name="some overlap",
input=total_plan_area_geometry_some_overlap,
expected=some_overlap_output_gdf,
),
TestCase(
name="total overlap",
input=total_plan_area_geometry_total_overlap,
expected=total_overlap_output_gdf,
),
]

for case in testcases:
actual = nodes.buildings_intersecting_plan_area(
building_id, building_height, building_geometry, building_area, case.input, crs
)
expected = case.expected
pd.testing.assert_frame_equal(
expected,
actual,
"failed test {} expected {}, actual {}".format(case.name, expected, actual),
)

def test_building_plan_area(self):
"""Test that the function building_plan_area returns the correct building plan area."""

polygon1 = Polygon([[0, 0], [0, 1], [1, 1], [1, 0]])
polygon2 = Polygon([[3, 3], [3, 4], [4, 4], [4, 3]])
building_id = pd.Series([0, 1])
building_height = pd.Series([5, 10])
building_geometry = pd.Series([polygon1, polygon2])
building_area = pd.Series([polygon1.area, polygon2.area])

total_plan_area_geometry_no_overlap = pd.Series([polygon1.buffer(1), polygon2.buffer(1)])
total_plan_area_geometry_some_overlap = pd.Series(
[
polygon1.buffer(2.5, join_style=JOIN_STYLE.mitre),
polygon2.buffer(2.5, join_style=JOIN_STYLE.mitre),
]
)
total_plan_area_geometry_total_overlap = pd.Series([polygon1.buffer(5), polygon2.buffer(5)])

no_overlap_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": building_id,
"building_id_target": building_id,
"building_height_target": building_height,
"building_area_target": building_area,
"building_geometry": building_geometry,
"building_buffered_target": gpd.GeoSeries(total_plan_area_geometry_no_overlap),
"index_neighbor": building_id,
"building_height_neighbor": building_height,
"building_area_neighbor": building_area,
"building_buffered_neighbor": total_plan_area_geometry_no_overlap,
"building_geometry_neighbor": gpd.GeoSeries(building_geometry),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

some_overlap_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": pd.Series([0, 0, 1, 1]),
"building_id_target": pd.Series([0, 1, 0, 1]),
"building_height_target": pd.Series([5, 10, 5, 10]),
"building_area_target": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_geometry": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[1],
building_geometry[0],
building_geometry[1],
]
),
"building_buffered_target": gpd.GeoSeries(
[
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
]
),
"index_neighbor": pd.Series([0, 0, 1, 1]),
"building_height_neighbor": pd.Series([5, 5, 10, 10]),
"building_area_neighbor": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_buffered_neighbor": pd.Series(
[
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[0],
total_plan_area_geometry_some_overlap[1],
total_plan_area_geometry_some_overlap[1],
]
),
"building_geometry_neighbor": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[0],
building_geometry[1],
building_geometry[1],
]
),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

total_overlap_gdf = gpd.GeoDataFrame(
{
"building_id_neighbor": pd.Series([0, 0, 1, 1]),
"building_id_target": pd.Series([0, 1, 0, 1]),
"building_height_target": pd.Series([5, 10, 5, 10]),
"building_area_target": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_geometry": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[1],
building_geometry[0],
building_geometry[1],
]
),
"building_buffered_target": gpd.GeoSeries(
[
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
]
),
"index_neighbor": pd.Series([0, 0, 1, 1]),
"building_height_neighbor": pd.Series([5, 5, 10, 10]),
"building_area_neighbor": pd.Series([1.0, 1.0, 1.0, 1.0]),
"building_buffered_neighbor": pd.Series(
[
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[0],
total_plan_area_geometry_total_overlap[1],
total_plan_area_geometry_total_overlap[1],
]
),
"building_geometry_neighbor": gpd.GeoSeries(
[
building_geometry[0],
building_geometry[0],
building_geometry[1],
building_geometry[1],
]
),
},
geometry="building_geometry",
).set_index("building_id_neighbor")

@dataclass
class TestCase:
name: str
input: gpd.GeoDataFrame
expected: List[float]

testcases = [
TestCase(name="no overlap", input=no_overlap_gdf, expected=[1.0, 1.0]),
TestCase(name="some overlap", input=some_overlap_gdf, expected=[1.25, 1.25]),
TestCase(name="total overlap", input=total_overlap_gdf, expected=[2.0, 2.0]),
]

for case in testcases:
actual = nodes.building_plan_area(case.input)
expected = pd.Series(case.expected)
pd.testing.assert_series_equal(
expected,
actual,
"failed test {} expected {}, actual {}".format(case.name, expected, actual),
)


if __name__ == "__main__":
unittest.main()

0 comments on commit 8a300b5

Please sign in to comment.