def test_fractional_no_round(self):
''' Test shared for a fractional aggregation where we do not round'''
df_target, df_source = load_dfs('target.shp', 'source_shared.shp')
source_cols = ['dist_col']
df = distribute_values(df_source, source_cols, df_target)
left = 0.9 + 0.7 + 0.55 + 0.4 + 0.2
assert round(df.at[0, 'dist_col'], 2) == round(left, 2)
assert round(df.at[1, 'dist_col'], 2) == round(5 - left, 2)
def test_custom_target_cols(self):
''' Check that the custom naming columns in target_cols list works '''
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
source_cols = ['dist_col']
target_cols = ['new_col']
df = distribute_values(df_source, source_cols, df_target, target_cols)
# Check that the proper aggregation occurred
assert df.at[0, 'new_col'] == 1
assert df.at[1, 'new_col'] == 0
def test_leftovers(self):
''' Test that the leftover values (i.e. area that does not intersect
with any geometry) gets assigned to target geometry with the
greatest area'''
df_target, df_source = load_dfs('target.shp', 'source_leftover.shp')
source_cols = ['dist_col']
df = distribute_values(df_source, source_cols, df_target)
assert round(df.at[0, 'dist_col'], 2) == round(5 / 6, 2)
assert round(df.at[1, 'dist_col'], 2) == round(1 / 6, 2)
def test_keep_existing_cols(self):
''' Ensure that aggregation function doesn't delete existing columns in
target_df'''
df_target, df_source = load_dfs('keep.shp', 'source_centroid.shp')
source_cols = ['dist_col']
df = distribute_values(df_source, source_cols, df_target)
# Check that the proper aggregation occurred
assert df.at[0, 'keep'] == 'keep'
assert df.at[1, 'keep'] == 'keep'
def test_delete_old_cols(self):
''' Check that aggregation function replaces existing columns with the
same name'''
df_target, df_source = load_dfs('reset.shp', 'source_centroid.shp')
source_cols = ['dist_col']
df = distribute_values(df_source, source_cols, df_target)
# Check that the proper aggregation occurred
assert df.at[0, 'dist_col'] == 1
assert df.at[1, 'dist_col'] == 0
def test_centroid(self):
''' Check when aggregating based on centroid. (i.e. no intersection)'''
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
source_cols = ['dist_col']
# get aggregation
df = distribute_values(df_source, source_cols, df_target)
# Value gets assigned to the proper target geometry (left)
assert df.at[0, 'dist_col'] == 1
assert df.at[1, 'dist_col'] == 0
def test_fractional_other_aggregation_attribute(self):
''' Test aggregating on the dummy column 'col_dist_on' rather than area
'''
df_target, df_source = load_dfs('target.shp', 'source_shared.shp')
source_cols = ['dist_col']
df = distribute_values(df_source,
source_cols,
df_target,
distribute_on='dist_on')
assert round(df.at[0, 'dist_col'], 2) == round(10 / 3, 2)
assert round(df.at[1, 'dist_col'], 2) == round(5 / 3, 2)
def test_shared_winnner_take_all(self):
'''Test shared winner take all'''
df_target, df_source = load_dfs('target.shp', 'source_shared.shp')
source_cols = ['dist_col']
df = distribute_values(df_source,
source_cols,
df_target,
distribute_type='winner take all')
# Check proper aggregation
assert df.at[0, 'dist_col'] == 3
assert df.at[1, 'dist_col'] == 2
def test_within_distribute_values(self):
''' Check aggregation intersection is with only one target geometry'''
df_target, df_source = load_dfs('target.shp', 'source_within.shp')
source_cols = ['dist_col']
# Apply aggregation
df = distribute_values(df_source, source_cols, df_target)
# Check that the proper aggregation occurred
assert df.at[0, 'dist_col'] == 2
assert df.at[1, 'dist_col'] == 1
def test_fractional_round(self):
''' Test shared for a fractional aggregation where we round'''
df_target, df_source = load_dfs('target.shp', 'source_shared.shp')
source_cols = ['dist_col']
df = distribute_values(df_source,
source_cols,
df_target,
distribute_round=True)
left = 0.9 + 0.7 + 0.55 + 0.4 + 0.2
assert df.at[0, 'dist_col'] == 3
assert df.at[1, 'dist_col'] == 2
def test_aggregation_attribute(self):
''' Check that false is returned when aggregating attribute for df_target
is not actually an attribute of df_target'''
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
source_cols = df_source.columns
# enter with incorrect aggregation attribute
df = distribute_values(df_source,
source_cols,
df_target,
distribute_on='other')
assert df == False
def test_attribute_subset(self):
''' Check that False is returned when the an element of source is
not in the dataframe itself'''
# load dataframe
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
# Get same length column sizes
source_cols = list(df_source.columns)
source_cols = [x + 'foo' for x in source_cols]
df = distribute_values(df_source, source_cols, df_target)
assert df == False
def test_attribute_list_size_comparison(self):
''' Check that False is returned when the size of attribute lists
are not of the same length'''
# load
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
# Get different length column sizes
target_cols = list(df_target.columns)
source_cols = ['a']
df = distribute_values(df_source, source_cols, df_target, target_cols)
assert df == False
def test_multiple_columns(self):
''' Check aggregation for multiple columns being aggregated'''
df_target, df_source = load_dfs('target.shp', 'source_within.shp')
source_cols = ['dist_col', 'other_col']
# Apply aggregation
df = distribute_values(df_source, source_cols, df_target)
# Check that the proper aggregation occurred
assert df.at[0, 'dist_col'] == 2
assert df.at[1, 'dist_col'] == 1
assert df.at[0, 'other_col'] == 4
assert df.at[1, 'other_col'] == 3
def test_aggregegation_type(self):
''' Check that false is returned when 'fractional' or 'winner take all'
is not entered into the distribute_type argument'''
# load dataframe
df_target, df_source = load_dfs('target.shp', 'source_centroid.shp')
source_cols = df_source.columns
# enter with incorrect aggregation type
df = distribute_values(df_source,
source_cols,
df_target,
distribute_type='other')
assert df == False
