def polygon_2_raster(polygon: gpd.GeoDataFrame, rst_fn: str, out_fn: str,
out_shape: List[int]) -> None:
'''This function take in a geopandas object of polygons and converts them into a raster
image based on the 'temp_rast' raster
Input:
polygon -- the geopandas object of polygons
rst_fn -- the filename of the raster that will be the template upon which the polygons will be burned
out_fn -- the output file name that the raster of the polygons will be written to
out_shape -- a two element list of the x and y shape of the output raster image
Output:
'''
with rio.open(rst_fn, 'r') as rst:
meta = rst.meta.copy()
meta.update(compress='lzw')
with rio.open(out_fn, 'w', **meta) as out:
#we create a generator of geom, value pairs to use in the features.rasterize function
shapes = ((feature['geometry'], 1)
for feature in polygon.iterfeatures())
labels = features.rasterize(shapes=shapes,
out_shape=out_shape,
transform=meta['transform'],
fill=0,
dtype='uint16',
all_touched=False)
out.write(labels, indexes=1)
def test_geodataframe_iterfeatures(self):
df = self.df.iloc[:1].copy()
df.loc[0, "BoroName"] = np.nan
# when containing missing values
# null: ouput the missing entries as JSON null
result = list(df.iterfeatures(na="null"))[0]["properties"]
assert result["BoroName"] is None
# drop: remove the property from the feature.
result = list(df.iterfeatures(na="drop"))[0]["properties"]
assert "BoroName" not in result.keys()
# keep: output the missing entries as NaN
result = list(df.iterfeatures(na="keep"))[0]["properties"]
assert np.isnan(result["BoroName"])
# test for checking that the (non-null) features are python scalars and
# not numpy scalars
assert type(df.loc[0, "Shape_Leng"]) is np.float64
# null
result = list(df.iterfeatures(na="null"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
# drop
result = list(df.iterfeatures(na="drop"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
# keep
result = list(df.iterfeatures(na="keep"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
# when only having numerical columns
df_only_numerical_cols = df[["Shape_Leng", "Shape_Area", "geometry"]]
assert type(df_only_numerical_cols.loc[0, "Shape_Leng"]) is np.float64
# null
result = list(df_only_numerical_cols.iterfeatures(na="null"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
# drop
result = list(df_only_numerical_cols.iterfeatures(na="drop"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
# keep
result = list(df_only_numerical_cols.iterfeatures(na="keep"))[0]
assert type(result["properties"]["Shape_Leng"]) is float
with pytest.raises(
ValueError,
match="GeoDataFrame cannot contain duplicated column names."):
df_with_duplicate_columns = df[[
"Shape_Leng", "Shape_Leng", "Shape_Area", "geometry"
]]
list(df_with_duplicate_columns.iterfeatures())
# geometry not set
df = GeoDataFrame({
"values": [0, 1],
"geom": [Point(0, 1), Point(1, 0)]
})
with pytest.raises(AttributeError):
list(df.iterfeatures())
