Issue #117: links_on_spatial_condition failing for large spatial area #126
Conversation
There was a problem hiding this comment.
A lot of code duplication happening here, I don't think you need to change anything in
Line 632 in 4fcc3d3
Now that we're not relying on these intersects and within methods, it would be cleaner and more general to get rid of the if statements and open up the predicate param: https://geopandas.org/en/stable/docs/reference/api/geopandas.sindex.SpatialIndex.valid_query_predicates.html so the user can choose from the geopandas list of operations and our code would simplify to:
def _find_ids_on_shapely_geometry(self, gdf, predicate, shapely_input):
geojson_input_gdf = ...
return list(gpd.sjoin(gdf, geojson_input_gdf, how='inner', predicate=predicate)['id'])|
|
||
| def _find_ids_on_shapely_geometry(self, gdf, how, shapely_input): | ||
| if how == 'intersect': | ||
| return list(gdf[gdf.intersects(shapely_input)]['id']) |
There was a problem hiding this comment.
as far as I understand that's the only line you need to change, the rest of the code falls in place
| if how == 'within': | ||
| geojson_input_gdf = gpd.GeoDataFrame(index=[0], crs=gdf.crs, geometry=[shapely_input]) | ||
| return list(gpd.sjoin(gdf, geojson_input_gdf, how='inner', op='intersects')['id']) | ||
| elif how == 'within': | ||
| return list(gdf[gdf.within(shapely_input)]['id']) |
There was a problem hiding this comment.
I assume this is at risk too and should be updated similarly to intersection?
| return list(gdf[gdf.intersects(shapely_input)]['id']) | ||
| if how == 'within': | ||
| geojson_input_gdf = gpd.GeoDataFrame(index=[0], crs=gdf.crs, geometry=[shapely_input]) | ||
| return list(gpd.sjoin(gdf, geojson_input_gdf, how='inner', op='intersects')['id']) |
There was a problem hiding this comment.
op is deprecated, predicate should be used instead: https://geopandas.org/en/stable/docs/reference/api/geopandas.sjoin.html
Addresses issue #117 :
Using the function links_on_spatial_condition with a geojson input results in an error due to a memory error:
The geojson contained a polygon which covered multiple cities. Possible to work around it by using
gpd.sjoin(network_gdf,urban_gdf,how='inner', op='intersects'). It might be worth changing the function links_on_spatial_condition to usegpd.sjoininstead ofgdf.intersects.