Merge pull request #186 from Deltares/feat/180-update-geo-dependencies

Feat/180 update geo dependencies

.config/docker_environment.yml

@@ -5,11 +5,9 @@ channels:
 dependencies:
   - conda-forge::python=3.10
   - conda-forge::gdal=3.5.1
-  - conda-forge::fiona
   # from here some dragons 
   - conda-forge::geopandas
   - conda-forge::rasterio
-  - conda-forge::geojson
   - conda-forge::osmnx
   - conda-forge::tqdm
   - conda-forge::geopy

.config/environment.yml

@@ -5,7 +5,6 @@ channels:
 dependencies:
   - conda-forge::python=3.10
   - conda-forge::gdal=3.5.1
-  - conda-forge::fiona
   # Install poetry 1.3.2 (1.4.*) gives issues.
   # If this does not work then exclude it from the environment and install it manually
   # with: conda install -c conda-forge poetry=1.3.2

.config/tc_environment.yml

@@ -5,5 +5,4 @@ channels:
 dependencies:
   - python=3.10
   - conda-forge::gdal=3.5.1
-  - conda-forge::fiona
   - teamcity-messages

pyproject.toml

@@ -29,8 +29,6 @@ run_ra2ce = "ra2ce.run:main"
 python = "^3.9, <3.11"
 affine = "^2.3.1"
 click = "^8.1.3"
-geojson = "^2.5.0"
-geopy = "^2.2.0"
 numpy = "^1.23.3"
 networkx = "^2.8.6"
 pandas = "^1.4.4"
@@ -39,18 +37,18 @@ rasterio = "^1.3.2"
 rasterstats = "^0.17.0"
 Rtree = "^1.0.0"
 seaborn = "^0.12.0"
-Shapely = "^1.8.4"
-tqdm = "^4.64.1"
 openpyxl = "^3.0.10"
 xlrd = "^2.0.1"
 xarray = "^2022.6.0"
-geopandas = "^0.11.1"
-osmnx = "0.16.*"
 scipy = "^1.9.1"
 GDAL = "3.5.1"
-fiona = "1.8.21"
 pygeos = "^0.14"
 momepy = "0.5.0"
+geopandas = "^0.14.0"
+shapely = "^2.0.1"
+osmnx = "^1.6.0"
+tqdm = "^4.66.1"
+geopy = "^2.4.0"
 pyogrio = "^0.6.0"
 joblib = "^1.3.2"
 

ra2ce/analyses/direct/cost_benefit_analysis.py

@@ -22,6 +22,7 @@
 
 import logging
 import os
+from pathlib import Path
 
 import geopandas as gpd
 import numpy as np
@@ -71,17 +72,17 @@ def _validate_input_params(self, analysis: dict, config: dict) -> None:
             )
 
     @staticmethod
-    def load_effectiveness_table(path):
+    def load_effectiveness_table(path: Path):
         """This function loads a CSV table containing effectiveness of the different aspects for a number of strategies"""
         file_path = path / "effectiveness_measures.csv"
         df_lookup = pd.read_csv(file_path, index_col="strategies")
         return df_lookup.transpose().to_dict()
 
     @staticmethod
-    def create_feature_table(file_path):
+    def create_feature_table(file_path: Path):
         """This function loads a table of features from the input folder"""
         logging.info("Loading feature dataframe...")
-        gdf = gpd.read_file(file_path)
+        gdf = gpd.read_file(file_path, engine="pyogrio")
         logging.info("Dataframe loaded...")
 
         # cleaning up dataframe

ra2ce/analyses/indirect/analyses_indirect.py

@@ -22,7 +22,6 @@
 
 import copy
 import logging
-import sys
 import time
 from pathlib import Path
 from typing import List, Tuple
@@ -536,8 +535,8 @@ def extract_od_nodes_from_graph(
     def _get_origin_destination_pairs(
         self, graph: nx.classes.MultiGraph
     ) -> list[tuple[int, str], tuple[int, str]]:
-        od_path = (
-            self.config["static"] / "output_graph" / "origin_destination_table.feather"
+        od_path = self.config["static"].joinpath(
+            "output_graph", "origin_destination_table.feather"
         )
         od = gpd.read_feather(od_path)
         od_pairs = [
@@ -782,7 +781,7 @@ def multi_link_origin_destination_regional_impact(self, gdf_ori):
         origin_fn = Path(self.config["static"]).joinpath(
             "output_graph", "origin_destination_table.gpkg"
         )
-        origin = gpd.read_file(origin_fn)
+        origin = gpd.read_file(origin_fn, engine="pyogrio")
         index = [type(x) == str for x in origin["o_id"]]
         origin = origin[index]
         origin.reset_index(inplace=True, drop=True)
@@ -983,15 +982,12 @@ def get_network_with_edge_fid(self, graph: nx.Graph) -> gpd.GeoDataFrame:
         """
         network = graph_to_gdf(graph)[0]
         # TODO: add making "edges_fid" (internal convention) to graph_to_gdf
-        if all(
-            c_idx in network.columns for c_idx in ["node_A", "node_B"]
-        ):  # shapefiles
+        if all(c_idx in network.index.names for c_idx in ["u", "v"]):
+            network["edge_fid"] = [f"{na}_{nb}" for (na, nb, _) in network.index]
+        elif all(c_idx in network.columns for c_idx in ["node_A", "node_B"]):
+            # shapefiles
             network["edge_fid"] = [
-                f"{na}_{nb}" for na, nb in network[["node_A", "node_B"]].values
-            ]
-        elif all(c_idx in network.columns for c_idx in ["u", "v"]):  # osm
-            network["edge_fid"] = [
-                f"{na}_{nb}" for na, nb in network[["u", "v"]].values
+                f"{na}_{nb}" for na, nb in network["node_A", "node_B"].values
             ]
         return network[["edge_fid", "geometry"]]
 
@@ -1032,7 +1028,7 @@ def execute(self):
             starttime = time.time()
             gdf = pd.DataFrame()
             opt_routes = None
-            output_path = self.config["output"] / analysis["analysis"]
+            output_path = self.config["output"].joinpath(analysis["analysis"])
 
             def _save_gpkg_analysis(
                 base_graph,
@@ -1041,16 +1037,16 @@ def _save_gpkg_analysis(
             ):
                 for to_save, save_name in zip(to_save_gdf, to_save_gdf_names):
                     if not to_save.empty:
-                        gpkg_path = output_path / (
+                        gpkg_path = output_path.joinpath(
                             analysis["name"].replace(" ", "_") + f"_{save_name}.gpkg"
                         )
                         save_gdf(to_save, gpkg_path)
 
                 # Save the Graph
-                gpkg_path_nodes = output_path / (
+                gpkg_path_nodes = output_path.joinpath(
                     analysis["name"].replace(" ", "_") + "_results_nodes.gpkg"
                 )
-                gpkg_path_edges = output_path / (
+                gpkg_path_edges = output_path.joinpath(
                     analysis["name"].replace(" ", "_") + "_results_edges.gpkg"
                 )
                 graph_to_gpkg(base_graph, gpkg_path_edges, gpkg_path_nodes)

ra2ce/configuration/config_wrapper.py

@@ -24,7 +24,6 @@
 
 import logging
 from pathlib import Path
-from typing import Optional
 
 from ra2ce.analyses.analysis_config_wrapper.analysis_config_wrapper_base import (
     AnalysisConfigWrapperBase,

ra2ce/graph/hazard/hazard_overlay.py

@@ -766,7 +766,7 @@ def create(self):
         #### Step 4: hazard overlay of the locations that are checked for isolation ###
         if self._isolation_locations:
             logging.info("Detected isolated locations, checking for hazard overlay.")
-            locations = gpd.read_file(self._isolation_locations)
+            locations = gpd.read_file(self._isolation_locations, engine="pyogrio")
             locations["i_id"] = locations.index
             locations_crs = pyproj.CRS.from_user_input(locations.crs)
             hazard_crs = pyproj.CRS.from_user_input(self._hazard_crs)

ra2ce/graph/network_wrappers/osm_network_wrapper/osm_network_wrapper.py

@@ -21,11 +21,12 @@
 import networkx as nx
 import osmnx
 import pandas as pd
-from geopandas import GeoDataFrame
+from geopandas import GeoDataFrame, read_file
 from networkx import MultiDiGraph, MultiGraph
 from shapely.geometry.base import BaseGeometry
 
 import ra2ce.graph.networks_utils as nut
+from shapely.geometry import shape
 from ra2ce.graph.exporters.json_exporter import JsonExporter
 from ra2ce.graph.network_config_data.network_config_data import NetworkConfigData
 from ra2ce.graph.network_wrappers.network_wrapper_protocol import NetworkWrapperProtocol
@@ -133,9 +134,9 @@ def get_clean_graph_from_osm(self) -> MultiDiGraph:
                 "No polygon_file file found at {}.".format(self.polygon_path)
             )
 
-        poly_dict = nut.read_geojson(geojson_file=self.polygon_path)
+        _normalized_polygon = nut.get_normalized_geojson_polygon(self.polygon_path)
         _complex_graph = self._download_clean_graph_from_osm(
-            polygon=nut.geojson_to_shp(poly_dict),
+            polygon=_normalized_polygon,
             network_type=self.network_type,
             road_types=self.road_types,
         )

ra2ce/graph/network_wrappers/osm_network_wrapper/osm_utils.py

@@ -44,7 +44,7 @@ def from_shapefile_to_poly(shapefile: Path, out_path: Path, outname: str = ""):
     Returns:
         .poly file for each region, in a new dir in the working directory (in the CRS of te input file)
     """
-    shp_file_gdf = gpd.read_file(str(shapefile))
+    shp_file_gdf = gpd.read_file(str(shapefile), engine="pyogrio")
 
     num = 0
     # iterate over the seperate polygons in the shapefile

ra2ce/graph/network_wrappers/shp_network_wrapper.py

@@ -68,12 +68,12 @@ def _read_merge_shp(self) -> gpd.GeoDataFrame:
             properties (pandas dataframe): attributes of shapefile(s), in order of the linestrings in lines
         """
         # concatenate all shapefile into one geodataframe and set analysis to 1 or 0 for diversions
-        lines = [gpd.read_file(shp) for shp in self.primary_files]
+        lines = [gpd.read_file(shp, engine="pyogrio") for shp in self.primary_files]
 
         if any(self.diversion_files):
             lines.extend(
                 [
-                    nut.check_crs_gdf(gpd.read_file(shp), self.crs)
+                    nut.check_crs_gdf(gpd.read_file(shp, engine="pyogrio"), self.crs)
                     for shp in self.diversion_files
                 ]
             )

ra2ce/graph/network_wrappers/vector_network_wrapper.py

@@ -110,7 +110,7 @@ def _read_files(self, file_list: list[Path]) -> gpd.GeoDataFrame:
             gpd.GeoDataFrame: GeoDataFrame representing the data.
         """
         # read file
-        gdf = gpd.GeoDataFrame(pd.concat(list(map(gpd.read_file, file_list))))
+        gdf = gpd.GeoDataFrame(pd.concat([gpd.read_file(_fl, engine="pyogrio") for _fl in file_list]))
         logging.info(
             "Read files {} into a 'GeoDataFrame'.".format(
                 ", ".join(map(str, file_list))