expose encoding with default utf-8 for reading xml files

CHANGELOG.md

@@ -3,6 +3,7 @@
 ## Unreleased
 
 - formally support Python 3.12
+- fix Windows-specific character encoding issues when reading XML files
 
 ## 1.7.1 (2023-10-29)
 

osmnx/features.py

@@ -336,7 +336,7 @@ def features_from_polygon(polygon, tags):
     return _create_gdf(response_jsons, polygon, tags)
 
 
-def features_from_xml(filepath, polygon=None, tags=None):
+def features_from_xml(filepath, polygon=None, tags=None, encoding="utf-8"):
     """
     Create a GeoDataFrame of OSM features in an OSM-formatted XML file.
 
@@ -367,13 +367,15 @@ def features_from_xml(filepath, polygon=None, tags=None):
         the area. `tags = {'amenity':True, 'landuse':['retail','commercial'],
         'highway':'bus_stop'}` would return all amenities, landuse=retail,
         landuse=commercial, and highway=bus_stop.
+    encoding : string
+        the XML file's character encoding
 
     Returns
     -------
     gdf : geopandas.GeoDataFrame
     """
     # transmogrify file of OSM XML data into JSON
-    response_jsons = [osm_xml._overpass_json_from_file(filepath)]
+    response_jsons = [osm_xml._overpass_json_from_file(filepath, encoding)]
 
     # create GeoDataFrame using this response JSON
     return _create_gdf(response_jsons, polygon=polygon, tags=tags)

osmnx/graph.py

@@ -530,7 +530,9 @@ def graph_from_polygon(
     return G
 
 
-def graph_from_xml(filepath, bidirectional=False, simplify=True, retain_all=False):
+def graph_from_xml(
+    filepath, bidirectional=False, simplify=True, retain_all=False, encoding="utf-8"
+):
     """
     Create a graph from data in a .osm formatted XML file.
 
@@ -550,13 +552,15 @@ def graph_from_xml(filepath, bidirectional=False, simplify=True, retain_all=Fals
     retain_all : bool
         if True, return the entire graph even if it is not connected.
         otherwise, retain only the largest weakly connected component.
+    encoding : string
+        the XML file's character encoding
 
     Returns
     -------
     G : networkx.MultiDiGraph
     """
     # transmogrify file of OSM XML data into JSON
-    response_jsons = [osm_xml._overpass_json_from_file(filepath)]
+    response_jsons = [osm_xml._overpass_json_from_file(filepath, encoding)]
 
     # create graph using this response JSON
     G = _create_graph(response_jsons, bidirectional=bidirectional, retain_all=retain_all)

osmnx/osm_xml.py

@@ -61,30 +61,32 @@ def endElement(self, name):
             self.object["elements"].append(self._element)
 
 
-def _overpass_json_from_file(filepath):
+def _overpass_json_from_file(filepath, encoding):
     """
     Read OSM XML from file and return Overpass-like JSON.
 
     Parameters
     ----------
     filepath : string or pathlib.Path
         path to file containing OSM XML data
+    encoding : string
+        the XML file's character encoding
 
     Returns
     -------
     OSMContentHandler object
     """
 
     # open the XML file, handling bz2 or regular XML
-    def _opener(filepath):
+    def _opener(filepath, encoding):
         if filepath.suffix == ".bz2":
-            return bz2.BZ2File(filepath)
+            return bz2.open(filepath, mode="rt", encoding=encoding)
 
         # otherwise just open it if it's not bz2
-        return filepath.open()
+        return filepath.open(encoding=encoding)
 
     # warn if this XML file was generated by OSMnx itself
-    with _opener(Path(filepath)) as f:
+    with _opener(Path(filepath), encoding) as f:
         root_attrs = ET.parse(f).getroot().attrib
         if "generator" in root_attrs and "OSMnx" in root_attrs["generator"]:
             warn(
@@ -97,7 +99,7 @@ def _opener(filepath):
             )
 
     # parse the XML to Overpass-like JSON
-    with _opener(Path(filepath)) as f:
+    with _opener(Path(filepath), encoding) as f:
         handler = _OSMContentHandler()
         xml.sax.parse(f, handler)
         return handler.object
@@ -187,7 +189,7 @@ def _save_graph_xml(
     precision=6,
 ):
     """
-    Save graph to disk as an OSM-formatted XML .osm file.
+    Save graph to disk as an OSM-formatted UTF-8 encoded XML .osm file.
 
     Parameters
     ----------