Skip to content

mamut_routing_tools.roadgraph.osmxml

Streaming OSM XML parsing for the road engine.

Reads only what the road graph needs: node coordinates, ways with their tags and node lists, and the <bounds> element. City extracts run to ~150 MB, so parsing is incremental with element recycling.

ensure_bounds 

ensure_bounds(osm_path: Path) -> None

Inject a <bounds> element computed from node coordinates when the extract ships without one (some Overpass responses omit it). Matches the Julia pipeline's behavior so both read identical files afterwards.

Source code in MAMUT-routing-tools/src/mamut_routing_tools/roadgraph/osmxml.py 
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
def ensure_bounds(osm_path: Path) -> None:
    """Inject a ``<bounds>`` element computed from node coordinates when the
    extract ships without one (some Overpass responses omit it). Matches the
    Julia pipeline's behavior so both read identical files afterwards."""
    with osm_path.open("r", encoding="utf-8") as handle:
        head = handle.read(64 * 1024)
    if re.search(r"<bounds\b", head):
        return
    min_lat = min_lon = float("inf")
    max_lat = max_lon = float("-inf")
    count = 0
    for _, elem in ET.iterparse(osm_path, events=("start",)):
        if elem.tag == "node":
            lat = float(elem.attrib["lat"])
            lon = float(elem.attrib["lon"])
            min_lat, max_lat = min(min_lat, lat), max(max_lat, lat)
            min_lon, max_lon = min(min_lon, lon), max(max_lon, lon)
            count += 1
        elem.clear()
    if count == 0:
        raise ValueError(f"OSM file '{osm_path}' contains no nodes; cannot derive bounds")
    bounds_line = (
        f'  <bounds minlat="{min_lat}" minlon="{min_lon}" maxlat="{max_lat}" maxlon="{max_lon}"/>\n'
    )
    text = osm_path.read_text(encoding="utf-8")
    match = re.search(r"<osm\b[^>]*>\n?", text)
    if match is None:
        raise ValueError(f"OSM file '{osm_path}' has no <osm> root element")
    insert_at = match.end()
    osm_path.write_text(text[:insert_at] + bounds_line + text[insert_at:], encoding="utf-8")

crop_to_bounds 

crop_to_bounds(osm_data: OsmData) -> None

Crop ways and nodes to the <bounds> rectangle, as get_map_data does before graph construction. Mutates in place.

Source code in MAMUT-routing-tools/src/mamut_routing_tools/roadgraph/osmxml.py 
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
def crop_to_bounds(osm_data: OsmData) -> None:
    """Crop ways and nodes to the ``<bounds>`` rectangle, as
    ``get_map_data`` does before graph construction. Mutates in place."""
    min_lat, min_lon, max_lat, max_lon = osm_data.bounds
    if min_lon > max_lon:
        raise ValueError("Antimeridian-crossing bounds are not supported")
    next_id = max(osm_data.nodes.keys(), default=0) + 1

    def allocate_node_id() -> int:
        nonlocal next_id
        value = next_id
        next_id += 1
        return value

    osm_data.ways = [
        way for way in osm_data.ways if not _crop_way(osm_data.nodes, osm_data.bounds, way, allocate_node_id)
    ]
    osm_data.nodes = {
        node_id: point for node_id, point in osm_data.nodes.items() if _inbounds(point, osm_data.bounds)
    }