Geograph-based shortest distance calculation for Python. 100% in Rust. Based on the routrs rust crate.
The distance calculation functions in this library are designed to work with various types of geographs (maritime, highway, or railway).
When provided with origin and destination coordinates, the library will attempt to find the nearest points on the respective geograph if the given coordinates are not directly on it.
Distance calculations are performed using the haversine formula between nodes in the geograph, ensuring accurate representations of distances on a spherical surface. Then, the shortest path is calculated using the Dijkstra's algorithm.
It's important to note that if no path connecting the two coordinates is found in the geograph, the function will return the direct haversine distance between the two points, along with a path containing only the origin and destination points.
Please be aware that this library is primarily intended for generating realistic distance estimates and paths for visualization or estimation purposes. It is not designed for precise navigation or routing in real-world scenarios. The calculated routes and distances should be used for approximations and visual representations rather than as definitive guidance for actual travel or transportation planning.
- Maritime: based on the MARNET geograph.
- Highways: based on the OpenStreetMap highways database
- Railways: based on the OpenStreetMap railways database
pip install pyroutrs
# or
poetry install pyroutrsfrom pyroutrs import maritime_distance, Geoloc
origin: Geoloc = (40.6759, -74.0504)
destination: Geoloc = (41.0067858, 28.9732219)
distance, path = maritime_distance(origin, destination)
print(distance) # 9224.95741604269
print(len(path)) # 118
print(path[0]) # (40.6759, -74.0504)
print(path[-1]) # (41.0067858, 28.9732219)
distance, path = highway_distance(origin, destination)
print(distance) # 57.237115955889074
print(len(path)) # 39
distance, path = railway_distance(origin, destination)
print(distance) # 749.4744344461568
print(len(path)) # 603For bulk distance calculations, you can use the parallel functions. They accept a list of Leg objects and return a list of tuples, where each tuple contains the distance and the path.
These parallel functions will leverage all available CPU cores to calculate distances, regardless of the GIL, since the distance calculations are performed in Rust.
from pyroutrs import par_maritime_distance, Leg
legs: list[Leg] = [(origin, destination), (origin, destination)]
distances = par_maritime_distance(legs)
for distance, path in distances:
print(distance) # 9224.95741604269
print(path[0]) # (40.6759, -74.0504)
print(path[-1]) # (41.0067858, 28.9732219)The Geoloc type is a tuple[float, float] representing a latitude and longitude.
The GeoPath type is a list[Geoloc] representing a path of coordinates.
The Leg type is a tuple[Geoloc, Geoloc] representing an origin and destination.
MIT
Contributions are welcome! Please open an issue or submit a PR. Contact Carlo Casorzo at carlo[a]deepblu.dev or in X @deepbludev
poetry install
# Build the package
maturin develop --manifest-path Cargo.toml
# or
poe build:dev
# Run the tests
poe testInspired by scgraph and searoute, including the use their datasets, which have been modified to work with this package.