Function graphalgs::metrics::weighted_eccentricity

pub fn weighted_eccentricity<G>(
    graph: G,
    start: G::NodeId
) -> Option<G::EdgeWeight>where
    G: NodeCount + IntoNodeIdentifiers + IntoEdges + NodeIndexable,
    G::EdgeWeight: FloatMeasure,

Weighted eccentricity.

Calculate the distance to the node farthest from start, given the edge weights. The function is based on the Bellman-Ford algorithm and has a time complexity of O(|V|*|E|). So if edge weight is not important it is better to use eccentricity() function.

Arguments

• graph: weighted graph.
• start: node whose eccentricity is to be calculated.

Returns

• Some(G::EdgeWeight): the eccentricity.
• None: if graph contains negative cycle.

Examples

use graphalgs::metrics::weighted_eccentricity;
use petgraph::Graph;

let inf = f32::INFINITY;

let graph = Graph::<(), f32>::from_edges(&[
    (0, 1, 2.0), (1, 2, 10.0), (1, 3, -5.0),
    (3, 2, 2.0), (2, 3, 20.0),
]);

assert_eq!(weighted_eccentricity(&graph, 0.into()), Some(2.0));
assert_eq!(weighted_eccentricity(&graph, 1.into()), Some(inf));
assert_eq!(weighted_eccentricity(&graph, 2.into()), Some(inf));
assert_eq!(weighted_eccentricity(&graph, 3.into()), Some(inf));

// Negative cycle.
let graph = Graph::<(), f32>::from_edges(&[
    (0, 1, 2.0), (1, 2, 2.0), (2, 0, -10.0)
]);

assert_eq!(weighted_eccentricity(&graph, 0.into()), None);
assert_eq!(weighted_eccentricity(&graph, 1.into()), None);
assert_eq!(weighted_eccentricity(&graph, 2.into()), None);