Function bipartite_iea_game 

pub fn bipartite_iea_game(
    n1: u32,
    n2: u32,
    m: u64,
    directed: bool,
    mode: BipartiteMode,
    seed: u64,
) -> IgraphResult<BipartiteGraph>

Generate a random bipartite multigraph through independent edge assignment (IEA).

Counterpart of igraph_bipartite_iea_game() from references/igraph/src/misc/bipartite.c:1476. Each of the m edges is assigned, independently and uniformly at random, to one of the max_edges(n1, n2, directed, mode) possible cross-partition vertex pairs. Because draws are with replacement, the result may contain parallel edges — unlike bipartite_game_gnm, which samples m distinct edges.

This model does not sample multigraphs uniformly: a multigraph is produced with probability proportional to (prod A_ij!)^(-1), so all simple graphs share one probability while non-simple ones are down-weighted by their edge multiplicities. See bipartite_game_gnm for uniform sampling of simple bipartite graphs.

• n1, n2, directed, mode, seed — see bipartite_game_gnp.
• m — exact edge count (no upper bound beyond u64; multi-edges are allowed, so m may exceed max_edges).

Errors

Returns IgraphError::InvalidArgument if n1 + n2 overflows u32, or if m > 0 while the pair space is empty (n1 == 0 or n2 == 0).

Examples

use rust_igraph::{bipartite_iea_game, BipartiteMode};
let bg = bipartite_iea_game(3, 4, 20, false, BipartiteMode::All, 11).unwrap();
assert_eq!(bg.graph.vcount(), 7);
assert_eq!(bg.graph.ecount(), 20); // multi-edges allowed → exactly m edges
// Every edge crosses the partition.
for eid in 0..bg.graph.ecount() {
    let (u, v) = bg.graph.edge(eid as u32).unwrap();
    assert_ne!(bg.types[u as usize], bg.types[v as usize]);
}