1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
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
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166

use crate::sparse::{Node, SparseMatrix};
use std::collections::VecDeque;

#[derive(Debug, Clone, Eq, PartialEq)]
struct PathHead {
    node: Node,
    parent: Option<Node>,
    path_length: usize,
}

impl PathHead {
    fn iter<'a>(&'a self, h: &'a SparseMatrix) -> impl Iterator<Item = PathHead> + 'a {
        self.node
            .iter(h)
            .filter(move |&x| {
                if let Some(parent) = self.parent {
                    x != parent
                } else {
                    true
                }
            })
            .map(move |x| PathHead {
                node: x,
                parent: Some(self.node),
                path_length: self.path_length + 1,
            })
    }
}

/// Results for BFS algorithm
///
/// This gives the distances of each of the nodes of the graph from the node
/// that was used as root for the BFS algorithm. Distances are represented
/// as `Option<usize>`, with the value `None` for nodes that are not reachable
/// from the root.
#[derive(Debug, Clone, Eq, PartialEq)]
pub struct BFSResults {
    /// The vector of distances from each of the row nodes to the root
    pub row_nodes_distance: Vec<Option<usize>>,
    /// The vector of distances from each of the column nodes to the root
    pub col_nodes_distance: Vec<Option<usize>>,
}

impl BFSResults {
    fn get_node_mut(&mut self, node: Node) -> &mut Option<usize> {
        match node {
            Node::Row(n) => &mut self.row_nodes_distance[n],
            Node::Col(n) => &mut self.col_nodes_distance[n],
        }
    }
}

pub struct BFSContext<'a> {
    results: BFSResults,
    to_visit: VecDeque<PathHead>,
    h: &'a SparseMatrix,
}

impl<'a> BFSContext<'a> {
    pub fn new(h: &SparseMatrix, node: Node) -> BFSContext {
        let mut to_visit = VecDeque::new();
        to_visit.push_back(PathHead {
            node,
            parent: None,
            path_length: 0,
        });
        let mut results = BFSResults {
            row_nodes_distance: vec![None; h.num_rows()],
            col_nodes_distance: vec![None; h.num_cols()],
        };
        results.get_node_mut(node).replace(0);
        BFSContext {
            results,
            to_visit,
            h,
        }
    }

    pub fn bfs(mut self) -> BFSResults {
        while let Some(head) = self.to_visit.pop_front() {
            for next_head in head.iter(self.h) {
                let next_dist = self.results.get_node_mut(next_head.node);
                if next_dist.is_none() {
                    *next_dist = Some(next_head.path_length);
                    self.to_visit.push_back(next_head);
                }
            }
        }
        self.results
    }

    pub fn local_girth(mut self, max: usize) -> Option<usize> {
        while let Some(head) = self.to_visit.pop_front() {
            for next_head in head.iter(self.h) {
                let next_dist = self.results.get_node_mut(next_head.node);
                if let Some(dist) = *next_dist {
                    let total = dist + next_head.path_length;
                    return if total <= max { Some(total) } else { None };
                } else {
                    *next_dist = Some(next_head.path_length);
                    if next_head.path_length < max {
                        self.to_visit.push_back(next_head);
                    }
                }
            }
        }
        None
    }
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn disconnected_2x2() {
        let mut h = SparseMatrix::new(2, 2);
        h.insert(0, 0);
        h.insert(1, 1);
        let r = h.bfs(Node::Col(0));
        assert_eq!(r.row_nodes_distance[0], Some(1));
        assert_eq!(r.row_nodes_distance[1], None);
        assert_eq!(r.col_nodes_distance[0], Some(0));
        assert_eq!(r.col_nodes_distance[1], None);
    }

    #[test]
    fn complete_nxm() {
        let n = 20;
        let m = 10;
        let mut h = SparseMatrix::new(n, m);
        for i in 0..n {
            for j in 0..m {
                h.insert(i, j);
            }
        }
        let r = h.bfs(Node::Row(0));
        assert_eq!(r.row_nodes_distance[0], Some(0));
        for i in 1..n {
            assert_eq!(r.row_nodes_distance[i], Some(2));
        }
        for i in 0..m {
            assert_eq!(r.col_nodes_distance[i], Some(1));
        }
    }

    #[test]
    fn circulant() {
        let n = 20;
        let mut h = SparseMatrix::new(n, n);
        for j in 0..n {
            h.insert(j, j);
            h.insert(j, (j + 1) % n);
        }
        let r = h.bfs(Node::Row(0));
        assert_eq!(r.row_nodes_distance[0], Some(0));
        for j in 1..n {
            let dist = std::cmp::min(2 * j, 2 * (n - j));
            assert_eq!(r.row_nodes_distance[j], Some(dist));
        }
        for j in 1..n + 1 {
            let dist = std::cmp::min(2 * j - 1, 2 * (n - j) + 1);
            assert_eq!(r.col_nodes_distance[j % n], Some(dist));
        }
    }
}