pmat 3.11.0

PMAT - Zero-config AI context generation and code quality toolkit (CLI, MCP, HTTP)
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

// Matrix operations: EdgeData weights, SimpleSparseMatrix impl, GraphMatrices conversions
// Included by types.rs - shares parent module scope (no `use` imports)

impl SimpleSparseMatrix {
    /// Create a new sparse matrix with given dimensions
    pub fn new(nrows: usize, ncols: usize) -> Self {
        Self {
            nrows,
            ncols,
            triplets: Vec::new(),
        }
    }

    /// Add a value at (row, col)
    pub fn push(&mut self, row: usize, col: usize, value: f64) {
        self.triplets.push((row, col, value));
    }

    /// Get row values as iterator
    pub fn row_values(&self, row: usize) -> impl Iterator<Item = (usize, f64)> + '_ {
        self.triplets
            .iter()
            .filter(move |(r, _, _)| *r == row)
            .map(|(_, c, v)| (*c, *v))
    }

    /// Get all values in a row as Vec
    pub fn row_as_vec(&self, row: usize) -> Vec<(usize, f64)> {
        self.row_values(row).collect()
    }
}

impl EdgeData {
    /// Convert heterogeneous edge types to numeric weights
    /// Complexity: 3 (simple match with arithmetic)
    pub fn to_numeric_weight(&self) -> f64 {
        match self {
            EdgeData::Import { weight, .. } => *weight * 2.0, // Imports weighted higher
            EdgeData::FunctionCall { count, .. } => *count as f64,
            EdgeData::TypeDependency { strength, .. } => *strength * 1.5,
            EdgeData::DataFlow { confidence, .. } => *confidence,
            EdgeData::Inheritance { depth } => 3.0 / (*depth as f64 + 1.0),
        }
    }
}

/// Conversion from DependencyGraph to matrix representations
/// Complexity: 8 (loop with matrix operations)
/// Sovereign AI Stack: Uses simple Vec-based sparse matrices
impl From<&DependencyGraph> for GraphMatrices {
    fn from(graph: &DependencyGraph) -> Self {
        let n = graph.node_count();
        let mut adjacency = SimpleSparseMatrix::new(n, n);
        let mut out_degrees = vec![0.0; n];
        let mut edges = Vec::new();

        // Build adjacency matrix with edge weights
        for edge in graph.edge_references() {
            let weight = edge.weight().to_numeric_weight();
            let source = edge.source().0 as usize;
            let target = edge.target().0 as usize;

            if source < n && target < n {
                edges.push((source, target, weight));
                adjacency.push(source, target, weight);
                out_degrees[source] += weight;
            }
        }

        // Create column-stochastic transition matrix
        let transition = Self::normalize_columns(&adjacency, &out_degrees);

        // Compute Laplacian L = D - A
        let laplacian = Self::compute_laplacian(&adjacency);

        GraphMatrices {
            adjacency,
            transition,
            laplacian,
            out_degrees,
            node_count: n,
            edges,
        }
    }
}

impl GraphMatrices {
    /// Normalize columns for stochastic matrix
    /// Complexity: 6 (nested loop with early exit)
    fn normalize_columns(
        adjacency: &SimpleSparseMatrix,
        out_degrees: &[f64],
    ) -> SimpleSparseMatrix {
        let n = adjacency.nrows;
        let mut result = SimpleSparseMatrix::new(n, n);

        for i in 0..n {
            if out_degrees[i] > 0.0 {
                for (col, value) in adjacency.row_values(i) {
                    result.push(i, col, value / out_degrees[i]);
                }
            }
        }

        result
    }

    /// Compute graph Laplacian
    /// Complexity: 6 (simplified matrix operations)
    fn compute_laplacian(adjacency: &SimpleSparseMatrix) -> SimpleSparseMatrix {
        let n = adjacency.nrows;
        let mut result = SimpleSparseMatrix::new(n, n);

        // Compute degree matrix D and build Laplacian L = D - A
        for i in 0..n {
            let row_vals: Vec<_> = adjacency.row_values(i).collect();
            let degree: f64 = row_vals.iter().map(|(_, v)| v).sum();

            // Add diagonal degree
            result.push(i, i, degree);

            // Subtract adjacency values
            for (col, value) in row_vals {
                result.push(i, col, -value);
            }
        }

        result
    }
}