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
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

impl MLQualityScorer {
    /// Create a new ML quality scorer
    pub fn new() -> Self {
        let mut heuristic_weights = HashMap::new();
        // Default heuristic weights (fallback when ML not trained)
        heuristic_weights.insert("loc".to_string(), 0.02);
        heuristic_weights.insert("nesting".to_string(), 0.5);
        heuristic_weights.insert("control_flow".to_string(), 0.3);
        heuristic_weights.insert("loops".to_string(), 0.2);

        Self {
            complexity_model: None,
            tdg_model: None,
            heuristic_weights,
            trained: false,
            training_samples: 0,
            feature_importance: HashMap::new(),
        }
    }

    /// Predict complexity score using ML model
    pub fn predict_complexity(&self, features: &ComplexityFeatures) -> Result<QualityPrediction> {
        let feature_vec = features.to_vector();

        let (score, ml_used) = if let Some(ref model) = self.complexity_model {
            // Use ML model
            let feature_vec_f32: Vec<f32> = feature_vec.iter().map(|&x| x as f32).collect();
            match Matrix::from_vec(1, 8, feature_vec_f32) {
                Ok(x) => {
                    let predictions = model.predict(&x);
                    let predicted = predictions.as_slice()[0].clamp(0.0, 100.0) as f64;
                    (predicted, true)
                }
                Err(_) => (self.heuristic_complexity(features), false),
            }
        } else {
            // Fallback to heuristics
            (self.heuristic_complexity(features), false)
        };

        let confidence = if ml_used { 0.85 } else { 0.5 };

        let mut feature_contributions = HashMap::new();
        let feature_names = [
            "loc",
            "nesting",
            "control_flow",
            "loops",
            "conditionals",
            "functions",
            "avg_size",
            "language",
        ];
        for (name, &value) in feature_names.iter().zip(feature_vec.iter()) {
            feature_contributions.insert(name.to_string(), value);
        }

        Ok(QualityPrediction {
            score,
            confidence,
            ml_used,
            feature_contributions,
        })
    }

    /// Predict TDG score using ML model
    pub fn predict_tdg(&self, features: &TDGFeatures) -> Result<QualityPrediction> {
        let feature_vec = features.to_vector();

        let (score, ml_used) = if let Some(ref model) = self.tdg_model {
            let feature_vec_f32: Vec<f32> = feature_vec.iter().map(|&x| x as f32).collect();
            match Matrix::from_vec(1, 8, feature_vec_f32) {
                Ok(x) => {
                    let predictions = model.predict(&x);
                    let predicted = predictions.as_slice()[0].clamp(0.0, 5.0) as f64;
                    (predicted, true)
                }
                Err(_) => (self.heuristic_tdg(features), false),
            }
        } else {
            (self.heuristic_tdg(features), false)
        };

        let confidence = if ml_used { 0.85 } else { 0.5 };

        let mut feature_contributions = HashMap::new();
        let feature_names = [
            "complexity",
            "churn",
            "coupling",
            "domain_risk",
            "duplication",
            "coverage",
            "age",
            "frequency",
        ];
        for (name, &value) in feature_names.iter().zip(feature_vec.iter()) {
            feature_contributions.insert(name.to_string(), value);
        }

        Ok(QualityPrediction {
            score,
            confidence,
            ml_used,
            feature_contributions,
        })
    }

    /// Heuristic complexity calculation (fallback)
    fn heuristic_complexity(&self, features: &ComplexityFeatures) -> f64 {
        // Traditional formula: base + nesting_penalty + control_flow
        let base = features.loc / 50.0;
        let nesting_penalty = features.max_nesting * 2.0;
        let control_flow_penalty = features.control_flow_count * 0.5;

        (base + nesting_penalty + control_flow_penalty).min(100.0)
    }

    /// Heuristic TDG calculation (fallback)
    fn heuristic_tdg(&self, features: &TDGFeatures) -> f64 {
        // Traditional weighted sum
        let score = features.complexity * 0.3
            + features.churn * 0.25
            + features.coupling * 0.2
            + features.domain_risk * 0.15
            + features.duplication * 0.1;

        score.clamp(0.0, 5.0)
    }

    /// Check if model is trained
    pub fn is_trained(&self) -> bool {
        self.trained
    }

    /// Get feature importance
    pub fn feature_importance(&self) -> &HashMap<String, f64> {
        &self.feature_importance
    }

    /// Save model to file
    pub fn save(&self, _path: &Path) -> Result<()> {
        // Serialization deferred until aprender supports it
        Ok(())
    }

    /// Load model from file
    pub fn load(_path: &Path) -> Result<Self> {
        // Deserialization deferred until aprender supports it
        Ok(Self::new())
    }
}

impl Default for MLQualityScorer {
    fn default() -> Self {
        Self::new()
    }
}