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
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283

#![cfg_attr(coverage_nightly, coverage(off))]
#[cfg(test)]
mod tests {
    use super::super::super::TursoVectorDB;
    use super::super::engine::ClusteringEngine;
    #[allow(unused_imports)]
    use super::super::types::{
        Cluster, ClusterFilters, ClusterMember, ClusterResult, ClusteringMethod, Dendrogram,
        DendrogramMerge, Linkage, OutlierPoint,
    };
    #[allow(unused_imports)]
    use std::collections::HashMap;
    use std::sync::Arc;

    /// Helper to create a clustering engine for tests
    async fn create_test_engine() -> ClusteringEngine {
        let db = TursoVectorDB::new_local(":memory:")
            .await
            .expect("Failed to create test database");
        ClusteringEngine::new(Arc::new(db))
    }

    // ==================== Euclidean Distance Tests ====================

    #[tokio::test]
    async fn test_euclidean_distance_unit_vector() {
        let engine = create_test_engine().await;

        let v1 = vec![0.0, 0.0, 0.0];
        let v2 = vec![1.0, 0.0, 0.0];

        let dist = engine.euclidean_distance(&v1, &v2);
        assert!((dist - 1.0).abs() < 1e-6);
    }

    #[tokio::test]
    async fn test_euclidean_distance_same_point() {
        let engine = create_test_engine().await;

        let v = vec![1.0, 2.0, 3.0];
        let dist = engine.euclidean_distance(&v, &v);
        assert!((dist - 0.0).abs() < 1e-6);
    }

    #[tokio::test]
    async fn test_euclidean_distance_3d() {
        let engine = create_test_engine().await;

        let v1 = vec![0.0, 0.0, 0.0];
        let v2 = vec![1.0, 1.0, 1.0];

        let dist = engine.euclidean_distance(&v1, &v2);
        // sqrt(1 + 1 + 1) = sqrt(3) ≈ 1.732
        assert!((dist - 3.0_f64.sqrt()).abs() < 1e-6);
    }

    #[tokio::test]
    async fn test_euclidean_distance_high_dimensional() {
        let engine = create_test_engine().await;

        let v1 = vec![0.0; 128];
        let mut v2 = vec![0.0; 128];
        v2[0] = 3.0;
        v2[1] = 4.0;

        let dist = engine.euclidean_distance(&v1, &v2);
        // sqrt(9 + 16) = 5
        assert!((dist - 5.0).abs() < 1e-6);
    }

    #[tokio::test]
    async fn test_euclidean_distance_different_lengths() {
        let engine = create_test_engine().await;

        let v1 = vec![1.0, 2.0, 3.0];
        let v2 = vec![1.0, 2.0];

        let dist = engine.euclidean_distance(&v1, &v2);
        assert_eq!(dist, f64::MAX);
    }

    #[tokio::test]
    async fn test_euclidean_distance_empty_vectors() {
        let engine = create_test_engine().await;

        let v1: Vec<f32> = vec![];
        let v2: Vec<f32> = vec![];

        let dist = engine.euclidean_distance(&v1, &v2);
        assert!((dist - 0.0).abs() < 1e-6);
    }

    // ==================== K-Means Tests ====================

    #[tokio::test]
    async fn test_kmeans_basic_clustering() {
        let engine = create_test_engine().await;

        // Two well-separated clusters
        let vectors = vec![
            vec![0.0, 0.0],
            vec![0.1, 0.1],
            vec![0.2, 0.0],
            vec![10.0, 10.0],
            vec![10.1, 10.1],
            vec![10.2, 10.0],
        ];

        let labels = engine.kmeans(&vectors, 2, 100).unwrap();

        assert_eq!(labels.len(), 6);

        // Points 0, 1, 2 should be in same cluster
        assert_eq!(labels[0], labels[1]);
        assert_eq!(labels[1], labels[2]);

        // Points 3, 4, 5 should be in same cluster
        assert_eq!(labels[3], labels[4]);
        assert_eq!(labels[4], labels[5]);

        // The two groups should be in different clusters
        assert_ne!(labels[0], labels[3]);
    }

    #[tokio::test]
    async fn test_kmeans_single_cluster() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0], vec![3.0, 4.0], vec![5.0, 6.0]];

        let labels = engine.kmeans(&vectors, 1, 100).unwrap();

        assert_eq!(labels.len(), 3);
        assert!(labels.iter().all(|&l| l == 0));
    }

    #[tokio::test]
    async fn test_kmeans_empty_vectors() {
        let engine = create_test_engine().await;

        let vectors: Vec<Vec<f32>> = vec![];
        let result = engine.kmeans(&vectors, 2, 100);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "Cannot cluster empty vector set");
    }

    #[tokio::test]
    async fn test_kmeans_k_zero() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0]];
        let result = engine.kmeans(&vectors, 0, 100);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "k must be greater than 0");
    }

    #[tokio::test]
    async fn test_kmeans_k_greater_than_points() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0], vec![3.0, 4.0]];
        let result = engine.kmeans(&vectors, 5, 100);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "Cannot have more clusters than points");
    }

    #[tokio::test]
    async fn test_kmeans_single_point() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0, 3.0]];
        let labels = engine.kmeans(&vectors, 1, 100).unwrap();

        assert_eq!(labels.len(), 1);
        assert_eq!(labels[0], 0);
    }

    #[tokio::test]
    async fn test_kmeans_k_equals_points() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0], vec![3.0, 4.0], vec![5.0, 6.0]];

        let labels = engine.kmeans(&vectors, 3, 100).unwrap();

        assert_eq!(labels.len(), 3);
        // Each point should be in its own cluster (or at least all labels used)
        let unique_labels: std::collections::HashSet<_> = labels.iter().collect();
        assert_eq!(unique_labels.len(), 3);
    }

    // ==================== K-Means with Seed Tests ====================

    #[tokio::test]
    async fn test_kmeans_with_seed_deterministic() {
        let engine = create_test_engine().await;

        let vectors = vec![
            vec![0.0, 0.0],
            vec![0.1, 0.1],
            vec![10.0, 10.0],
            vec![10.1, 10.1],
        ];

        let labels1 = engine.kmeans_with_seed(&vectors, 2, 100, 42).unwrap();
        let labels2 = engine.kmeans_with_seed(&vectors, 2, 100, 42).unwrap();

        assert_eq!(labels1, labels2);
    }

    #[tokio::test]
    async fn test_kmeans_with_seed_different_seeds() {
        let engine = create_test_engine().await;

        // Using well-separated clusters to ensure clustering is deterministic
        let vectors = vec![
            vec![0.0, 0.0],
            vec![0.01, 0.01],
            vec![100.0, 100.0],
            vec![100.01, 100.01],
        ];

        let labels1 = engine.kmeans_with_seed(&vectors, 2, 100, 42).unwrap();
        let labels2 = engine.kmeans_with_seed(&vectors, 2, 100, 123).unwrap();

        // Both should correctly cluster (same groupings even if labels differ)
        assert_eq!(labels1[0], labels1[1]);
        assert_eq!(labels1[2], labels1[3]);
        assert_ne!(labels1[0], labels1[2]);

        assert_eq!(labels2[0], labels2[1]);
        assert_eq!(labels2[2], labels2[3]);
        assert_ne!(labels2[0], labels2[2]);
    }

    #[tokio::test]
    async fn test_kmeans_with_seed_empty_vectors() {
        let engine = create_test_engine().await;

        let vectors: Vec<Vec<f32>> = vec![];
        let result = engine.kmeans_with_seed(&vectors, 2, 100, 42);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "Cannot cluster empty vector set");
    }

    #[tokio::test]
    async fn test_kmeans_with_seed_k_zero() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0]];
        let result = engine.kmeans_with_seed(&vectors, 0, 100, 42);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "k must be greater than 0");
    }

    #[tokio::test]
    async fn test_kmeans_with_seed_k_greater_than_points() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0]];
        let result = engine.kmeans_with_seed(&vectors, 3, 100, 42);

        assert!(result.is_err());
        assert_eq!(result.unwrap_err(), "Cannot have more clusters than points");
    }

    #[tokio::test]
    async fn test_kmeans_with_seed_single_cluster() {
        let engine = create_test_engine().await;

        let vectors = vec![vec![1.0, 2.0], vec![3.0, 4.0], vec![5.0, 6.0]];

        let labels = engine.kmeans_with_seed(&vectors, 1, 100, 42).unwrap();

        assert_eq!(labels.len(), 3);
        assert!(labels.iter().all(|&l| l == 0));
    }
}