aprender-core 0.29.2

Next-generation machine learning library in pure Rust
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

//! Encryption property tests.

#![allow(unused_imports)]

use super::super::*;
pub(crate) use proptest::prelude::*;

pub(crate) use super::*;
pub(crate) use proptest::prelude::*;

/// Strategy for generating valid passwords (8-64 chars)
pub(super) fn arb_password() -> impl Strategy<Value = String> {
    proptest::collection::vec(any::<u8>(), 8..64)
        .prop_map(|bytes| bytes.iter().map(|b| (b % 94 + 33) as char).collect())
}

/// Strategy for generating test model data
pub(super) fn arb_model_data() -> impl Strategy<Value = Vec<f32>> {
    proptest::collection::vec(-100.0f32..100.0, 1..100)
}

// 3 cases for encryption tests (Argon2id has high computational cost by design)
proptest! {
    #![proptest_config(ProptestConfig::with_cases(3))]

    /// Property: Encryption roundtrip preserves data (in-memory)
    #[test]
    fn prop_encryption_roundtrip_preserves_data(
        password in arb_password(),
        data in arb_model_data()
    ) {
        use tempfile::tempdir;

        #[derive(Debug, serde::Serialize, serde::Deserialize, PartialEq)]
        struct Model { weights: Vec<f32> }

        let model = Model { weights: data.clone() };
        let dir = tempdir().expect("tempdir");
        let path = dir.path().join("test.apr");

        save_encrypted(&model, ModelType::Custom, &path, SaveOptions::default(), &password)
            .expect("save");
        let loaded: Model = load_encrypted(&path, ModelType::Custom, &password)
            .expect("load");

        prop_assert_eq!(loaded.weights, data);
    }

    /// Property: Wrong password fails decryption
    #[test]
    fn prop_wrong_password_fails(
        password in arb_password(),
        wrong_password in arb_password()
    ) {
        // Skip if passwords happen to be the same
        if password == wrong_password {
            return Ok(());
        }

        use tempfile::tempdir;

        #[derive(Debug, serde::Serialize, serde::Deserialize)]
        struct Model { value: i32 }

        let model = Model { value: 42 };
        let dir = tempdir().expect("tempdir");
        let path = dir.path().join("test.apr");

        save_encrypted(&model, ModelType::Custom, &path, SaveOptions::default(), &password)
            .expect("save");
        let result: Result<Model> = load_encrypted(&path, ModelType::Custom, &wrong_password);

        prop_assert!(result.is_err(), "Wrong password should fail");
    }

    /// Property: Encrypted files have ENCRYPTED flag set
    #[test]
    fn prop_encrypted_flag_set(password in arb_password()) {
        use tempfile::tempdir;

        #[derive(Debug, serde::Serialize, serde::Deserialize)]
        struct Model { v: i32 }

        let model = Model { v: 1 };
        let dir = tempdir().expect("tempdir");
        let path = dir.path().join("test.apr");

        save_encrypted(&model, ModelType::Custom, &path, SaveOptions::default(), &password)
            .expect("save");
        let info = inspect(&path).expect("inspect");

        prop_assert!(info.encrypted, "ENCRYPTED flag must be set");
    }

    /// Property: load_from_bytes_encrypted roundtrip works
    #[test]
    fn prop_bytes_encrypted_roundtrip(
        password in arb_password(),
        data in arb_model_data()
    ) {
        use tempfile::tempdir;

        #[derive(Debug, serde::Serialize, serde::Deserialize, PartialEq)]
        struct Model { weights: Vec<f32> }

        let model = Model { weights: data.clone() };
        let dir = tempdir().expect("tempdir");
        let path = dir.path().join("test.apr");

        save_encrypted(&model, ModelType::Custom, &path, SaveOptions::default(), &password)
            .expect("save");

        let bytes = std::fs::read(&path).expect("read");
        let loaded: Model = load_from_bytes_encrypted(&bytes, ModelType::Custom, &password)
            .expect("load from bytes");

        prop_assert_eq!(loaded.weights, data);
    }
}