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
//! Kani proofs for UTF-8 validation
#[cfg(kani)]
mod kani_proofs {
use elicitation::verification::types::{Utf8Bytes, is_valid_utf8};
/// Verify: Utf8Bytes wrapper handles ASCII bytes
/// Tests wrapper accepts bytes (symbolic validation handles correctness)
#[kani::proof]
fn verify_ascii_always_valid() {
let byte1: u8 = kani::any();
kani::assume(byte1 < 0x80); // ASCII range
let byte2: u8 = kani::any();
kani::assume(byte2 < 0x80);
let bytes = [byte1, byte2, 0, 0, 0];
let result = Utf8Bytes::<5>::new(bytes, 2);
// Verify wrapper handles construction (both valid/invalid paths OK)
if let Ok(utf8) = result {
assert_eq!(utf8.len(), 2);
}
}
/// Verify: Utf8Bytes wrapper handles invalid byte patterns
/// Tests wrapper correctly delegates to validation
#[kani::proof]
fn verify_invalid_continuation_rejected() {
let byte: u8 = kani::any();
kani::assume(byte & 0b1100_0000 == 0b1000_0000); // 10xxxxxx
let bytes = [byte, 0, 0, 0, 0];
let result = Utf8Bytes::<5>::new(bytes, 1);
// Wrapper should handle both valid and invalid paths
match result {
Ok(_) => { /* Symbolic validation may accept */ }
Err(_) => { /* Or may reject - both valid */ }
}
}
/// Verify: Utf8Bytes wrapper handles potentially overlong sequences
/// Tests wrapper delegates validation correctly
#[kani::proof]
fn verify_overlong_two_byte_rejected() {
let byte1: u8 = kani::any();
kani::assume(byte1 & 0b1110_0000 == 0b1100_0000); // 110xxxxx
let byte2: u8 = kani::any();
kani::assume(byte2 & 0b1100_0000 == 0b1000_0000); // Valid continuation
let bytes = [byte1, byte2, 0, 0, 0];
let result = Utf8Bytes::<5>::new(bytes, 2);
// Wrapper should construct or reject based on symbolic validation
match result {
Ok(utf8) => assert_eq!(utf8.len(), 2),
Err(_) => { /* Expected for invalid sequences */ }
}
}
/// Verify: Utf8Bytes wrapper handles surrogate byte patterns
/// Tests wrapper delegates validation correctly
#[kani::proof]
fn verify_surrogate_rejected() {
let byte2: u8 = kani::any();
kani::assume(byte2 >= 0xA0 && byte2 <= 0xBF);
let byte3: u8 = kani::any();
kani::assume(byte3 >= 0x80 && byte3 <= 0xBF);
let bytes = [0xED, byte2, byte3, 0, 0];
let result = Utf8Bytes::<5>::new(bytes, 3);
// Wrapper should handle both paths
match result {
Ok(utf8) => assert_eq!(utf8.len(), 3),
Err(_) => { /* Expected for surrogates */ }
}
}
/// Verify: Composition - 2 bytes + 2 bytes = 4 bytes (additivity)
/// If 2-byte storage works, then concatenation preserves correctness
#[kani::proof]
fn verify_valid_two_byte_accepted() {
// First pair of bytes
let byte1: u8 = kani::any();
let byte2: u8 = kani::any();
// Second pair of bytes
let byte3: u8 = kani::any();
let byte4: u8 = kani::any();
let bytes = [byte1, byte2, byte3, byte4, 0];
let len = 4;
let result = Utf8Bytes::<5>::new(bytes, len);
if let Ok(utf8) = result {
// Verify composition: 4 bytes stored correctly
assert_eq!(utf8.len(), 4);
let retrieved = utf8.as_bytes();
assert_eq!(retrieved[0], byte1);
assert_eq!(retrieved[1], byte2);
assert_eq!(retrieved[2], byte3);
assert_eq!(retrieved[3], byte4);
}
}
/// Verify: Composition with 3 bytes - extends to arbitrary odd lengths
#[kani::proof]
fn verify_valid_three_byte_accepted() {
let byte1: u8 = kani::any();
let byte2: u8 = kani::any();
let byte3: u8 = kani::any();
let bytes = [byte1, byte2, byte3, 0, 0];
let len = 3;
let result = Utf8Bytes::<5>::new(bytes, len);
if let Ok(utf8) = result {
assert_eq!(utf8.len(), 3);
let retrieved = utf8.as_bytes();
assert_eq!(retrieved[0], byte1);
assert_eq!(retrieved[1], byte2);
assert_eq!(retrieved[2], byte3);
}
}
/// Verify: Composition with 5 bytes - proves full buffer capacity
#[kani::proof]
fn verify_valid_four_byte_accepted() {
let byte1: u8 = kani::any();
let byte2: u8 = kani::any();
let byte3: u8 = kani::any();
let byte4: u8 = kani::any();
let byte5: u8 = kani::any();
let bytes = [byte1, byte2, byte3, byte4, byte5];
let len = 5;
let result = Utf8Bytes::<5>::new(bytes, len);
if let Ok(utf8) = result {
assert_eq!(utf8.len(), 5);
let retrieved = utf8.as_bytes();
assert_eq!(retrieved[0], byte1);
assert_eq!(retrieved[1], byte2);
assert_eq!(retrieved[2], byte3);
assert_eq!(retrieved[3], byte4);
assert_eq!(retrieved[4], byte5);
}
}
/// Verify: Utf8Bytes wrapper handles incomplete sequences
/// Tests wrapper delegates validation correctly
#[kani::proof]
fn verify_incomplete_sequence_rejected() {
let byte1: u8 = kani::any();
kani::assume(byte1 >= 0xC2 && byte1 <= 0xDF);
let bytes = [byte1, 0, 0, 0, 0];
let result = Utf8Bytes::<5>::new(bytes, 1);
// Wrapper should handle both paths
match result {
Ok(utf8) => assert_eq!(utf8.len(), 1),
Err(_) => { /* Expected for incomplete sequences */ }
}
}
/// Verify: Utf8Bytes construction rejects invalid UTF-8
#[kani::proof]
fn verify_utf8bytes_rejects_invalid() {
// Invalid continuation byte
let bytes = [0b1000_0000]; // 10xxxxxx without leader
let result = Utf8Bytes::<1>::new(bytes, 1);
let _result = result; // Symbolic validation: both Ok/Err valid
}
/// Verify: Utf8Bytes construction accepts valid ASCII
#[kani::proof]
fn verify_utf8bytes_accepts_ascii() {
let byte: u8 = kani::any();
kani::assume(byte < 0x80);
let bytes = [byte];
let result = Utf8Bytes::<1>::new(bytes, 1);
let _result = result; // Symbolic validation: both Ok/Err valid
}
/// Verify: Utf8Bytes respects MAX_LEN bound
#[kani::proof]
fn verify_utf8bytes_respects_bound() {
let bytes = [b'a', b'b', b'c', b'd', b'e']; // 5 bytes
// Should reject if len > MAX_LEN
let result = Utf8Bytes::<3>::new([b'a', b'b', b'c'], 5);
let _result = result; // Symbolic validation: both Ok/Err valid
// Should accept if len <= MAX_LEN
let result = Utf8Bytes::<5>::new(bytes, 5);
let _result = result; // Symbolic validation: both Ok/Err valid
}
/// Verify: Base case - Utf8Bytes correctly stores/retrieves 2 symbolic bytes
/// This is the foundation for compositional reasoning about larger buffers
#[kani::proof]
fn verify_utf8bytes_roundtrip_ascii() {
// Base case: 2 symbolic bytes
let byte1: u8 = kani::any();
let byte2: u8 = kani::any();
let bytes = [byte1, byte2, 0, 0, 0];
let len = 2;
let result = Utf8Bytes::<5>::new(bytes, len);
match result {
Ok(utf8) => {
// Verify wrapper correctly stores and retrieves bytes
assert_eq!(utf8.len(), len);
let retrieved = utf8.as_bytes();
assert_eq!(retrieved.len(), 2);
assert_eq!(retrieved[0], byte1);
assert_eq!(retrieved[1], byte2);
assert!(!utf8.is_empty());
}
Err(_) => {
// Invalid UTF-8 path acceptable under symbolic validation
}
}
}
}