soyokaze 0.6.3

HTTP/1/2/3 Library Crate
Documentation 
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
284
285
286
287
288
289
290

use soyokaze::helpers::fields::{Entry, Error, HeaderField, Integer, Mark, StaticIndex, StringLiteral};
use soyokaze::helpers::{hpack, qpack};

fn field(name: &str, value: &str) -> HeaderField {
    HeaderField::new(name, value)
}

#[test]
fn encodes_the_specification_integers() {
    let mut out = Vec::new();
    Integer::encode(&mut out, 10, 5, 0);
    assert_eq!(out, [10]);

    let mut out = Vec::new();
    Integer::encode(&mut out, 1337, 5, 0);
    assert_eq!(out, [31, 154, 10]);

    let mut out = Vec::new();
    Integer::encode(&mut out, 42, 8, 0);
    assert_eq!(out, [42]);
}

#[test]
fn decodes_the_specification_integers() {
    assert_eq!(Integer::decode(&[10], 5), Ok((1, 10)));
    assert_eq!(Integer::decode(&[31, 154, 10], 5), Ok((3, 1337)));
    assert_eq!(Integer::decode(&[42], 8), Ok((1, 42)));
}

#[test]
fn keeps_the_flag_bits_above_the_prefix() {
    let mut out = Vec::new();
    Integer::encode(&mut out, 2, 6, 0x40);
    assert_eq!(out, [0x42]);
    assert_eq!(Integer::decode(&out, 6), Ok((1, 2)));
}

#[test]
fn round_trips_integers_at_every_prefix() {
    let values = [0, 1, 2, 14, 15, 16, 30, 31, 32, 127, 128, 255, 256, 16_383, 1 << 32, u64::MAX];

    for prefix_bits in 1..=8u8 {
        for value in values {
            let mut out = Vec::new();
            Integer::encode(&mut out, value, prefix_bits, 0);
            assert_eq!(Integer::decode(&out, prefix_bits), Ok((out.len(), value)), "{value} at {prefix_bits}");
        }
    }
}

#[test]
fn reports_a_truncated_integer() {
    assert_eq!(Integer::decode(&[], 5), Err(Error::Incomplete));
    assert_eq!(Integer::decode(&[31, 154], 5), Err(Error::Incomplete));
}

#[test]
fn refuses_an_integer_that_does_not_fit() {
    let overflowing = [0xff, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x80, 0x7f];
    assert_eq!(Integer::decode(&overflowing, 8), Err(Error::IntegerOverflow));
}

#[test]
fn round_trips_strings_both_ways() {
    for value in [&b""[..], b"a", b"custom-key", b"www.example.com", &[0xff, 0x00, 0x80]] {
        for huffman in [false, true] {
            let mut out = Vec::new();
            StringLiteral::encode(&mut out, value, 7, 0x00, huffman);

            assert_eq!(out.first().is_some_and(|first| first & 0x80 != 0), huffman);
            assert_eq!(StringLiteral::decode(&out, 7), Ok((out.len(), value.to_vec())));
        }
    }
}

#[test]
fn moves_the_huffman_mark_with_the_prefix() {
    for prefix_bits in [3, 4, 5, 6, 7u8] {
        for huffman in [false, true] {
            let mut out = Vec::new();
            StringLiteral::encode(&mut out, b"www.example.com", prefix_bits, 0x00, huffman);

            assert_eq!(out.first().is_some_and(|first| first & 1 << prefix_bits != 0), huffman);
            assert_eq!(StringLiteral::decode(&out, prefix_bits), Ok((out.len(), b"www.example.com".to_vec())));
        }
    }
}

#[test]
fn reports_a_string_that_ends_early() {
    assert_eq!(StringLiteral::decode(&[0x05, b'a', b'b'], 7), Err(Error::Incomplete));
    assert_eq!(StringLiteral::decode(&[], 7), Err(Error::Incomplete));
}

#[test]
fn refuses_a_string_whose_length_cannot_be_addressed() {
    let huge = [0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x01];
    assert!(matches!(StringLiteral::decode(&huge, 7), Err(Error::Incomplete | Error::IntegerOverflow)));
}

#[test]
fn prefers_huffman_only_when_it_helps() {
    assert!(StringLiteral::prefers_huffman(b"www.example.com"));
    assert!(!StringLiteral::prefers_huffman(&[0xc0; 8]));
}

#[test]
fn every_sensitive_name_is_recognised() {
    // Recognised by length and first octet rather than by walking the list, so
    // the two are checked against each other over names near the list as well
    // as the list itself.
    let corpus: Vec<String> = HeaderField::SENSITIVE
        .iter()
        .flat_map(|name| {
            let name = name.to_string();
            [name.clone(), name[..name.len() - 1].to_string(), format!("{name}x"), format!("x{}", &name[1..])]
        })
        .chain(["".into(), "accept".into(), "cookies".into(), "content-type".into()])
        .collect();

    for name in &corpus {
        assert_eq!(
            field(name, "value").sensitive(),
            HeaderField::SENSITIVE.contains(&name.as_str()),
            "{name:?} is not classified as the list says"
        );
    }
}

#[test]
fn a_field_costs_its_octets_plus_the_overhead() {
    assert_eq!(field("name", "value").size(), 4 + 5 + HeaderField::OVERHEAD);
    assert_eq!(field("", "").size(), HeaderField::OVERHEAD);
}

#[test]
fn errors_describe_themselves() {
    assert_eq!(Error::IntegerOverflow.to_string(), "integer representation overflowed");
    assert_eq!(Error::Incomplete.to_string(), "representation ends before the input does");
}

// ---------------------------------------------------------------- conformance

#[test]
fn a_field_name_is_a_non_empty_token() {
    for name in ["content-length", "x-custom", "a!#$%&'*+-.^_`|~9"] {
        assert!(HeaderField::is_name(name), "{name:?} is a token");
    }

    for name in ["", "bad name", "a:b", "a\r\nb", "a\0b", "a(b", "a@b", "a\u{7f}b"] {
        assert!(!HeaderField::is_name(name), "RFC 9110 5.6.2: {name:?} is not a token, so it is not a field name");
    }

    assert!(HeaderField::is_lowercase_name("x-custom"));
    assert!(!HeaderField::is_lowercase_name("X-Custom"), "RFC 9113 8.2.1 and RFC 9114 4.2 want a lowercase name");
}

#[test]
fn a_field_value_carries_no_control_octet_and_no_surrounding_space() {
    for value in ["", "plain", "a\tb", "a b", "caf\u{e9}"] {
        assert!(HeaderField::is_value(value), "{value:?} is a field value");
    }

    for value in ["a\r\nb", "a\rb", "a\nb", "a\0b", "a\u{7f}b", " leading", "trailing ", "\ttab", "tab\t"] {
        assert!(!HeaderField::is_value(value), "RFC 9110 5.5: {value:?} is not a field value");
    }
}

#[test]
fn a_string_literal_leaves_room_for_its_huffman_mark() {
    let max = StringLiteral::MAX_PREFIX_BITS;
    assert_eq!(max, 7, "the mark sits just above the prefix, so eight bits would push it out of the octet");

    for prefix_bits in [0u8, 1, 7] {
        assert_eq!(StringLiteral::huffman_mark(prefix_bits), 1 << prefix_bits);
    }

    for prefix_bits in [8u8, 63, 200, 255] {
        assert_eq!(
            StringLiteral::huffman_mark(prefix_bits),
            1 << max,
            "a prefix wider than {max} is read as {max} rather than shifting the mark out of the octet"
        );
    }
}

#[test]
fn a_prefix_no_representation_uses_never_takes_the_process_with_it() {
    let mut out = Vec::new();
    StringLiteral::encode(&mut out, b"hello", 8, 0x00, true);
    assert!(!out.is_empty(), "an over-wide prefix must be read as the widest one, not overflow a shift");

    let mut scratch = Vec::new();
    for prefix_bits in [8u8, 64, 255] {
        let _ = StringLiteral::decode_into_ascii(&out, prefix_bits, &mut scratch);
        let _ = StringLiteral::decode_into_ascii(&[0x05, b'a'], prefix_bits, &mut scratch);
    }
}

#[test]
fn the_token_table_marks_lowercase_tokens() {
    // RFC 9110 §5.6.2 gives the token characters; RFC 9113 §8.2.1 and RFC 9114
    // §4.2 add that a field name carries no uppercase letter. The table holds
    // both answers, and the second must be exactly the first without A-Z.
    for value in 0..=255u8 {
        let token = value.is_ascii_alphanumeric()
            || matches!(value, b'!' | b'#' | b'$' | b'%' | b'&' | b'\'' | b'*' | b'+' | b'-' | b'.' | b'^' | b'_' | b'`' | b'|' | b'~');

        let marks = HeaderField::TOKENS[value as usize];
        assert_eq!(marks & HeaderField::TOKEN != 0, token, "{value:#04x} is marked wrongly as a token");
        assert_eq!(
            marks & HeaderField::LOWERCASE != 0,
            token && !value.is_ascii_uppercase(),
            "{value:#04x} is marked wrongly as a lowercase token"
        );
    }
}

#[test]
fn a_lowercase_name_is_a_token_carrying_no_capital() {
    for value in 0..=255u8 {
        let token = HeaderField::TOKENS[value as usize] & HeaderField::TOKEN != 0;

        for length in 1..=20usize {
            for at in 0..length {
                let mut name = vec![b'a'; length];
                name[at] = value;
                let name = String::from_utf8_lossy(&name).into_owned();

                assert_eq!(HeaderField::is_name(&name), token, "{name:?} is not classified as a token");
                assert_eq!(
                    HeaderField::is_lowercase_name(&name),
                    token && !value.is_ascii_uppercase(),
                    "{name:?} is not classified as a lowercase token"
                );
            }
        }
    }

    assert!(!HeaderField::is_name(""), "an empty name is not a token");
    assert!(!HeaderField::is_lowercase_name(""), "an empty name is not a lowercase token");
}

#[test]
fn a_mark_stands_for_exactly_its_name() {
    let names = [
        "", "a", "b", "te", "host", "accept", "accept-encoding", "content-type", "content-length",
        "strict-transport-security", "x-a-rather-long-header-field-name-indeed", "ab", "ba", "abc", "acb",
    ];

    for left in names {
        assert_eq!(Mark::of(left), Mark::of(left), "the same name marks differently twice");

        for right in names {
            if left != right {
                assert_ne!(Mark::of(left), Mark::of(right), "{left:?} and {right:?} share a mark");
            }
        }
    }
}

#[test]
fn an_entry_confirms_a_mark_against_the_name_itself() {
    let entry = Entry::of(field("content-type", "text/html"));

    assert!(entry.named(Mark::of("content-type"), &field("content-type", "text/plain")));
    assert!(entry.valued(&field("anything", "text/html")));
    assert!(!entry.valued(&field("content-type", "text/plain")));

    // A mark that matched but a name that did not must not be taken for a hit,
    // whatever the mark says.
    assert!(!entry.named(Mark::of("content-type"), &field("content-length", "10")));
}

#[test]
fn the_static_index_finds_every_entry_it_was_built_from() {
    for (entries, base) in [(hpack::StaticTable::entries().as_slice(), 1usize), (qpack::StaticTable::entries().as_slice(), 0)] {
        let index = StaticIndex::new(entries, base);

        for (offset, entry) in entries.iter().enumerate() {
            let (named, exact) = index.lookup(&entry.name, &entry.value);

            let first = entries.iter().position(|other| other.name == entry.name).expect("the name is in the table");
            assert_eq!(named, Some(base + first), "{:?} does not resolve to its lowest index", entry.name);
            assert_eq!(exact, Some(base + offset), "{:?}: {:?} does not resolve to its own index", entry.name, entry.value);
        }

        assert_eq!(index.lookup("x-not-in-any-static-table", "?"), (None, None));
    }
}