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
|
use std::error::Error;
use regex_automata::{
dfa::{dense, Automaton, OverlappingState},
nfa::thompson,
HalfMatch, Input, MatchError,
};
// Tests that quit bytes in the forward direction work correctly.
#[test]
fn quit_fwd() -> Result<(), Box<dyn Error>> {
let dfa = dense::Builder::new()
.configure(dense::Config::new().quit(b'x', true))
.build("[[:word:]]+$")?;
assert_eq!(
Err(MatchError::quit(b'x', 3)),
dfa.try_search_fwd(&Input::new(b"abcxyz"))
);
assert_eq!(
dfa.try_search_overlapping_fwd(
&Input::new(b"abcxyz"),
&mut OverlappingState::start()
),
Err(MatchError::quit(b'x', 3)),
);
Ok(())
}
// Tests that quit bytes in the reverse direction work correctly.
#[test]
fn quit_rev() -> Result<(), Box<dyn Error>> {
let dfa = dense::Builder::new()
.configure(dense::Config::new().quit(b'x', true))
.thompson(thompson::Config::new().reverse(true))
.build("^[[:word:]]+")?;
assert_eq!(
Err(MatchError::quit(b'x', 3)),
dfa.try_search_rev(&Input::new(b"abcxyz"))
);
Ok(())
}
// Tests that if we heuristically enable Unicode word boundaries but then
// instruct that a non-ASCII byte should NOT be a quit byte, then the builder
// will panic.
#[test]
#[should_panic]
fn quit_panics() {
dense::Config::new().unicode_word_boundary(true).quit(b'\xFF', false);
}
// This tests an intesting case where even if the Unicode word boundary option
// is disabled, setting all non-ASCII bytes to be quit bytes will cause Unicode
// word boundaries to be enabled.
#[test]
fn unicode_word_implicitly_works() -> Result<(), Box<dyn Error>> {
let mut config = dense::Config::new();
for b in 0x80..=0xFF {
config = config.quit(b, true);
}
let dfa = dense::Builder::new().configure(config).build(r"\b")?;
let expected = HalfMatch::must(0, 1);
assert_eq!(Ok(Some(expected)), dfa.try_search_fwd(&Input::new(b" a")));
Ok(())
}
|
