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

extern crate deunicode;

use deunicode::deunicode_char;

/// Convert any unicode string to an ascii "slug" (useful for file names/url components)
///
/// The returned "slug" will consist of a-z, 0-9, and '-'. Furthermore, a slug will
/// never contain more than one '-' in a row and will never start or end with '-'.
///
/// ```rust
/// use self::slug::slugify;
///
/// assert_eq!(slugify("My Test String!!!1!1"), "my-test-string-1-1");
/// assert_eq!(slugify("test\nit   now!"), "test-it-now");
/// assert_eq!(slugify("  --test_-_cool"), "test-cool");
/// assert_eq!(slugify("Æúű--cool?"), "aeuu-cool");
/// assert_eq!(slugify("You & Me"), "you-me");
/// assert_eq!(slugify("user@example.com"), "user-example-com");
/// ```
pub fn slugify<S: AsRef<str>>(s: S) -> String {
    _slugify(s.as_ref())
}

// avoid unnecessary monomorphizations
fn _slugify(s: &str) -> String {
    let mut slug: Vec<u8> = Vec::with_capacity(s.len());
    // Starts with true to avoid leading -
    let mut prev_is_dash = true;
    {
        let mut push_char = |x: char| {
            match x {
                'a'...'z' | '0'...'9' => {
                    prev_is_dash = false;
                    slug.push(x as u8);
                }
                'A'...'Z' => {
                    prev_is_dash = false;
                    // Manual lowercasing as Rust to_lowercase() is unicode
                    // aware and therefore much slower
                    slug.push((x as u8) - b'A' + b'a');
                }
                _ => {
                    if !prev_is_dash {
                        slug.push(b'-');
                        prev_is_dash = true;
                    }
                }
            }
        };

        for c in s.chars() {
            if c.is_ascii() {
                (push_char)(c);
            } else {
                for cx in deunicode_char(c).unwrap_or("-").chars() {
                    (push_char)(cx);
                }
            }
        }
    }

    // It's not really unsafe in practice, we know we have ASCII
    let mut string = unsafe { String::from_utf8_unchecked(slug) };
    if string.ends_with('-') {
        string.pop();
    }
    // We likely reserved more space than needed.
    string.shrink_to_fit();
    string
}