headless_lms_utils/
strings.rs

1use once_cell::sync::Lazy;
2use rand::{RngExt, distr::Alphanumeric, rng};
3use regex::Regex;
4
5static IETF_LANGUAGE_CODE_REGEX: Lazy<Regex> = Lazy::new(|| {
6    Regex::new(r"^[a-z]{2,3}(-[A-Z][a-z]{3})?(-[A-Z]{2})?$")
7        .expect("Invalid IETF language code regex.")
8});
9
10static HTML_TAG_REGEX: Lazy<Regex> =
11    Lazy::new(|| Regex::new(r"<[^>]*>").expect("Invalid HTML tag regex."));
12
13pub fn generate_random_string(length: usize) -> String {
14    rng()
15        .sample_iter(Alphanumeric)
16        .take(length)
17        .map(char::from)
18        .collect()
19}
20
21pub fn generate_easily_writable_random_string(length: usize) -> String {
22    rng()
23        .sample_iter(Alphanumeric)
24        .filter(|c: &u8| c.is_ascii_lowercase() || c.is_ascii_digit())
25        // Filter out characters that might be confused with each other
26        .filter(|c| c != &b'l' && c != &b'1' && c != &b'o' && c != &b'0')
27        .take(length)
28        .map(char::from)
29        .collect()
30}
31
32/// Checks whether the string is IETF language code where subtags are separated with underscore.
33pub fn is_ietf_language_code_like(string: &str) -> bool {
34    IETF_LANGUAGE_CODE_REGEX.is_match(string)
35}
36
37/// Removes all HTML tags from the input, leaving only the text content.
38pub fn strip_html_tags(input: &str) -> String {
39    HTML_TAG_REGEX.replace_all(input, "").into_owned()
40}
41
42/// Truncates UTF-8 text to a max byte length at a valid char boundary.
43pub fn truncate_utf8_at_boundary(s: &str, max_bytes: usize) -> &str {
44    if s.len() <= max_bytes {
45        return s;
46    }
47    let mut idx = max_bytes;
48    while idx > 0 && !s.is_char_boundary(idx) {
49        idx -= 1;
50    }
51    &s[..idx]
52}
53
54#[cfg(test)]
55mod test {
56    use super::*;
57
58    #[test]
59    fn ietf_language_code_validation_works() {
60        // Invalid scenarios
61        assert!(!is_ietf_language_code_like(""));
62        assert!(!is_ietf_language_code_like("en_us"));
63        assert!(!is_ietf_language_code_like("en_US"));
64        assert!(!is_ietf_language_code_like("in-cans"));
65        assert!(!is_ietf_language_code_like("in-cans-ca"));
66
67        // Valid scenarios
68        assert!(is_ietf_language_code_like("en"));
69        assert!(is_ietf_language_code_like("eng"));
70        assert!(is_ietf_language_code_like("en-US"));
71        assert!(is_ietf_language_code_like("in-Cans-CA"));
72    }
73
74    #[test]
75    fn strip_html_tags_removes_all_tags() {
76        assert_eq!(
77            strip_html_tags("<em>Intro</em> to <strong>X</strong>"),
78            "Intro to X"
79        );
80        assert_eq!(strip_html_tags(r#"<a href="/x">link</a>"#), "link");
81        assert_eq!(strip_html_tags("plain text"), "plain text");
82        assert_eq!(strip_html_tags("<br>"), "");
83    }
84
85    #[test]
86    fn truncate_utf8_at_boundary_returns_original_when_short() {
87        let input = "heillä";
88        let result = truncate_utf8_at_boundary(input, 255);
89        assert_eq!(result, input);
90    }
91
92    #[test]
93    fn truncate_utf8_at_boundary_handles_finnish_characters() {
94        let input = format!("{}äz", "a".repeat(254));
95        let result = truncate_utf8_at_boundary(&input, 255);
96        assert_eq!(result.as_bytes().len(), 254);
97        assert!(result.is_char_boundary(result.len()));
98        assert_eq!(result, "a".repeat(254));
99    }
100
101    #[test]
102    fn truncate_utf8_at_boundary_handles_emoji() {
103        let input = format!("{}😀z", "a".repeat(254));
104        let result = truncate_utf8_at_boundary(&input, 255);
105        assert_eq!(result.as_bytes().len(), 254);
106        assert!(result.is_char_boundary(result.len()));
107        assert_eq!(result, "a".repeat(254));
108    }
109}
headless_lms_utils/strings.rs

headless_lms_utils/
strings.rs
