parse_hyperlinks/
lib.rs

1//! Library and application for parsing hyperlinks and link reference
2//! definitions in Markdown, reStructuredText, Asciidoc and HTML format. The
3//! library implements the
4//! [CommonMark Specification 0.30](https://spec.commonmark.org/0.30/),
5//! [reStructuredText Markup Specification](https://docutils.sourceforge.io/docs/ref/rst/restructuredtext.html)
6//! (revision 8571, date 2020-10-28),
7//! [Asciidoctor User Manual, chapter 26](https://asciidoctor.org/docs/user-manual/#url) (date 2020-12-03),
8//! the
9//! [HTML 5.2: section 4.5.](https://www.w3.org/TR/html52/textlevel-semantics.html#the-a-element)
10//! specification
11//! and the [Wikitext v1.0.0](https://www.mediawiki.org/wiki/Specs/wikitext/1.0.0)
12//! specification.
13#![allow(dead_code)]
14
15pub mod iterator;
16pub mod parser;
17pub mod renderer;
18
19use nom::Err;
20use nom::IResult;
21use nom::error::Error;
22use nom::error::ErrorKind;
23use nom::error::ParseError;
24
25/// A parser similar to `nom::bytes::complete::take_until()`, except that this
26/// one does not stop at balanced opening and closing tags. It is designed to
27/// work inside the `nom::sequence::delimited()` parser.
28///
29/// # Basic usage
30/// ```
31/// use nom::bytes::complete::tag;
32/// use nom::sequence::delimited;
33/// use nom::Parser;
34/// use parse_hyperlinks::take_until_unbalanced;
35///
36/// let mut parser = delimited(tag("<"), take_until_unbalanced('<', '>'), tag(">"));
37/// assert_eq!(parser.parse("<<inside>inside>abc"), Ok(("abc", "<inside>inside")));
38/// ```
39/// It skips nested brackets until it finds an extra unbalanced closing bracket. Escaped brackets
40/// like `\<` and `\>` are not considered as brackets and are not counted. This function is
41/// very similar to `nom::bytes::complete::take_until(">")`, except it also takes nested brackets.
42pub fn take_until_unbalanced(
43    opening_bracket: char,
44    closing_bracket: char,
45) -> impl Fn(&str) -> IResult<&str, &str> {
46    move |i: &str| {
47        let mut index = 0;
48        let mut bracket_counter = 0;
49        while let Some(n) = &i[index..].find(&[opening_bracket, closing_bracket, '\\'][..]) {
50            index += n;
51            let mut it = i[index..].chars();
52            match it.next() {
53                Some('\\') => {
54                    // Skip the escape char `\`.
55                    index += '\\'.len_utf8();
56                    // Skip also the following char.
57                    if let Some(c) = it.next() {
58                        index += c.len_utf8();
59                    }
60                }
61                Some(c) if c == opening_bracket => {
62                    bracket_counter += 1;
63                    index += opening_bracket.len_utf8();
64                }
65                Some(c) if c == closing_bracket => {
66                    // Closing bracket.
67                    bracket_counter -= 1;
68                    index += closing_bracket.len_utf8();
69                }
70                // Can not happen.
71                _ => unreachable!(),
72            };
73            // We found the unmatched closing bracket.
74            if bracket_counter == -1 {
75                // We do not consume it.
76                index -= closing_bracket.len_utf8();
77                return Ok((&i[index..], &i[0..index]));
78            };
79        }
80
81        if bracket_counter == 0 {
82            Ok(("", i))
83        } else {
84            Err(Err::Error(Error::from_error_kind(i, ErrorKind::TakeUntil)))
85        }
86    }
87}
88
89#[cfg(test)]
90mod tests {
91    use super::*;
92    use nom::error::ErrorKind;
93
94    #[test]
95    fn test_take_until_unmatched() {
96        assert_eq!(take_until_unbalanced('(', ')')("abc"), Ok(("", "abc")));
97        assert_eq!(
98            take_until_unbalanced('(', ')')("url)abc"),
99            Ok((")abc", "url"))
100        );
101        assert_eq!(
102            take_until_unbalanced('(', ')')("url)abc\\"),
103            Ok((")abc\\", "url"))
104        );
105        assert_eq!(
106            take_until_unbalanced('(', ')')("u()rl)abc"),
107            Ok((")abc", "u()rl"))
108        );
109        assert_eq!(
110            take_until_unbalanced('(', ')')("u(())rl)abc"),
111            Ok((")abc", "u(())rl"))
112        );
113        assert_eq!(
114            take_until_unbalanced('(', ')')("u\\(())rl)abc"),
115            Ok((")rl)abc", "u\\(()"))
116        );
117        assert_eq!(
118            take_until_unbalanced('(', ')')("u(()\\)rl)abc"),
119            Ok(("", "u(()\\)rl)abc"))
120        );
121        assert_eq!(
122            take_until_unbalanced('(', ')')("u(())r()l)abc"),
123            Ok((")abc", "u(())r()l"))
124        );
125        assert_eq!(
126            take_until_unbalanced('(', ')')("u(())r()labc"),
127            Ok(("", "u(())r()labc"))
128        );
129        assert_eq!(
130            take_until_unbalanced('(', ')')(r#"u\((\))r()labc"#),
131            Ok(("", r#"u\((\))r()labc"#))
132        );
133        assert_eq!(
134            take_until_unbalanced('(', ')')("u(())r(labc"),
135            Err(nom::Err::Error(nom::error::Error::new(
136                "u(())r(labc",
137                ErrorKind::TakeUntil
138            )))
139        );
140        assert_eq!(
141            take_until_unbalanced('€', 'ü')("€uü€€üürlüabc"),
142            Ok(("üabc", "€uü€€üürl"))
143        );
144    }
145}
parse_hyperlinks/lib.rs

parse_hyperlinks/
lib.rs
