regex_syntax/
debug.rs

1/// A type that wraps a single byte with a convenient fmt::Debug impl that
2/// escapes the byte.
3pub(crate) struct Byte(pub(crate) u8);
4
5impl core::fmt::Debug for Byte {
6    fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result {
7        // Special case ASCII space. It's too hard to read otherwise, so
8        // put quotes around it. I sometimes wonder whether just '\x20' would
9        // be better...
10        if self.0 == b' ' {
11            return write!(f, "' '");
12        }
13        // 10 bytes is enough to cover any output from ascii::escape_default.
14        let mut bytes = [0u8; 10];
15        let mut len = 0;
16        for (i, mut b) in core::ascii::escape_default(self.0).enumerate() {
17            // capitalize \xab to \xAB
18            if i >= 2 && b'a' <= b && b <= b'f' {
19                b -= 32;
20            }
21            bytes[len] = b;
22            len += 1;
23        }
24        write!(f, "{}", core::str::from_utf8(&bytes[..len]).unwrap())
25    }
26}
27
28/// A type that provides a human readable debug impl for arbitrary bytes.
29///
30/// This generally works best when the bytes are presumed to be mostly UTF-8,
31/// but will work for anything.
32///
33/// N.B. This is copied nearly verbatim from regex-automata. Sigh.
34pub(crate) struct Bytes<'a>(pub(crate) &'a [u8]);
35
36impl<'a> core::fmt::Debug for Bytes<'a> {
37    fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result {
38        write!(f, "\"")?;
39        // This is a sad re-implementation of a similar impl found in bstr.
40        let mut bytes = self.0;
41        while let Some(result) = utf8_decode(bytes) {
42            let ch = match result {
43                Ok(ch) => ch,
44                Err(byte) => {
45                    write!(f, r"\x{byte:02x}")?;
46                    bytes = &bytes[1..];
47                    continue;
48                }
49            };
50            bytes = &bytes[ch.len_utf8()..];
51            match ch {
52                '\0' => write!(f, "\\0")?,
53                // ASCII control characters except \0, \n, \r, \t
54                '\x01'..='\x08'
55                | '\x0b'
56                | '\x0c'
57                | '\x0e'..='\x19'
58                | '\x7f' => {
59                    write!(f, "\\x{:02x}", u32::from(ch))?;
60                }
61                '\n' | '\r' | '\t' | _ => {
62                    write!(f, "{}", ch.escape_debug())?;
63                }
64            }
65        }
66        write!(f, "\"")?;
67        Ok(())
68    }
69}
70
71/// Decodes the next UTF-8 encoded codepoint from the given byte slice.
72///
73/// If no valid encoding of a codepoint exists at the beginning of the given
74/// byte slice, then the first byte is returned instead.
75///
76/// This returns `None` if and only if `bytes` is empty.
77pub(crate) fn utf8_decode(bytes: &[u8]) -> Option<Result<char, u8>> {
78    fn len(byte: u8) -> Option<usize> {
79        if byte <= 0x7F {
80            return Some(1);
81        } else if byte & 0b1100_0000 == 0b1000_0000 {
82            return None;
83        } else if byte <= 0b1101_1111 {
84            Some(2)
85        } else if byte <= 0b1110_1111 {
86            Some(3)
87        } else if byte <= 0b1111_0111 {
88            Some(4)
89        } else {
90            None
91        }
92    }
93
94    if bytes.is_empty() {
95        return None;
96    }
97    let len = match len(bytes[0]) {
98        None => return Some(Err(bytes[0])),
99        Some(len) if len > bytes.len() => return Some(Err(bytes[0])),
100        Some(1) => return Some(Ok(char::from(bytes[0]))),
101        Some(len) => len,
102    };
103    match core::str::from_utf8(&bytes[..len]) {
104        Ok(s) => Some(Ok(s.chars().next().unwrap())),
105        Err(_) => Some(Err(bytes[0])),
106    }
107}
regex_syntax/debug.rs

regex_syntax/
debug.rs
