1 /// A type that provides a human readable debug impl for arbitrary bytes.
2 ///
3 /// This generally works best when the bytes are presumed to be mostly UTF-8,
4 /// but will work for anything.
5 ///
6 /// N.B. This is copied nearly verbatim from regex-automata. Sigh.
7 pub(crate) struct Bytes<'a>(pub(crate) &'a [u8]);
8
9 impl<'a> core::fmt::Debug for Bytes<'a> {
fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result10 fn fmt(&self, f: &mut core::fmt::Formatter) -> core::fmt::Result {
11 write!(f, "\"")?;
12 // This is a sad re-implementation of a similar impl found in bstr.
13 let mut bytes = self.0;
14 while let Some(result) = utf8_decode(bytes) {
15 let ch = match result {
16 Ok(ch) => ch,
17 Err(byte) => {
18 write!(f, r"\x{:02x}", byte)?;
19 bytes = &bytes[1..];
20 continue;
21 }
22 };
23 bytes = &bytes[ch.len_utf8()..];
24 match ch {
25 '\0' => write!(f, "\\0")?,
26 // ASCII control characters except \0, \n, \r, \t
27 '\x01'..='\x08'
28 | '\x0b'
29 | '\x0c'
30 | '\x0e'..='\x19'
31 | '\x7f' => {
32 write!(f, "\\x{:02x}", u32::from(ch))?;
33 }
34 '\n' | '\r' | '\t' | _ => {
35 write!(f, "{}", ch.escape_debug())?;
36 }
37 }
38 }
39 write!(f, "\"")?;
40 Ok(())
41 }
42 }
43
44 /// Decodes the next UTF-8 encoded codepoint from the given byte slice.
45 ///
46 /// If no valid encoding of a codepoint exists at the beginning of the given
47 /// byte slice, then the first byte is returned instead.
48 ///
49 /// This returns `None` if and only if `bytes` is empty.
utf8_decode(bytes: &[u8]) -> Option<Result<char, u8>>50 pub(crate) fn utf8_decode(bytes: &[u8]) -> Option<Result<char, u8>> {
51 fn len(byte: u8) -> Option<usize> {
52 if byte <= 0x7F {
53 return Some(1);
54 } else if byte & 0b1100_0000 == 0b1000_0000 {
55 return None;
56 } else if byte <= 0b1101_1111 {
57 Some(2)
58 } else if byte <= 0b1110_1111 {
59 Some(3)
60 } else if byte <= 0b1111_0111 {
61 Some(4)
62 } else {
63 None
64 }
65 }
66
67 if bytes.is_empty() {
68 return None;
69 }
70 let len = match len(bytes[0]) {
71 None => return Some(Err(bytes[0])),
72 Some(len) if len > bytes.len() => return Some(Err(bytes[0])),
73 Some(1) => return Some(Ok(char::from(bytes[0]))),
74 Some(len) => len,
75 };
76 match core::str::from_utf8(&bytes[..len]) {
77 Ok(s) => Some(Ok(s.chars().next().unwrap())),
78 Err(_) => Some(Err(bytes[0])),
79 }
80 }
81