Skip to content

Commit

Permalink
Improve escape methods.
Browse files Browse the repository at this point in the history
  • Loading branch information
@reitermarkus
reitermarkus committed May 9, 2024
1 parent 16981ba commit 4edf12d
Show file tree
Hide file tree
Showing 3 changed files with 30 additions and 40 deletions.
2 changes: 1 addition & 1 deletion library/core/src/ascii.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -102,7 +102,7 @@ impl EscapeDefault {

#[inline]
pub(crate) fn empty() -> Self {
EscapeDefault(escape::EscapeIterInner::empty())
Self(escape::EscapeIterInner::empty())
}

#[inline]
Expand Down
7 changes: 1 addition & 6 deletions library/core/src/char/mod.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -220,19 +220,14 @@ pub struct EscapeDefault(escape::EscapeIterInner<10>);
impl EscapeDefault {
#[inline]
const fn printable(c: ascii::Char) -> Self {
Self::ascii(c.to_u8())
Self(escape::EscapeIterInner::ascii(c.to_u8()))
}

#[inline]
const fn backslash(c: ascii::Char) -> Self {
Self(escape::EscapeIterInner::backslash(c))
}

#[inline]
const fn ascii(c: u8) -> Self {
Self(escape::EscapeIterInner::ascii(c))
}

#[inline]
const fn unicode(c: char) -> Self {
Self(escape::EscapeIterInner::unicode(c))
Expand Down
61 changes: 28 additions & 33 deletions library/core/src/escape.rs
View file
Original file line number Diff line number Diff line change
Expand Up @@ -7,21 +7,21 @@ use crate::ops::Range;
const HEX_DIGITS: [ascii::Char; 16] = *b"0123456789abcdef".as_ascii().unwrap();

#[inline]
const fn backslash<const N: usize>(a: ascii::Char) -> ([ascii::Char; N], u8) {
const fn backslash<const N: usize>(a: ascii::Char) -> ([ascii::Char; N], Range<u8>) {
const { assert!(N >= 2) };

let mut output = [ascii::Char::Null; N];

output[0] = ascii::Char::ReverseSolidus;
output[1] = a;

(output, 2)
(output, 0..2)
}

/// Escapes an ASCII character.
///
/// Returns a buffer and the length of the escaped representation.
const fn escape_ascii<const N: usize>(byte: u8) -> ([ascii::Char; N], u8) {
const fn escape_ascii<const N: usize>(byte: u8) -> ([ascii::Char; N], Range<u8>) {
const { assert!(N >= 4) };

match byte {
Expand All @@ -38,7 +38,7 @@ const fn escape_ascii<const N: usize>(byte: u8) -> ([ascii::Char; N], u8) {
&& !byte.is_ascii_control()
{
output[0] = c;
(output, 1)
(output, 0..1)
} else {
let hi = HEX_DIGITS[(byte >> 4) as usize];
let lo = HEX_DIGITS[(byte & 0xf) as usize];
Expand All @@ -48,7 +48,7 @@ const fn escape_ascii<const N: usize>(byte: u8) -> ([ascii::Char; N], u8) {
output[2] = hi;
output[3] = lo;

(output, 4)
(output, 0..4)
}
}
}
Expand All @@ -57,34 +57,28 @@ const fn escape_ascii<const N: usize>(byte: u8) -> ([ascii::Char; N], u8) {
/// Escapes a character `\u{NNNN}` representation.
///
/// Returns a buffer and the length of the escaped representation.
const fn escape_unicode<const N: usize>(c: char) -> ([ascii::Char; N], u8) {
const { assert!(N >= 10) };
const fn escape_unicode<const N: usize>(c: char) -> ([ascii::Char; N], Range<u8>) {
const { assert!(N >= 10 && N < u8::MAX as usize) };

let c = c as u32;
let c = u32::from(c);

// OR-ing `1` ensures that for `c == 0` the code computes that
// one digit should be printed.
let u_len = (8 - (c | 1).leading_zeros() / 4) as usize;

let closing_paren_offset = 3 + u_len;
let start = (c | 1).leading_zeros() as usize / 4 - 2;

let mut output = [ascii::Char::Null; N];

output[0] = ascii::Char::ReverseSolidus;
output[1] = ascii::Char::SmallU;
output[2] = ascii::Char::LeftCurlyBracket;

output[3 + u_len.saturating_sub(6)] = HEX_DIGITS[((c >> 20) & 0x0f) as usize];
output[3 + u_len.saturating_sub(5)] = HEX_DIGITS[((c >> 16) & 0x0f) as usize];
output[3 + u_len.saturating_sub(4)] = HEX_DIGITS[((c >> 12) & 0x0f) as usize];
output[3 + u_len.saturating_sub(3)] = HEX_DIGITS[((c >> 8) & 0x0f) as usize];
output[3 + u_len.saturating_sub(2)] = HEX_DIGITS[((c >> 4) & 0x0f) as usize];
output[3 + u_len.saturating_sub(1)] = HEX_DIGITS[((c >> 0) & 0x0f) as usize];

output[closing_paren_offset] = ascii::Char::RightCurlyBracket;

let len = (closing_paren_offset + 1) as u8;
(output, len)
output[3] = HEX_DIGITS[((c >> 20) & 15) as usize];
output[4] = HEX_DIGITS[((c >> 16) & 15) as usize];
output[5] = HEX_DIGITS[((c >> 12) & 15) as usize];
output[6] = HEX_DIGITS[((c >> 8) & 15) as usize];
output[7] = HEX_DIGITS[((c >> 4) & 15) as usize];
output[8] = HEX_DIGITS[((c >> 0) & 15) as usize];
output[9] = ascii::Char::RightCurlyBracket;
output[start + 0] = ascii::Char::ReverseSolidus;
output[start + 1] = ascii::Char::SmallU;
output[start + 2] = ascii::Char::LeftCurlyBracket;

(output, (start as u8)..(N as u8))
}

/// An iterator over an fixed-size array.
Expand All @@ -102,25 +96,26 @@ pub(crate) struct EscapeIterInner<const N: usize> {

impl<const N: usize> EscapeIterInner<N> {
pub const fn backslash(c: ascii::Char) -> Self {
let (data, len) = backslash(c);
Self { data, alive: 0..len }
let (data, range) = backslash(c);
Self { data, alive: range }
}

pub const fn ascii(c: u8) -> Self {
let (data, len) = escape_ascii(c);
Self { data, alive: 0..len }
let (data, range) = escape_ascii(c);
Self { data, alive: range }
}

pub const fn unicode(c: char) -> Self {
let (data, len) = escape_unicode(c);
Self { data, alive: 0..len }
let (data, range) = escape_unicode(c);
Self { data, alive: range }
}

#[inline]
pub const fn empty() -> Self {
Self { data: [ascii::Char::Null; N], alive: 0..0 }
}

#[inline]
pub fn as_ascii(&self) -> &[ascii::Char] {
// SAFETY: `self.alive` is guaranteed to be a valid range for indexing `self.data`.
unsafe {
Expand Down

0 comments on commit 4edf12d

Please sign in to comment.