Skip to content

cmp: stop allocating for byte printing #153

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kov wants to merge 1 commit into
base: main
Choose a base branch
from
+59 −37
Open

cmp: stop allocating for byte printing #153

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
96 changes: 59 additions & 37 deletions src/cmp.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -500,12 +500,6 @@ pub fn main(opts: Peekable<ArgsOs>) -> ExitCode {
}
}

#[inline]
fn is_ascii_printable(byte: u8) -> bool {
let c = byte as char;
c.is_ascii() && !c.is_ascii_control()
}

#[inline]
fn format_octal(byte: u8, buf: &mut [u8; 3]) -> &str {
*buf = [b' ', b' ', b'0'];
Expand All @@ -525,32 +519,67 @@ fn format_octal(byte: u8, buf: &mut [u8; 3]) -> &str {
}

#[inline]
fn format_byte(byte: u8) -> String {
let mut byte = byte;
let mut quoted = vec![];

if !is_ascii_printable(byte) {
if byte >= 128 {
quoted.push(b'M');
quoted.push(b'-');
byte -= 128;
fn write_visible_byte(output: &mut Vec<u8>, byte: u8) -> usize {
Copy link
Collaborator

@sylvestre sylvestre Jan 1, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

write_visible_byte returns usize, but write_visible_byte_padded returns ()

match byte {
// Control characters: ^@, ^A, ..., ^_
0..=31 => {
output.push(b'^');
output.push(byte + 64);
2
}

if byte < 32 {
quoted.push(b'^');
byte += 64;
} else if byte == 127 {
quoted.push(b'^');
byte = b'?';
// Printable ASCII (space through ~)
32..=126 => {
output.push(byte);
1
}
// DEL: ^?
127 => {
output.extend_from_slice(b"^?");
2
}
// High bytes with control equivalents: M-^@, M-^A, ..., M-^_
128..=159 => {
output.push(b'M');
output.push(b'-');
output.push(b'^');
output.push(byte - 64);
4
}
// High bytes: M-<space>, M-!, ..., M-~
160..=254 => {
output.push(b'M');
output.push(b'-');
output.push(byte - 128);
3
}
// Byte 255: M-^?
255 => {
output.extend_from_slice(b"M-^?");
4
}
assert!((byte as char).is_ascii());
}
}

quoted.push(byte);
/// Writes a byte in visible form with right-padding to the specified width
#[inline]
fn write_visible_byte_padded(output: &mut Vec<u8>, byte: u8, width: usize) {
let display_width = write_visible_byte(output, byte);

// SAFETY: the checks and shifts we do above match what cat and GNU
// Add right-padding spaces
let padding = width.saturating_sub(display_width);
Copy link
Collaborator

@sylvestre sylvestre Jan 1, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could use extend_from_slice with a constant array instead of a loop

for _ in 0..padding {
output.push(b' ');
}
}

/// Formats a byte as a visible string (for non-performance-critical path)
#[inline]
fn format_visible_byte(byte: u8) -> String {
let mut result = Vec::with_capacity(4);
write_visible_byte(&mut result, byte);
// SAFETY: the checks and shifts in write_visible_byte match what cat and GNU
// cmp do to ensure characters fall inside the ascii range.
unsafe { String::from_utf8_unchecked(quoted) }
unsafe { String::from_utf8_unchecked(result) }
}

// This function has been optimized to not use the Rust fmt system, which
Expand Down Expand Up @@ -588,22 +617,15 @@ fn format_verbose_difference(

output.push(b' ');

let from_byte_str = format_byte(from_byte);
let from_byte_padding = 4 - from_byte_str.len();

output.extend_from_slice(from_byte_str.as_bytes());

for _ in 0..from_byte_padding {
output.push(b' ')
}
write_visible_byte_padded(output, from_byte, 4);

output.push(b' ');

output.extend_from_slice(format_octal(to_byte, &mut to_oct).as_bytes());

output.push(b' ');

output.extend_from_slice(format_byte(to_byte).as_bytes());
write_visible_byte(output, to_byte);

output.push(b'\n');
} else {
Expand Down Expand Up @@ -706,9 +728,9 @@ fn report_difference(from_byte: u8, to_byte: u8, at_byte: usize, at_line: usize,
print!(
" is {:>3o} {:char_width$} {:>3o} {:char_width$}",
from_byte,
format_byte(from_byte),
format_visible_byte(from_byte),
to_byte,
format_byte(to_byte)
format_visible_byte(to_byte)
);
}
println!();
Expand Down
Loading