mirror of
https://github.com/uutils/coreutils
synced 2024-12-13 14:52:41 +00:00
Merge pull request #3740 from resistor/main
Implement wc fast paths that skip Unicode decoding.
This commit is contained in:
commit
2fa4d6a2bb
3 changed files with 71 additions and 36 deletions
|
@ -117,32 +117,21 @@ pub(crate) fn count_bytes_fast<T: WordCountable>(handle: &mut T) -> (usize, Opti
|
|||
}
|
||||
}
|
||||
|
||||
pub(crate) fn count_bytes_and_lines_fast<R: Read>(
|
||||
handle: &mut R,
|
||||
) -> (WordCount, Option<io::Error>) {
|
||||
let mut total = WordCount::default();
|
||||
let mut buf = [0; BUF_SIZE];
|
||||
loop {
|
||||
match handle.read(&mut buf) {
|
||||
Ok(0) => return (total, None),
|
||||
Ok(n) => {
|
||||
total.bytes += n;
|
||||
total.lines += bytecount::count(&buf[..n], b'\n');
|
||||
}
|
||||
Err(ref e) if e.kind() == ErrorKind::Interrupted => continue,
|
||||
Err(e) => return (total, Some(e)),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// Returns a WordCount that counts the number of Unicode characters encoded in UTF-8 read via a Reader.
|
||||
/// Returns a WordCount that counts the number of bytes, lines, and/or the number of Unicode characters encoded in UTF-8 read via a Reader.
|
||||
///
|
||||
/// This corresponds to the `-m` command line flag to wc.
|
||||
/// This corresponds to the `-c`, `-l` and `-m` command line flags to wc.
|
||||
///
|
||||
/// # Arguments
|
||||
///
|
||||
/// * `R` - A Reader from which the UTF-8 stream will be read.
|
||||
pub(crate) fn count_chars_fast<R: Read>(handle: &mut R) -> (WordCount, Option<io::Error>) {
|
||||
pub(crate) fn count_bytes_chars_and_lines_fast<
|
||||
R: Read,
|
||||
const COUNT_BYTES: bool,
|
||||
const COUNT_CHARS: bool,
|
||||
const COUNT_LINES: bool,
|
||||
>(
|
||||
handle: &mut R,
|
||||
) -> (WordCount, Option<io::Error>) {
|
||||
/// Mask of the value bits of a continuation byte
|
||||
const CONT_MASK: u8 = 0b0011_1111u8;
|
||||
/// Value of the tag bits (tag mask is !CONT_MASK) of a continuation byte
|
||||
|
@ -154,10 +143,18 @@ pub(crate) fn count_chars_fast<R: Read>(handle: &mut R) -> (WordCount, Option<io
|
|||
match handle.read(&mut buf) {
|
||||
Ok(0) => return (total, None),
|
||||
Ok(n) => {
|
||||
total.chars += buf[..n]
|
||||
.iter()
|
||||
.filter(|&&byte| (byte & !CONT_MASK) != TAG_CONT_U8)
|
||||
.count();
|
||||
if COUNT_BYTES {
|
||||
total.bytes += n;
|
||||
}
|
||||
if COUNT_CHARS {
|
||||
total.chars += buf[..n]
|
||||
.iter()
|
||||
.filter(|&&byte| (byte & !CONT_MASK) != TAG_CONT_U8)
|
||||
.count();
|
||||
}
|
||||
if COUNT_LINES {
|
||||
total.lines += bytecount::count(&buf[..n], b'\n');
|
||||
}
|
||||
}
|
||||
Err(ref e) if e.kind() == ErrorKind::Interrupted => continue,
|
||||
Err(e) => return (total, Some(e)),
|
||||
|
|
|
@ -13,7 +13,7 @@ extern crate uucore;
|
|||
mod count_fast;
|
||||
mod countable;
|
||||
mod word_count;
|
||||
use count_fast::{count_bytes_and_lines_fast, count_bytes_fast, count_chars_fast};
|
||||
use count_fast::{count_bytes_chars_and_lines_fast, count_bytes_fast};
|
||||
use countable::WordCountable;
|
||||
use unicode_width::UnicodeWidthChar;
|
||||
use utf8::{BufReadDecoder, BufReadDecoderError};
|
||||
|
@ -315,7 +315,7 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
) {
|
||||
// Specialize scanning loop to improve the performance.
|
||||
(false, false, false, false, false) => unreachable!(),
|
||||
(false, true, false, false, false) => count_chars_fast(&mut reader),
|
||||
|
||||
(true, false, false, false, false) => {
|
||||
// Fast path when only show_bytes is true.
|
||||
let (bytes, error) = count_bytes_fast(&mut reader);
|
||||
|
@ -327,10 +327,27 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
error,
|
||||
)
|
||||
}
|
||||
(false, false, true, false, false) | (true, false, true, false, false) => {
|
||||
// Fast path when only (show_bytes || show_lines) is true.
|
||||
count_bytes_and_lines_fast(&mut reader)
|
||||
|
||||
// Fast paths that can be computed without Unicode decoding.
|
||||
(false, false, true, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, false, false, true>(&mut reader)
|
||||
}
|
||||
(false, true, false, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, false, true, false>(&mut reader)
|
||||
}
|
||||
(false, true, true, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, false, true, true>(&mut reader)
|
||||
}
|
||||
(true, false, true, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, true, false, true>(&mut reader)
|
||||
}
|
||||
(true, true, false, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, true, true, false>(&mut reader)
|
||||
}
|
||||
(true, true, true, false, false) => {
|
||||
count_bytes_chars_and_lines_fast::<_, true, true, true>(&mut reader)
|
||||
}
|
||||
|
||||
(_, false, false, false, true) => {
|
||||
word_count_from_reader_specialized::<_, false, false, false, true>(reader)
|
||||
}
|
||||
|
@ -349,9 +366,6 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
(_, false, true, true, true) => {
|
||||
word_count_from_reader_specialized::<_, false, true, true, true>(reader)
|
||||
}
|
||||
(_, true, false, false, false) => {
|
||||
word_count_from_reader_specialized::<_, true, false, false, false>(reader)
|
||||
}
|
||||
(_, true, false, false, true) => {
|
||||
word_count_from_reader_specialized::<_, true, false, false, true>(reader)
|
||||
}
|
||||
|
@ -361,9 +375,6 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
(_, true, false, true, true) => {
|
||||
word_count_from_reader_specialized::<_, true, false, true, true>(reader)
|
||||
}
|
||||
(_, true, true, false, false) => {
|
||||
word_count_from_reader_specialized::<_, true, true, false, false>(reader)
|
||||
}
|
||||
(_, true, true, false, true) => {
|
||||
word_count_from_reader_specialized::<_, true, true, false, true>(reader)
|
||||
}
|
||||
|
|
|
@ -104,6 +104,33 @@ fn test_utf8_chars() {
|
|||
.stdout_is("442\n");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_utf8_bytes_chars() {
|
||||
new_ucmd!()
|
||||
.arg("-cm")
|
||||
.pipe_in_fixture("UTF_8_weirdchars.txt")
|
||||
.run()
|
||||
.stdout_is(" 442 513\n");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_utf8_bytes_lines() {
|
||||
new_ucmd!()
|
||||
.arg("-cl")
|
||||
.pipe_in_fixture("UTF_8_weirdchars.txt")
|
||||
.run()
|
||||
.stdout_is(" 25 513\n");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_utf8_bytes_chars_lines() {
|
||||
new_ucmd!()
|
||||
.arg("-cml")
|
||||
.pipe_in_fixture("UTF_8_weirdchars.txt")
|
||||
.run()
|
||||
.stdout_is(" 25 442 513\n");
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_utf8_chars_words() {
|
||||
new_ucmd!()
|
||||
|
|
Loading…
Reference in a new issue