mirror of
https://github.com/RGBCube/uutils-coreutils
synced 2025-07-28 03:27:44 +00:00
wc: add lines() method for iterating over lines
Add the `WordCountable::lines()` method that returns an iterator over lines of a file-like object. This mirrors the `std::io::BufRead::lines()` method, with some minor differences due to the particular use case of `wc`. This commit also creates a new module, `countable.rs`, to contain the `WordCountable` trait and the new `Lines` struct returned by `lines()`.
This commit is contained in:
parent
0dfb3e36ab
commit
0a3e2216d7
2 changed files with 81 additions and 46 deletions
72
src/uu/wc/src/countable.rs
Normal file
72
src/uu/wc/src/countable.rs
Normal file
|
@ -0,0 +1,72 @@
|
|||
//! Traits and implementations for iterating over lines in a file-like object.
|
||||
//!
|
||||
//! This module provides a [`WordCountable`] trait and implementations
|
||||
//! for some common file-like objects. Use the [`WordCountable::lines`]
|
||||
//! method to get an iterator over lines of a file-like object.
|
||||
use std::fs::File;
|
||||
use std::io::{self, BufRead, BufReader, Read, StdinLock};
|
||||
|
||||
#[cfg(unix)]
|
||||
use std::os::unix::io::AsRawFd;
|
||||
|
||||
#[cfg(unix)]
|
||||
pub trait WordCountable: AsRawFd + Read {
|
||||
type Buffered: BufRead;
|
||||
fn lines(self) -> Lines<Self::Buffered>;
|
||||
}
|
||||
|
||||
#[cfg(not(unix))]
|
||||
pub trait WordCountable: Read {
|
||||
type Buffered: BufRead;
|
||||
fn lines(self) -> Lines<Self::Buffered>;
|
||||
}
|
||||
|
||||
impl WordCountable for StdinLock<'_> {
|
||||
type Buffered = Self;
|
||||
|
||||
fn lines(self) -> Lines<Self::Buffered>
|
||||
where
|
||||
Self: Sized,
|
||||
{
|
||||
Lines { buf: self }
|
||||
}
|
||||
}
|
||||
impl WordCountable for File {
|
||||
type Buffered = BufReader<Self>;
|
||||
|
||||
fn lines(self) -> Lines<Self::Buffered>
|
||||
where
|
||||
Self: Sized,
|
||||
{
|
||||
Lines {
|
||||
buf: BufReader::new(self),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// An iterator over the lines of an instance of `BufRead`.
|
||||
///
|
||||
/// Similar to [`io::Lines`] but yields each line as a `Vec<u8>` and
|
||||
/// includes the newline character (`\n`, the `0xA` byte) that
|
||||
/// terminates the line.
|
||||
///
|
||||
/// [`io::Lines`]:: io::Lines
|
||||
pub struct Lines<B> {
|
||||
buf: B,
|
||||
}
|
||||
|
||||
impl<B: BufRead> Iterator for Lines<B> {
|
||||
type Item = io::Result<Vec<u8>>;
|
||||
|
||||
fn next(&mut self) -> Option<Self::Item> {
|
||||
let mut line = Vec::new();
|
||||
|
||||
// reading from a TTY seems to raise a condition on, rather than return Some(0) like a file.
|
||||
// hence the option wrapped in a result here
|
||||
match self.buf.read_until(b'\n', &mut line) {
|
||||
Ok(0) => None,
|
||||
Ok(_n) => Some(Ok(line)),
|
||||
Err(e) => Some(Err(e)),
|
||||
}
|
||||
}
|
||||
}
|
|
@ -11,17 +11,17 @@
|
|||
extern crate uucore;
|
||||
|
||||
mod count_bytes;
|
||||
mod countable;
|
||||
use count_bytes::count_bytes_fast;
|
||||
use countable::WordCountable;
|
||||
|
||||
use clap::{App, Arg, ArgMatches};
|
||||
use thiserror::Error;
|
||||
|
||||
use std::cmp::max;
|
||||
use std::fs::File;
|
||||
use std::io::{self, BufRead, BufReader, Read, StdinLock, Write};
|
||||
use std::io::{self, Write};
|
||||
use std::ops::{Add, AddAssign};
|
||||
#[cfg(unix)]
|
||||
use std::os::unix::io::AsRawFd;
|
||||
use std::path::Path;
|
||||
use std::str::from_utf8;
|
||||
|
||||
|
@ -82,32 +82,6 @@ impl Settings {
|
|||
}
|
||||
}
|
||||
|
||||
#[cfg(unix)]
|
||||
trait WordCountable: AsRawFd + Read {
|
||||
type Buffered: BufRead;
|
||||
fn get_buffered(self) -> Self::Buffered;
|
||||
}
|
||||
#[cfg(not(unix))]
|
||||
trait WordCountable: Read {
|
||||
type Buffered: BufRead;
|
||||
fn get_buffered(self) -> Self::Buffered;
|
||||
}
|
||||
|
||||
impl WordCountable for StdinLock<'_> {
|
||||
type Buffered = Self;
|
||||
|
||||
fn get_buffered(self) -> Self::Buffered {
|
||||
self
|
||||
}
|
||||
}
|
||||
impl WordCountable for File {
|
||||
type Buffered = BufReader<Self>;
|
||||
|
||||
fn get_buffered(self) -> Self::Buffered {
|
||||
BufReader::new(self)
|
||||
}
|
||||
}
|
||||
|
||||
#[derive(Debug, Default, Copy, Clone)]
|
||||
struct WordCount {
|
||||
bytes: usize,
|
||||
|
@ -270,25 +244,16 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
let mut byte_count: usize = 0;
|
||||
let mut char_count: usize = 0;
|
||||
let mut longest_line_length: usize = 0;
|
||||
let mut raw_line = Vec::new();
|
||||
let mut ends_lf: bool;
|
||||
|
||||
// reading from a TTY seems to raise a condition on, rather than return Some(0) like a file.
|
||||
// hence the option wrapped in a result here
|
||||
let mut buffered_reader = reader.get_buffered();
|
||||
loop {
|
||||
match buffered_reader.read_until(LF, &mut raw_line) {
|
||||
Ok(n) => {
|
||||
if n == 0 {
|
||||
break;
|
||||
}
|
||||
}
|
||||
Err(ref e) => {
|
||||
if !raw_line.is_empty() {
|
||||
show_warning!("Error while reading {}: {}", path, e);
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
for line_result in reader.lines() {
|
||||
let raw_line = match line_result {
|
||||
Ok(l) => l,
|
||||
Err(e) => {
|
||||
show_warning!("Error while reading {}: {}", path, e);
|
||||
continue;
|
||||
}
|
||||
};
|
||||
|
||||
|
@ -317,8 +282,6 @@ fn word_count_from_reader<T: WordCountable>(
|
|||
longest_line_length = current_char_count - (ends_lf as usize);
|
||||
}
|
||||
}
|
||||
|
||||
raw_line.truncate(0);
|
||||
}
|
||||
|
||||
Ok(WordCount {
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue