如何使用Rust从stdin创建一个高效的char迭代器? [英] How can I create an efficient iterator of chars from stdin with Rust?
问题描述
现在, Read :: chars
迭代器已被正式弃用,在像stdin这样的 Reader
来的字符上获取迭代器的正确方法是什么,而又不将整个流都读入内存?
Now that the Read::chars
iterator has been officially deprecated, what is the the proper way to obtain an iterator over the chars coming from a Reader
like stdin without reading the entire stream into memory?
推荐答案
正如其他人提到的那样,可以复制已弃用的 Read :: chars
以在您自己的代码中使用。对于我来说,这是否真正理想取决于您的用例-对我来说,这证明目前已经足够好了,尽管我的应用程序在不久的将来可能会超过此方法。
As a couple others have mentioned, it is possible to copy the deprecated implementation of Read::chars
for use in your own code. Whether this is truly ideal or not will depend on your use-case--for me, this proved to be good enough for now although it is likely that my application will outgrow this approach in the near-future.
为说明如何完成此操作,让我们来看一个具体示例:
To illustrate how this can be done, let's look at a concrete example:
use std::io::{self, Error, ErrorKind, Read};
use std::result;
use std::str;
struct MyReader<R> {
inner: R,
}
impl<R: Read> MyReader<R> {
fn new(inner: R) -> MyReader<R> {
MyReader {
inner,
}
}
#[derive(Debug)]
enum MyReaderError {
NotUtf8,
Other(Error),
}
impl<R: Read> Iterator for MyReader<R> {
type Item = result::Result<char, MyReaderError>;
fn next(&mut self) -> Option<result::Result<char, MyReaderError>> {
let first_byte = match read_one_byte(&mut self.inner)? {
Ok(b) => b,
Err(e) => return Some(Err(MyReaderError::Other(e))),
};
let width = utf8_char_width(first_byte);
if width == 1 {
return Some(Ok(first_byte as char));
}
if width == 0 {
return Some(Err(MyReaderError::NotUtf8));
}
let mut buf = [first_byte, 0, 0, 0];
{
let mut start = 1;
while start < width {
match self.inner.read(&mut buf[start..width]) {
Ok(0) => return Some(Err(MyReaderError::NotUtf8)),
Ok(n) => start += n,
Err(ref e) if e.kind() == ErrorKind::Interrupted => continue,
Err(e) => return Some(Err(MyReaderError::Other(e))),
}
}
}
Some(match str::from_utf8(&buf[..width]).ok() {
Some(s) => Ok(s.chars().next().unwrap());
None => Err(MyReaderError::NotUtf8),
})
}
}
以上代码也需要 read_one_byte
和 utf8_char_width
待实施。这些应该类似于:
The above code also requires read_one_byte
and utf8_char_width
to be implemented. Those should look something like:
static UTF8_CHAR_WIDTH: [u8; 256] = [
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, // 0x1F
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, // 0x3F
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, // 0x5F
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, // 0x7F
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 0x9F
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 0xBF
0,0,2,2,2,2,2,2,2,2,2,2,2,2,2,2,
2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, // 0xDF
3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3, // 0xEF
4,4,4,4,4,0,0,0,0,0,0,0,0,0,0,0, // 0xFF
];
fn utf8_char_width(b: u8) -> usize {
return UTF8_CHAR_WIDTH[b as usize] as usize;
}
fn read_one_byte(reader: &mut Read) -> Option<io::Result<u8>> {
let mut buf = [0];
loop {
return match reader.read(&mut buf) {
Ok(0) => None,
Ok(..) => Some(Ok(buf[0])),
Err(ref e) if e.kind() == ErrorKind::Interrupted => continue,
Err(e) => Some(Err(e)),
};
}
}
现在我们可以使用 MyReader
实现可在某些阅读器上生成 char
s的迭代器,例如 io :: stdin :: Stdin
:
Now we can use the MyReader
implementation to produce an iterator of char
s over some reader, like io::stdin::Stdin
:
fn main() {
let stdin = io::stdin();
let mut reader = MyReader::new(stdin.lock());
for c in reader {
println!("{}", c);
}
}
在原始问题线程。 一个值得关注的问题但值得指出的是,该迭代器将无法正确处理非UTF-8编码的流。
The limitations of this approach are discussed at length in the original issue thread. One particular concern worth pointing out however is that this iterator will not handle non-UTF-8 encoded streams correctly.
这篇关于如何使用Rust从stdin创建一个高效的char迭代器?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!