whitfin / bytelines Goto Github PK

There are many other popular runtimes besides Tokio, we should add modules for these runtimes too. The two in mind are smol and async_std but there are probably others.

At the same time all of these runtimes should be off by default, and opt-in via features in the dependencies. Obviously this will all be a major bump, so we can consider any changes in APIs here too.

It would be great if there would be a function with identical usage, but that also includes the byte offset in the iterator item. I need this because i need to remove some characters from a specific line, but cant because the byte offset i have (from .take(n).fold(0, |count, v| count + v.len())) would not take into consideration the bytes of the actual newlines characters.

There are many MIT variants out there and it would be very helpful if you would put specific one into the LICENSE file and release new version. This is basically requirement for packaging this crate in Fedora.

At the very least we should include a counterpart for Tokio, and perhaps even a Stream implementation if possible.

use std::io;
extern crate bytelines;
use bytelines::*;

fn main() {
    let stdin = io::stdin();
    for _ in stdin.lock().byte_lines() {}
}

$ echo | cargo run
thread 'main' panicked at 'attempt to subtract with overflow', …/bytelines-2.2.1/src/lib.rs:129:36

And the line that panics is:

// also "pop" a leading \r
if self.buffer[n - 1] == b'\r' {

Apologies if I'm not understanding some Rust conventions but it seems to me that that the docs are misleading about the types yielded when iterating over lines. E.g.

// do something with the line, which is &[u8]

whereas really (AIUI) it's Result<&[u8], Error>. The same is true of several of the comments (and thus in the official docs on crates.io).