OsStr::starts_with

[lolbinarycat]

Proposal

Problem statement

There is no easy and obviously correct way to check if an OsStr has some prefix.

Motivating examples or use cases

Checking if a file is a "dotfile" (ie. its filename starts with ascii .)

basic argument parsing through args_os.

Solution sketch

impl OsStr {
  pub fn starts_with<S: AsRef<OsStr>>(&self, s: &S) -> bool {
    self.as_encoded_bytes().starts_with(s.as_encoded_bytes())
}

Alternatives

• as_encoded_bytes().starts_with(s.as_bytes()): this is equivalent, but verbose and not immediately obvious as correct, as as_encoded_bytes can cause differences in behavior, and projects with a focus on correctness will likely want to avoid its use when possible. Even without the self-synchronization of utf-8, checking for a string prefix in this way is trivial, and thus should not divert reviewer attention.

Links and related work

What happens now?

This issue contains an API change proposal (or ACP) and is part of the libs-api team feature lifecycle. Once this issue is filed, the libs-api team will review open proposals as capability becomes available. Current response times do not have a clear estimate, but may be up to several months.

Possible responses

The libs team may respond in various different ways. First, the team will consider the problem (this doesn't require any concrete solution or alternatives to have been proposed):

• We think this problem seems worth solving, and the standard library might be the right place to solve it.
• We think that this probably doesn't belong in the standard library.

Second, if there's a concrete solution:

• We think this specific solution looks roughly right, approved, you or someone else should implement this. (Further review will still happen on the subsequent implementation PR.)
• We're not sure this is the right solution, and the alternatives or other materials don't give us enough information to be sure about that. Here are some questions we have that aren't answered, or rough ideas about alternatives we'd want to see discussed.

[Amanieu]

This was already accepted as part of #311, but it seems no tracking issue has been created for it. Feel free to open one.

[lolbinarycat]

Unless I'm missing something, #311 (and rust-lang/rfcs#2295) looks like it is unimplementable, due to the issue of negative patterns.

If that is the case, would it be possible to reopen this as an alternate, simpler proposal? Although that would require either changing the name of this function to something like OsStr::starts_with_str, or doing a de-RFC on RFC 2295.

[kennytm]

The "negative pattern" issue only affects FnMut(char) -> bool, everything else (str, char, [char]) is still part of OsStrPattern in #311.

So the difference between #311 and this ACP is only that this accepts an OsStr "pattern" but #311 does not (yet). The issue with accepting OsStr pattern is how to deal with unpaired surrogate in Windows though.

use std::ffi::{OsStr, OsString};
use std::os::windows::ffi::OsStringExt;

fn main() {
    let a = OsStr::new("\u{10000}");
    let b = OsString::from_wide(&[0xd800]);
    assert!(a.starts_with(&b));
    // ^ currently implementation in #611 using .as_encoded_bytes() will return false.
}