```rust
fn main() {
let output: Cow
// `output` contains no NUL bytes.
} ```
There's also a MString and mstr struct.
These are the counterparts to String and str within the standard library.
rust
fn main() {
let data = mstr::from_utf8(b"\0");
assert_eq!(data.len(), 2);
}
This crate allows converting UTF-8 to and from MUTF-8.
Some data formats, such as the JVM classfile, make use of an altered UTF-8 encoding.
This one in particular is the MUTF-8 variant.
It allows a NUL byte to be encoded without using the NUL byte itself.
The algorithm itself is done, and useable.
It works as well as any other.
The reason I still call this crate WIP is because of the two String structs.
I'm not happy with them.
I do use this crate for a couple of projects, but none of them make use of the structs themselves.
I typically use this crate as just a jump from a [u8] to a Cow<str>.
So, until I work out where I want to go with this crate, it's probably going to stay like this.