How do I change characters at a specific index within a string in rust?

Question

I am trying to change a single character at a specific index in a string, but I do not know how to in rust. For example, how would I change the 4th character in "hello world" to 'x', so that it would be "helxo world"?

Answer 1

The easiest way is to use the replace_range() method like this:

let mut hello = String::from("hello world");
hello.replace_range(3..4,"x");
println!("hello: {}", hello);

Output: hello: helxo world ( Playground )

Please note that this will panic if the range to be replaced does not start and end on UTF-8 codepoint boundaries. Eg this will panic:

let mut hello2 = String::from("hell😀 world");
hello2.replace_range(4..5,"x"); // panics because 😀 needs more than one byte in UTF-8

If you want to replace the nth UTF-8 code point, you have to do something like this:

pub fn main() {
    let mut hello = String::from("hell😀 world");
    hello.replace_range(
        hello
            .char_indices()
            .nth(4)
            .map(|(pos, ch)| (pos..pos + ch.len_utf8()))
            .unwrap(),
        "x",
    );
    println!("hello: {}", hello);
}

( Playground )

Answer 2

The standard way of representing a string in Rust is as a contiguous range of bytes encoded as a UTF-8 string. UTF-8 codepoints can be from one to 4 bytes long, so generally you can't simply replace one UTF-8 codepoint with another because the length might change. You also can't do simple pointer arithmetic to index into a Rust String to the nth character, because again codepoint encodings can be from 1 to 4 bytes long.

So one safe but slow way to do it would be like this, iterating through the characters of the source string, replacing the one you want, then creating a new string:

fn replace_nth_char(s: &str, idx: usize, newchar: char) -> String {
    s.chars().enumerate().map(|(i,c)| if i == idx { newchar } else { c }).collect()
}

But we can do it in O(1) if we manually make sure the old and new character are single-byte ascii.

fn replace_nth_char_safe(s: &str, idx: usize, newchar: char) -> String {
    s.chars().enumerate().map(|(i,c)| if i == idx { newchar } else { c }).collect()
}

fn replace_nth_char_ascii(s: &mut str, idx: usize, newchar: char) {
    let s_bytes: &mut [u8] = unsafe { s.as_bytes_mut() };
    assert!(idx < s_bytes.len());
    assert!(s_bytes[idx].is_ascii());
    assert!(newchar.is_ascii());
    // we've made sure this is safe.
    s_bytes[idx] = newchar as u8;
}
fn main() {
    let s = replace_nth_char_safe("Hello, world!", 3, 'x');
    assert_eq!(s, "Helxo, world!");
    
    let mut s = String::from("Hello, world!");
    replace_nth_char_ascii(&mut s, 3, 'x');
    assert_eq!(s, "Helxo, world!");
}

Keep in mind that idx parameter in replace_nth_char_ascii is not a character index, but instead a byte index. If there are any multibyte characters earlier in the string, then the byte index and the character index will not correspond.