String, str and Unicode text

Student names do not always fit one byte per letter. Rust stores String and str text as UTF-8. String owns a growable buffer; str is a text slice type, commonly accessed through &str. Understanding that distinction prevents incorrect indexing and unnecessary ownership transfers.

Use push_str to append borrowed text to a mutable String. len() counts bytes and chars() yields Unicode scalar values. Rust does not provide direct integer indexing for String because a byte position does not consistently identify a character. User-perceived grapheme clusters can contain multiple scalar values.

Connect this idea with references and read-only borrowing and slices and reference lifetimes.

Save the complete source as main.rs in the Rust environment. No standard input is needed.

Rust 2024 · complete program

fn main() {
  let mut name = String::from("Zoë");
  println!("Bytes: {}", name.len());
  println!("Scalars: {}", name.chars().count());
  name.push_str(" - A");
  println!("{name}");
}

Expected output

Bytes: 4
Scalars: 3
Zoë - A

Follow the execution

Z and o each use one byte, while ë uses two in UTF-8. The text therefore has four bytes and three scalar values. Appending the class label grows the owned String. The original accent remains valid text.

Try the complete program in the Rust compiler. For tests, Cargo commands and local files, use the specified local environment.

Your task: Is name.len() a reliable count of letters shown to a reader?

Rust 2024 · complete program

fn main() {
  let mut name = String::from("Zoë");
  println!("Bytes: {}", name.len());
  println!("Scalars: {}", name.chars().count());
  name.push_str(" - A");
  println!("{name}");
}

Worked answer

No. It counts UTF-8 bytes. Even chars().count() counts scalar values rather than every user-perceived grapheme cluster.

Slicing a string with byte positions can panic if a boundary splits a UTF-8 encoding. Grapheme processing needs a suitable library. Prefer &str for a function that only reads text and String when the function needs owned, growable text.

API details: official Rust documentation for this boundary.

No. It counts UTF-8 bytes. Even chars().count() counts scalar values rather than every user-perceived grapheme cluster. Slicing a string with byte positions can panic if a boundary splits a UTF-8 encoding.

Further reading: the official Rust Book on string, str and unicode text.