// sample-unicode-bom-crlf.rs: non-ASCII text, saved with a UTF-8 byte
// order mark and Windows (CRLF) line endings.
// Free sample from 99helpers.com/file-types/rs
//
// rustc skips the byte order mark and treats CRLF like LF, so this file
// compiles and prints the same output as an LF copy without a BOM.
// Run it: rustc --edition 2024 sample-unicode-bom-crlf.rs && ./sample-unicode-bom-crlf

// 1. Identifiers may use non-ASCII letters (since Rust 1.53).
const GRÜẞE: &str = "Grüße aus Köln";

fn main() {
    // 2. A `str` is UTF-8. `len()` counts bytes, `chars().count()` counts
    //    Unicode scalar values; they differ for anything beyond ASCII.
    let samples = [
        "plain ASCII",
        GRÜẞE,
        "Ελληνικά",
        "日本語のテキスト",
        "مرحبا بالعالم",
        "emoji 🦀✨",
    ];
    for text in samples {
        println!(
            "{:>2} bytes {:>2} chars  {text}",
            text.len(),
            text.chars().count()
        );
    }

    // 3. A multi-line string literal. The file's CRLF line breaks are read
    //    as LF, so this string contains "\n" and no "\r".
    let poem = "first line
second line";
    assert!(!poem.contains('\r'));
    println!("{poem:?}");

    // 4. Escapes: \u{...} names a character by its code point.
    let crab = '\u{1F980}';
    println!(
        "{crab} is U+{:04X}, {} bytes in UTF-8",
        crab as u32,
        crab.len_utf8()
    );

    // 5. Slicing a string must land on a character boundary.
    //    `&GRÜẞE[..3]` would take "G", "r" and the first of the two bytes
    //    of "ü", and panic at run time, so check with is_char_boundary.
    let cut = (0..=3)
        .rev()
        .find(|&i| GRÜẞE.is_char_boundary(i))
        .unwrap_or(0);
    println!("first {cut} bytes: {:?}", &GRÜẞE[..cut]);
}