Function read_source_file

src/sync.rs:12–43

Read a source file to a UTF-8 string, transparently handling UTF-16 LE/BE (detected via BOM). Returns an IO error only when the file genuinely cannot be read or decoded.

(path: &Path)

/// (detected via BOM). Returns an IO error only when the file genuinely cannot
/// be read or decoded.
pub fn read_source_file(path: &Path) -> std::io::Result<String> {
    let bytes = read_file_bytes(path)?;

    // UTF-16 LE BOM: FF FE
    if bytes.starts_with(&[0xFF, 0xFE]) {
        let u16s: Vec<u16> = bytes[2..]
            .chunks_exact(2)
            .map(|pair| u16::from_le_bytes([pair[0], pair[1]]))
            .collect();
        return String::from_utf16(&u16s)
            .map_err(|e| std::io::Error::new(std::io::ErrorKind::InvalidData, e));
    }

    // UTF-16 BE BOM: FE FF
    if bytes.starts_with(&[0xFE, 0xFF]) {
        let u16s: Vec<u16> = bytes[2..]
            .chunks_exact(2)
            .map(|pair| u16::from_be_bytes([pair[0], pair[1]]))
            .collect();
        return String::from_utf16(&u16s)
            .map_err(|e| std::io::Error::new(std::io::ErrorKind::InvalidData, e));
    }

    // Strip UTF-8 BOM if present, then validate
    let start = if bytes.starts_with(&[0xEF, 0xBB, 0xBF]) {
        3
    } else {
        0
    };
    String::from_utf8(bytes[start..].to_vec())
        .map_err(|e| std::io::Error::new(std::io::ErrorKind::InvalidData, e))
}

/// Reads a file's bytes, absorbing transient Windows file locks.
///