Hmbown/CodeWhale · warning · io::Error

file appears to be binary

Error message

file appears to be binary

What it means

read_text_prefix reads the beginning of a file for @-mention content and rejects files whose buffered bytes contain a NUL byte (a strong binary signal). This InvalidData error ('file appears to be binary') is thrown before attempting UTF-8 conversion so binary media is never injected as text.

Solutions

1. Mention a text file instead, or convert the binary asset to text/summary first
2. Check the file type before mentioning (e.g. `file` command) and skip binary outputs
3. Extend is_media_path coverage if a known binary extension slips through
4. Handle InvalidData from read_file_content and show a 'binary file skipped' note

Example fix

// before
let (text, _) = read_file_content(&path)?;
// after
let (text, _) = match read_file_content(&path) {
    Ok(t) => t,
    Err(e) if e.kind() == io::ErrorKind::InvalidData => return Ok("[binary file omitted]".into()),
    Err(e) => return Err(e.into()),
};

let head = std::fs::read(&path)?;
if head.iter().take(8192).any(|&b| b == 0) {
    // binary; skip before calling read_file_content
}

fn looks_like_text(bytes: &[u8]) -> bool {
    !bytes.iter().take(8192).any(|&b| b == 0)
}

match read_file_content(&path) {
    Err(e) if e.kind() == io::ErrorKind::InvalidData => "[binary file omitted]".to_string(),
    other => other?.0,
}

When it happens

Trigger: Calling read_file_content / read_text_prefix on a file whose first bytes include a 0x00 byte — images, compiled objects, archives, PDFs, or any binary blob that is not filtered by is_media_path.

Common situations: Mentioning a binary file whose extension is not in the media-extension list (e.g. extensionless executables, .dat, .bin); a corrupted text file with embedded NULs.

Understand the failure class

Background: "failed to read file", EACCES, ENOENT and "could not read <path>" errors: when a program can't read a file from disk — this error's family across 49 libraries.

Related errors

• file is not UTF-8
• stream did not contain valid UTF-8
• artifact name is not portable UTF-8
• Cannot open session : its queued input is already open in…
• Checkpoint schema v is newer than supported v

AI-assisted analysis of Hmbown/CodeWhale@73e0f67d83 (2026-09-22). Data as JSON: /api/errors/ca440812b259277b. Report an issue: GitHub.

Appendix: source

Thrown at crates/tui/src/tui/file_mention.rs:1413

        .take(MAX_MENTION_FILE_BYTES + 1)
        .read_to_end(&mut buffer)?;
    let truncated = buffer.len() as u64 > MAX_MENTION_FILE_BYTES;
    if truncated {
        buffer.truncate(MAX_MENTION_FILE_BYTES as usize);
        // Round down to the nearest valid UTF-8 character boundary so a
        // multi-byte sequence (CJK, emoji, etc.) is never split at the cut point.
        // Only adjust when error_len() is None — that means truncation landed
        // mid-sequence (incomplete tail).  A Some(_) error_len means the file
        // genuinely contains invalid UTF-8 bytes; leave the buffer intact so
        // the from_utf8 call below returns the correct "file is not UTF-8" error.
        if let Err(e) = std::str::from_utf8(&buffer)
            && e.error_len().is_none()
        {
            buffer.truncate(e.valid_up_to());
        }
    }
    if buffer.contains(&0) {
        return Err(std::io::Error::new(
            std::io::ErrorKind::InvalidData,
            "file appears to be binary",
        ));
    }
    let text = std::str::from_utf8(&buffer)
        .map_err(|_| std::io::Error::new(std::io::ErrorKind::InvalidData, "file is not UTF-8"))?
        .to_string();
    Ok((text, truncated))
}

fn is_media_path(path: &Path) -> bool {
    let Some(ext) = path.extension().and_then(|ext| ext.to_str()) else {
        return false;
    };
    matches!(
        ext.to_ascii_lowercase().as_str(),
        "png"
            | "jpg"

View on GitHub (pinned to 73e0f67d83)