Function isBinaryBuffer

src/utils/binary.ts:38–63

(buf: Buffer, sampleSize = 8000)

}

export function isBinaryBuffer(buf: Buffer, sampleSize = 8000): boolean {
  if (buf.length === 0) return false;

  // Fast path: known binary signatures that the byte heuristics would miss.
  if (hasBinaryMagic(buf)) return true;

  const sample = buf.subarray(0, Math.min(sampleSize, buf.length));

  // Heuristic 1: any null byte in the first sample → binary
  // (Valid utf-8 text files do not contain null bytes.)
  for (let i = 0; i < sample.length; i++) {
    if (sample[i] === 0x00) return true;
  }

  // Heuristic 2: high proportion of control / non-printable bytes
  let nonPrintable = 0;
  for (let i = 0; i < sample.length; i++) {
    const b = sample[i]!;
    // Tab(9), LF(10), CR(13) are printable; everything 0..31 except those is suspect.
    // 0x7F (DEL) is also suspect. 0x80+ is fine (utf-8 multi-byte).
    if ((b < 0x09) || (b > 0x0d && b < 0x20) || b === 0x7f) {
      nonPrintable++;
    }
  }
  return nonPrintable / sample.length > 0.3;
}

/** Quick check by extension for common binary file types. */
const BINARY_EXTENSIONS = new Set([