Function detectEncoding

src/utils/encodings.js:92–152

(buffer)

}

export async function detectEncoding(buffer) {
	if (!buffer || buffer.byteLength === 0) {
		const def = settings.value.defaultFileEncoding;
		return def === "auto" ? "UTF-8" : def || "UTF-8";
	}

	const bytes = new Uint8Array(buffer);

	const bomEncoding = detectBOM(bytes);
	if (bomEncoding) return bomEncoding;

	const sample = bytes.subarray(0, Math.min(2048, bytes.length));
	let nulls = 0,
		ascii = 0;

	for (const byte of sample) {
		if (byte === 0) nulls++;
		else if (byte < 0x80) ascii++;
	}

	if (nulls > sample.length * 0.3) return "UTF-16LE";

	if (isValidUTF8(sample)) return "UTF-8";

	const encodings = [
		...new Set([
			"UTF-8",
			settings.value.defaultFileEncoding === "auto"
				? "UTF-8"
				: settings.value.defaultFileEncoding || "UTF-8",
			"windows-1252",
			"ISO-8859-1",
		]),
	];

	const testSample = sample.subarray(0, 512);
	const testBuffer = testSample.buffer.slice(
		testSample.byteOffset,
		testSample.byteOffset + testSample.byteLength,
	);

	for (const encoding of encodings) {
		try {
			const encodingObj = getEncoding(encoding);
			if (!encodingObj) continue;

			const text = await execDecode(testBuffer, encodingObj.name);
			if (
				!text.includes("\uFFFD") &&
				!/[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]/.test(text)
			) {
				return encoding;
			}
		} catch (error) {
			continue;
		}
	}