package org.springblade.mdm.utils;
import org.apache.tika.detect.Detector;
import org.apache.tika.io.TikaInputStream;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.mime.MediaType;
import java.io.IOException;
import java.io.InputStream;

public class AnsiTextDetector  implements Detector {
	// 可打印字符范围（ASCII + 扩展ANSI字符）
	private static final int PRINTABLE_START = 32;
	private static final int PRINTABLE_END = 126;
	private static final int EXTENDED_ANSI_START = 128;
	private static final int EXTENDED_ANSI_END = 255;

	// 文本判定阈值：可打印字符占比 >= 80%
	private static final double TEXT_THRESHOLD = 0.8;

	@Override
	public MediaType detect(InputStream input, Metadata metadata) throws IOException {
		TikaInputStream tis = TikaInputStream.get(input);
		byte[] buffer = new byte[1024];
		int bytesRead = tis.read(buffer);
		if (bytesRead <= 0) {
			return MediaType.OCTET_STREAM;
		}

		int total = 0;
		int printable = 0;
		for (int i = 0; i < bytesRead; i++) {
			byte b = buffer[i];
			int unsigned = b & 0xFF; // 转为无符号字节
			total++;
			// 判断是否为可打印字符（含扩展ANSI）
			if ((unsigned >= PRINTABLE_START && unsigned <= PRINTABLE_END) ||
				(unsigned >= EXTENDED_ANSI_START && unsigned <= EXTENDED_ANSI_END)) {
				printable++;
			}
		}

		// 若可打印字符占比达标，判定为文本
		if ((double) printable / total >= TEXT_THRESHOLD) {
			return MediaType.TEXT_PLAIN;
		}
		return MediaType.OCTET_STREAM;
	}
}