TextStoreUtil

TextStoreUtil.java

package cn.agilean.valuekanban.shared;

import org.springframework.stereotype.Component;

import java.nio.charset.StandardCharsets;

public class TextStoreUtil {

    private int defaultMaxByteCountOfText = 500;

    public String getMaxStorableSubString(String source) {
        return getMaxStorableSubString(source, this.defaultMaxByteCountOfText);
    }

    public String getMaxStorableSubString(String source, int maxByteCount) {
        if (getByteCount(source) <= maxByteCount) {
            return source;
        }
        int cutIndex = getCutIndex(source, maxByteCount);
        if (cutIndex == -1) {
            // 全部按照汉字计算
            return source.substring(0, maxByteCount / 3);
        }
        return source.substring(0, cutIndex);
    }

    private static int getCutIndex(final String source, final int maxByteCount) {
        final int sourceLength = source.length();
        int cutIndex = sourceLength / 2;
        int lastOutOfBoundIndex = sourceLength;
        int maxRetry = sourceLength / 3;
        int retryCount = 0;
        while (true) {
            int byteCount = getByteCount(source.substring(0, cutIndex));
            if (byteCount > maxByteCount) {
                // 当前截取的字符串的字节数超出限制
                lastOutOfBoundIndex = cutIndex;
                cutIndex = cutIndex / 2;
            } else if (maxByteCount - byteCount > 3) {
                // 当前截取的字符串的字节数和限制差3个以内认为是可以了（实际这里最糟糕的情况下会少两个字母，数字）
                cutIndex = (cutIndex + lastOutOfBoundIndex) / 2;
            } else {
                return cutIndex;
            }
            retryCount++;
            if (retryCount > maxRetry) {
                return -1;
            }
        }
    }

    private static int getByteCount(String str) {
        return str.getBytes(StandardCharsets.UTF_8).length;
    }

}