利用二分查找法查找数组中的连续的一段数据（前提数组有序）

最新推荐文章于 2024-05-05 09:41:45 发布

dutylemon

最新推荐文章于 2024-05-05 09:41:45 发布

阅读量89

点赞数 1

文章标签： elasticsearch java 大数据

本文链接：https://blog.csdn.net/qq_41842163/article/details/131294169

版权

##利用二分查找法查找数组中的连续的一段数据（前提数组有序）
最近在做一个项目，向ElasticSearch（简称ES）发送http请求，然后ES返回数据,由于ES返回的数据量非常大，又需要从其中查询业务数据，因此想起了二分查找法查询数据。
算法需求：任意给两个数据A，B（A<=B）,在数组arry[]中，找到集合(a,b)<=(A,B),返回子集合(a,b)。
代码如下：
二分查找算法：

    public Pair<Integer, Integer> Binary_search(List<Integer> srcList, int begin, int end) {
        int beginLeft = 0;
        int beginRight = srcList.size() - 1;
        int endLeft = 0;
        int endRight = srcList.size() - 1;
        if (begin <= srcList.get(beginLeft) && end >= srcList.get(endRight)) {
            return new Pair<>(beginLeft, endRight);
        }
        while ((beginRight - beginLeft > 1) || (endRight - endLeft > 1)) {
            int beginMid = (beginLeft + beginRight) / 2;
            int endMid = (endLeft + endRight) / 2;
            if (begin >= srcList.get(beginMid)) {
                beginLeft = beginMid;
            } else {
                beginRight = beginMid;
            }
            if (end >= srcList.get(endMid)) {
                endLeft = endMid;
            } else {
                endRight = endMid;
            }
        }
        long beginLeftDiff = Math.abs(srcList.get(beginLeft) - begin);
        long beginRightDiff = Math.abs(srcList.get(beginRight) - begin);
        long endLeftDiff = Math.abs(srcList.get(endLeft) - end);
        long endRightDiff = Math.abs(srcList.get(endRight) - end);
        int left = beginLeftDiff <= beginRightDiff ? beginLeft : beginRight;
        int right = endLeftDiff <= endRightDiff ? endLeft : endRight;
        return new Pair<>(left, right);
    }

测试函数：

@Test
    public void lemon4() {
        List<Integer> srcList = new ArrayList<>();
        Collections.addAll(srcList, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20);
        Pair<Integer, Integer> ret = Binary_search(srcList, 14, 16);
        List<Integer> retList = srcList.subList(ret.getKey(), ret.getValue() + 1);
        System.out.println(retList);
    }