两种查找算法的比较

Posted 2021-02-16 s0611163

1、普通查找：双层循环遍历，第二层循环中找到即break，查找时间复杂度O(M*N/2)

List<PtCameraInfo> cameraList = new List<PtCameraInfo>();
List<string> cameraIdList = dataIds.Split(‘,‘).ToList();
List<PtCameraInfo> oldList = this.cameraList.Cameras.ToList();

for (int i = 0; i < cameraIdList.Count; i++)
{
    string cameraId = cameraIdList[i];
    for (int j = 0; j < _cameraList.Count; j++)
    {
        PtCameraInfo camera = _cameraList[j];
        if (camera.ID == cameraId && !oldList.Exists(a => a.ID == camera.ID))
        {
            cameraList.Add(camera);
            break;
        }
    }
}

View Code

2、高效查找：排序虽然费时，但数量级低，所以耗时很低，排序时间复杂度O(M*logM+N*logN)，查找过程也是双层循环，但第二层循环比较省时，查找时间复杂度O(M*logN)，总的时间复杂度：O(M*logM+N*logN+M*logN)

List<PtCameraInfo> cameraList = new List<PtCameraInfo>();
List<string> cameraIdList = dataIds.Split(‘,‘).ToList();
List<PtCameraInfo> oldList = this.cameraList.Cameras.ToList();

//排序
this._cameraList.Sort(new Comparison<PtCameraInfo>((a, b) =>
{
    return string.Compare(a.ID.PadLeft(36, ‘0‘), b.ID.PadLeft(36, ‘0‘));
}));
cameraIdList.Sort((a, b) =>
{
    return string.Compare(a.PadLeft(36, ‘0‘), b.PadLeft(36, ‘0‘));
});

//高效查找
int k = 0;
for (int i = 0; i < cameraIdList.Count; i++)
{
    string cameraId = cameraIdList[i];
    for (int j = k; j < _cameraList.Count; j++)
    {
        PtCameraInfo camera = _cameraList[j];
        if (camera.ID == cameraId && !oldList.Exists(a => a.ID == camera.ID))
        {
            cameraList.Add(camera);
            k = j;
            break;
        }
    }
}

View Code

说明：cameraList的数量级是2万，当cameraIdList数量从1到大约1000时，高效查找耗时始终是0.1秒多左右，而普通查找，当cameraIdList数量很少时，很快，0.05秒，当cameraIdList数量接近1000时，大约2秒。