hadoop - 如何对Hive中存储的数据进行滞后/超前操作？

Question

是否可以对Hive中存储的数据进行滞后、超前操作？

任何指示将不胜感激!!!

Answer 1

您可以使用 LAG 函数获取之前的时间戳示例如下:

NVL( LAG (currentTimestamp, 1) OVER (PARTITION BY [partinioned columns] ORDER BY currentTimestamp ASC)  , 0 ) AS previousTimestamp;

与 LEAD 相同，默认情况下也在这里一行 NVL( LEAD (currentTimestamp, 1) OVER (PARTITION BY [分区列] ORDER BY currentTimestamp ASC) , 0 ) AS NextTimestamp;