Question

我想替换重复的字段（hits.customdimensions.value），而其余的数据应该保持不变。让我们说我想用MD5哈希hits.customdimension.value，其中hits.customdimension.index在1和5之间，并用散列的值替换原始值

如果是customdimensions.value，它似乎可以解决这个问题：

select x.* except (customdimensions), cd.index as cdindex, MD5(cd.value) as 
cdvalue
from `datasetx.tabley` x, unnest(customdimensions) as cd
where cd.index between 1 and 5

不幸的是，如果我想用hits.customdimensions.value做类似的事情，我必须在不使用（hit.customdimensions）之前使用except（命中），这会使其他列消失。

select x.* except (hits), hitcd.index as hitcdindex, MD5(hitcd.value) as 
hitcdvalue 
from `datasetx.tabley` x, unnest(hits) as hit, unnest(hit.customdimensions) 
as hitcd
where hitcd.index between 1 and 5

这有什么简单的解决方案吗？

Answer 1

下面完全保留整个表格结构/数据，并且hits.customdimension.value的哈希值hits.customdimension.index介于1到5之间

#standardSQL
SELECT * 
  REPLACE( ARRAY(
    SELECT AS STRUCT * 
      REPLACE( ARRAY(
        SELECT AS STRUCT 
          index, 
          IF(index BETWEEN 1 AND 5, TO_BASE64(MD5(value)), value) value 
        FROM UNNEST(customdimensions)      
      ) AS customdimensions
    )
    FROM UNNEST(hits)
  ) AS hits
)
FROM `datasetx.tabley`

替换嵌套/重复的字段

1 个答案: