Update nested field in BigQuery table

Question

I am trying to perform what, you would think, is a trivial operation in BigQuery; I am trying to update a nested field in a BigQuery table that is the result of a 360 export

Answer 1

hits is an array, so you need to use an array subquery to assign to it. It would look something like this:

#standardSQL
UPDATE `dataset_name`.`ga_sessions_20170705`
SET hits =
  ARRAY(
    SELECT AS STRUCT * REPLACE(
      (SELECT AS STRUCT eventInfo.* REPLACE('some string' AS eventLabel)) AS eventInfo)
    FROM UNNEST(hits)
  )
WHERE TRUE;

Answer 2

Here is to mask PII data in ga sessions having an email

UPDATE
     `<project-id>.<dataset-name>.<table-name>`
 
SET hits =
  ARRAY(SELECT AS STRUCT * REPLACE (
    -- correcting pages here
    IF(REGEXP_CONTAINS(page.pagePath, r"@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
    ,STRUCT(
        REGEXP_REPLACE(page.pagePath, r"@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+", "[EMAIL]")
        ,page.pagePathLevel1
        ,page.pagePathLevel2
        ,page.pagePathLevel3
        ,page.pagePathLevel4
        ,page.hostname
        ,page.pageTitle
        ,page.searchKeyword
        ,page.searchCategory
    ), page) AS page)
    
    FROM UNNEST(hits)
  ) 
WHERE ( -- only relevant sessions
  SELECT COUNT(1) > 0 
  FROM UNNEST(hits) AS hits
  WHERE totals.visits = 1
    AND hits.type = 'PAGE'
    AND REGEXP_CONTAINS(hits.page.pagePath, r"@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+") = true
)

Answer 3

If you need to modify a given custom dimension you can use this:

#standardSQL
UPDATE `tablename`
SET hits = 
  ARRAY(
    SELECT AS STRUCT * REPLACE(
      ARRAY(
        SELECT AS STRUCT cd.index,
          CASE WHEN cd.index = index_number THEN 'new value'
          ELSE cd.value
          END
        FROM UNNEST(customDimensions) AS cd
      ) AS customDimensions)
    FROM UNNEST(hits) hit
  )
WHERE TRUE

But it takes a while to run.