BigQuery SQL：无法查询重复字段的叉积/

Question

我的目标是过滤所有访问者，仅分析客户（位于 customDimension.index =2 中，然后进一步仅过滤客户的特定类型的综合浏览量。

SELECT customDimensions.value AS CustomerID,
SUM(totals.pageviews) as page_views,
SUM(CASE WHEN hits.type = 'PAGE' AND hits.contentGroup.contentGroup2 = 'important' THEN 1 ELSE 0 END) AS important_pageviews
FROM `xxxxxxxx.ga_sessions_20180415`
WHERE customDimensions.index = 2
GROUP BY CustomerID 

我收到以下错误（使用 StandardSQL）：

Error: Cannot access field index on a value with type
ARRAY<STRUCT<index INT64, value STRING>> at [5:24]

对于旧版 SQL：

错误：无法查询重复字段的叉积 customDimensions.index 和 hits.contentGroup.contentGroup2。

编辑：

SELECT cd.value AS CustomerID,
SUM(totals.pageviews) as page_views,
SUM(CASE WHEN hits.type = 'PAGE' AND hits.contentGroup.contentGroup2 = 'important' THEN 1 ELSE 0 END) AS important_pageviews
FROM `xxxxxxxx.ga_sessions_20180415`,
UNNEST(customDimensions) AS cd
WHERE cd.index = 2
GROUP BY CustomerID 

返回：

Error: Cannot access field type on a value with type ARRAY<STRUCT<hitNumber INT64, time INT64, hour INT64, ...>> at [3:20]

我尝试使用 UNNEST(hits.type) = 'PAGE' AND UNNEST(hitscontentGroup.contentGroup2) = 'important' 更正 3:20 行，这给出了 Error: Syntax error: Unexpected keyword UNNEST at [3:15]

Answer 1

由于 customDimensions 是一个数组，您需要 unnest this 才能引用它的内容，请参阅下面的 StandardSQL 示例，我在 BigQuery 中从 Google Analytics 数据中取消嵌套 UserID：

SELECT customDimension.value AS UserID
FROM `my.project.data` AS t
  CROSS JOIN UNNEST(t.customdimensions) AS customDimension
  WHERE customDimension.index = 2