Question

当字符串'DebugData'包含数字时，我必须进行子选择。但在google big query中，我收到错误“在select子句中不允许Subselect”。任何帮助将不胜感激!!

 %%sql --module Test2
 DEFINE QUERY Test2
 SELECT 
 HardwareId, DebugReason, DebugData, 
 CASE 
  WHEN lower(DebugData) LIKE 'ver%' THEN 'Verizon'
  WHEN lower(DebugData) LIKE 'ro%' THEN 'Rogers'
  WHEN REGEXP_MATCH(DebugData,'\\d+') THEN c.Network
   (SELECT Network from [red-road-574:RawDebug.CarrierDetails] Where Mcc = SUBSTR(DebugData,0,3) and Mnc = SUBSTR(DebugData,4,7))
  ELSE REGEXP_REPLACE(DebugData,'\\?',' ')
 END
  as ActualDebugData
  FROM (TABLE_DATE_RANGE([bigdata:RawDebug.T], TIMESTAMP ('2016-05-15'),TIMESTAMP('2016-05-15')))
  CROSS JOIN [bigdata:RawDebug.CarrierDetails] c
  WHERE DebugReason = 50013
  LIMIT 200

我尝试了这个，但这不起作用：

 SELECT 
   DebugData, 
   CASE 
     WHEN lower(DebugData) LIKE 'jas%' THEN 'Jasper' 
     WHEN REGEXP_MATCH(DebugData,'\\d+') THEN c.Network
     ELSE REGEXP_REPLACE(DebugData,'\\?',' ')
   END
  as ActualDebugData 
  FROM (TABLE_DATE_RANGE([bigdata:RawDebug.T],TIMESTAMP ('2016-05-15'),TIMESTAMP('2016-05-15')) d)
  CROSS JOIN [bigdata:RawDebug.CarrierDetails] c
  WHERE d.DebugReason = 50013 and c.Mcc =    SUBSTR(DebugData,0,3)
 LIMIT 200

Answer 1

改为使用LEFT JOIN。

SELECT 
   DebugData, 
   CASE 
     WHEN lower(DebugData) LIKE 'jas%' THEN 'Jasper' 
     WHEN REGEXP_MATCH(DebugData,'\\d+') THEN c.Network
     ELSE REGEXP_REPLACE(DebugData,'\\?',' ')
   END
  as ActualDebugData 
  FROM (TABLE_DATE_RANGE([bigdata:RawDebug.T],TIMESTAMP ('2016-05-15'),TIMESTAMP('2016-05-15')) d)
  LEFT JOIN [bigdata:RawDebug.CarrierDetails] c ON c.Mcc =    SUBSTR(DebugData,0,3)
  WHERE d.DebugReason = 50013 
 LIMIT 200

使用LEFT JOIN可以让它从第一个表中返回与匹配条件不匹配的行。

在谷歌大查询中选择内部案例陈述

1 个答案: