贪婪匹配和非贪婪匹配的区别 .* 和 .*?
--取|分割最后一段,.*贪婪匹配到最后一个|
SELECT regexp_extract(creativename,'.*\|(.*)',1) as "Creative Type",
creativename FROM "ads"."ads_dsp_detail_df"
--取|分割除去第一段之外的所有字段
SELECT regexp_extract(creativename,'.*?\|(.*)',1) as "Creative Type",
creativename FROM "ads"."ads_dsp_detail_df"
-- hive中是从0下标开始,Athena中从1开始,如果不确定分割后的字段个数,不可以用split,不然会报错数组越界:INVALID_FUNCTION_ARGUMENT: Array subscript out of bounds
SELECT split(creativename,'|')[2] as "Creative Type", creativename FROM "ads"."ads_dsp_detail_df"
--当要截取的字段可能是最后一个也可能不是的时候,在Athena中先用cardinality判断数组的长度再取值
select case when cardinality(split(creativename,'|'))<3 then null else split(creativename,'|')[3] end as "Creative Type",
creativename FROM "ads"."ads_dsp_detail_df"