A better way is to define PERIOD in your Hive view, and load the Hive view as the fact table.
2017-05-25 14:47 GMT+08:00 jiawei chen <[email protected]>: > Hi kylin's developers, > I am using kylin-1.6.0 now. I have to use a sql query like: > > SELECT > SUBSTR(TO_CHAR(`time`,'YYYYMMDDHH24MISS'),9,2) AS PERIOD, > COUNT(*) AS LL, > FROM table_name > GROUP BY > SUBSTR(TO_CHAR(`time`,'YYYYMMDDHH24MISS'),9,2); > > and I got a failed response that shows no match found for function > signature TO_CHAR, (If I remove to_char, SUBSTR still not found). > > Is there any plane to support these transform functions? > Or what alternative way can I choose? > > Thank you! >
