For example, I have a table partitioned by column ym
(202001).
Now, there is a SQL converting ym
to other time format:
select * from table
where from_unixtime(unix_timestamp(`table`.ym , 'yyyyMM')) >= '2020-01-01 00:00:00')
AND from_unixtime(unix_timestamp(`table`.ym , 'yyyyMM')) < '2020-01-05 00:00:00');
does it still have partition effect while the partition column is not as the original value?
Whether Hive scan only the 202001
partition range or the whole table range?
is there any official document for this? Thanks.