Reputation: 143
I am having a table in hive with below values
ID value
1
1
ID value
1
1 2
while doing sum i need the output as
select id,sum(val) from table group by id;
first required output
id sum
1
Second table Output
id sum
1 2
Upvotes: 1
Views: 7849
Reputation: 1483
In math default 2+0=2
so its anyway going to work .Don't worry about this hive
will be default will take care this.
hive> create table first (Id int,value int);
OK
Time taken: 3.895 seconds
hive> select * from first;
OK
1 2
1 NULL
hive> select id, sum(value) as sum from first group by id;
Total MapReduce CPU Time Spent: 4 seconds 610 msec
OK
1 2
Time taken: 83.483 seconds, Fetched: 1 row(s)
Upvotes: 1
Reputation: 38335
If you need to filter out rows with null sum, use having
:
select id, sum(value) from table group by id having sum(value) is not null;
Upvotes: 0