Reputation: 417
I have a table that is like this:
col1 col2 num
a <string1> 5
a <string2> 10
a <string3> 0
a <string4> 7
b <string1> 6
b <string2> 3
b <string3> 20
b <string4> 1
I want to select the min num for each value of col1 (or eventually col2), so the desired output would be:
col1 col2 num
a <string3> 0
b <string4> 1
How can I achieve this? I'm trying to do this in BigQuery.
Upvotes: 0
Views: 43
Reputation: 172994
Below is for BigQuery Standard SQL
#standardSQL
select as value array_agg(t order by num limit 1)[offset(0)]
from `project.dataset.table` t
group by col1
if to apply to sample data from your question as in below example
#standardSQL
WITH `project.dataset.table` AS (
select 'a' col1, '<string1>' col2, 5 num union all
select 'a', '<string2>', 10 union all
select 'a', '<string3>', 0 union all
select 'a', '<string4>', 7 union all
select 'b', '<string1>', 6 union all
select 'b', '<string2>', 3 union all
select 'b', '<string3>', 20 union all
select 'b', '<string4>', 1
)
select as value array_agg(t order by num limit 1)[offset(0)]
from `project.dataset.table` t
group by col1
the output is
Upvotes: 1
Reputation: 222432
If you are running Postgres (as tagged), you can use distinct on
:
select distinct on (col1) t.*
from mytable t
order by col1, num
In BigQuery (as mentionned in the question), you can do this with arrays:
select array_agg(t order by num limit 1)[offset(0)].*
from mytable t
group by col1
Upvotes: 0