sql - Postgres query with limit that selects all records with similar identifier

Question

I have a table that looks something like this:

customer_id	data
1	123
1	456
2	789
2	101
2	121
2	123
3	123
4	456

Answer 1

In Postgres 13 can use with ties:

select t.*
from t
order by customer_id
fetch first 3 rows with ties;

In earlier versions you can use in:

select t.*
from t
where t.customer_id in (select t2.customer_id
                        from t t2
                        order by t2.customer_id
                        limit 3
                       );