I have two tables A and B.
A has an id for rows in table B, and a counter:
b_id integer not null references B(id),
count integer not null default 1
Is there a way in Postgres to return query of A with the rows in B where the rows are repeated count times?
Yes. Use generate_series():
select t.*, n
from t cross join lateral
generate_series(1, t.count, 1) gs(n);
The above is actually the verbose way of writing the logic. I prefer the above, because it is quite explicit about what is happening. However, you can simplify this to:
select t.*, generate_series(1, t.count, 1) as n
from t;
select B.* from A join B on A.b_id=B.id, generate_series(1, A.count,1) n
from clause. Use an explicit cross join. But that should work.