问题
In PostgreSQL, is it possible to generate a series of repeating numbers? For example, I want to generate the numbers 1 to 10, with each number repeated 3 times:
1
1
1
2
2
2
3
3
3
.. and so on.
回答1:
You could cross join it to a series of 3:
SELECT a.n
from generate_series(1, 100) as a(n), generate_series(1, 3)
回答2:
You could try integer division like this:
SELECT generate_series(3, 100) / 3
回答3:
For such small numbers CROSS JOIN
two VALUES expressions:
SELECT n
FROM (VALUES (1),(2),(3)) x(r) -- repetitions (values are not used)
,(VALUES (1),(2),(3),(4),(5),(6),(7),(8),(9),(10)) y(n); -- numbers
SQL Fiddle.
This works for any sequence of numbers (including repeated or irregular patterns).
For anything bigger and with regular sequential pattern use generate_series() as has been suggested.
回答4:
SELECT * FROM (SELECT generate_series(1, 10)) A
JOIN (
SELECT generate_series(1, 3)
) B ON (TRUE)
http://www.sqlfiddle.com/#!12/d41d8
回答5:
I don't know if you can do use generate_series like that in PostgreSQL, but i would try a cross join:
SELECT x FROM
(SELECT generate_series(1, 10) AS x) t1,
(SELECT generate_series(1, 3) as y) t2
Edit:
As generate_series already returns a table there's no need for SELECT in a Derived Table:
SELECT x FROM
generate_series(1, 10) AS x,
generate_series(1, 3) as y
回答6:
Just another options:
select generate_series(1, 3) from generate_series(1, 10)
select generate_series(1, 30) % 10 + 1
回答7:
SELECT a.x from generate_series(0, 100) as a(x), generate_series(1, 3)
来源:https://stackoverflow.com/questions/23358333/how-can-i-generate-a-series-of-repeating-numbers-in-postgresql