I need to generate a list of users' departments codes. If a user doesn't have code get its manager code, and so on up the hierarchy.
The initial table looks like this:
manager emp code
-----------------------
boss subboss AAA
boss subsub SUBCODE
subboss john ABC
subboss alan (null)
(null) boss ZZZ
subsub steve (null)
steve rick (null)
CREATE TABLE Users
(
[manager] varchar(10),
[emp] varchar(10),
[code] varchar(10)
);
INSERT INTO Users ([manager], [emp], [code])
VALUES
('boss', 'subboss', 'AQQ'),
('boss', 'subsub', 'SUBSUB'),
('subboss', 'john', 'ABC'),
('subboss', 'alan', null),
(null, 'boss', 'ZZZ'),
('subsub', 'steve', null),
('steve', 'rick', null);
Desired result is this:
manager emp code
------------------------
boss subboss AAA
boss subsub SUBCODE
subboss john ABC
subboss alan AAA
(null) boss ZZZ
subsub steve SUBCODE
steve rick SUBCODE
My first attempt is:
select
manager, emp,
coalesce(code, (select code from Users u1 where u.manager = u1.code))
from
Users u;
But it returns only the direct manager code.
I'd appreciate tips on how to do it recursively with CTE.
CodePudding user response:
Try something like this:
WITH Hierarchy AS
(
-- create the "anchor" - the toplevel node(s)
SELECT
u.emp, u.manager, u.code, 0 AS Level
FROM
Users u
WHERE
u.manager IS NULL
UNION ALL
-- recursive part - join subordinate to manager, one level up
SELECT
u.emp, u.manager, u.code, h.Level 1
FROM
Hierarchy h
INNER JOIN
Users u ON u.manager = h.emp
)
SELECT
*
FROM
Hierarchy
This results in a dataset like this:
emp manager code Level
-----------------------------
boss NULL ZZZ 0
subboss boss AQQ 1
subsub boss SUBSUB 1
steve subsub NULL 2
rick steve NULL 3
john subboss ABC 2
alan subboss NULL 2