问题
I am having trouble succinctly describing what I need, so any help editing the title is appreciated!
I have 3 Tables:
Jobs(hasJobIDPK)JobsDetail(hasJobIDFK andDepartmentsIDFK)Departments(hasDepartmentsIDPK)
Departments will have a low number of records (~10)
If I have 10 Department rows, I need all 10 rows repeated for each Jobs record. DepartmentIDs that aren't in the JobsDetail records for the Job will show a NULL value in the Job column. With one Jobs record, a typical LEFT JOIN works the way I want it to:
Select d.Department, jobs.JobIdentifier
From Departments d
LEFT JOIN (Select Distinct j.JobID, j.JobIdentifier, DepartmentID,
From Jobs j
Join JobsDetail jd on j.JobID = jd.JobID) jobs on d.DepartmentID = jobs.DepartmentID
yields these results:
Department JobIdentifier
310 NULL
320 NULL
430 NULL
450 NULL
460 NULL
500 NULL
530 1000
533 1000
534 1000
535 NULL
Of course, when another Jobs record is added, I see this:
Department JobIdentifier
310 2000
320 NULL
430 NULL
450 NULL
460 2000
500 NULL
530 1000
533 1000
534 1000
534 2000
535 NULL
What I NEED is something like this:
Department JobIdentifier
310 NULL
320 NULL
430 NULL
450 NULL
460 NULL
500 NULL
530 1000
533 1000
534 1000
535 NULL
310 2000
320 NULL
430 NULL
450 NULL
460 2000
500 NULL
530 NULL
533 NULL
534 2000
535 NULL
How can this be accomplished?
Here is some test data from the tables:
JobID SubPlantID JobIdentifier PartFamilyID OrderDate OrderedBy OrderQuantity DueDate SpecialInstructions PrintDate PrintedBy StartDate StartedBy ProducedQuantity ReprintNumber CompletedDate Location
-------------------- ----------- -------------------------------------------------- ------------ ----------------------- -------------------------------------------------- ------------- ----------------------- ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ----------------------- -------------------------------------------------- ----------------------- -------------------------------------------------- ---------------- ------------- ----------------------- --------------------------------------------------
2 1 1000 1 2012-12-21 13:20:00.000 Keith 1 2012-12-28 00:00:00.000 NULL NULL 2012-12-28 00:00:00.000 NULL NULL 0 NULL NULL
3 1 2000 1 2013-01-03 00:00:00.000 Jon 10 2013-01-10 00:00:00.000 NULL NULL NULL NULL NULL NULL 0 NULL NULL
JobsDetailID JobID Operation FirstStartDate OperationQuantity OperationStatusTypeID OperationDescription DepartmentID WorkCenterID UserName
-------------------- -------------------- --------- ----------------------- ----------------- --------------------- -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ------------ ------------ --------------------------------------------------
1 2 10 NULL 1 2 Weld 1 2 keith
2 2 20 NULL 1 3 Punch 1 1 keith
3 2 30 NULL 1 2 Form 3 5 jon
4 2 40 NULL 1 3 Paint 2 4 jon
6 2 50 NULL 1 3 Glue 2 4 jwood
9 2 60 NULL 1 2 Eat 3 5 jon
10 3 20 NULL 10 1 Punch 10 18 Jon Wrye
11 3 30 NULL 10 0 Form 15 29 Jon Wrye
12 3 40 NULL 10 0 Weld 13 24 Jon Wrye
13 3 10 NULL 10 2 Start 1 1 jwrye
14 3 50 NULL 10 0 Finish 1 2 jwrye
DepartmentID Department
------------ --------------------------------------------------
1 534
2 533
3 530
4 535
5 500
6 450
7 430
8 200
9 240
10 232
11 220
12 300
13 460
14 320
15 310
回答1:
You should start with a cross join between jobs and departments, and then build the query from there:
Select d.Department, jobs.JobIdentifier
From Departments d cross join
(select distinct JobIdentifier from Jobs j) ji LEFT JOIN
(Select Distinct j.JobID, j.JobIdentifier, DepartmentID,
From Jobs j Join
JobsDetail jd
on j.JobID = jd.JobID
) jobs
on d.DepartmentID = jobs.DepartmentID and
ji.jobidentifer = jobs.jobIDentifier
I'm a bit unclear on the distinction between JobId and JobIdentifier, so this may not be quite right.
If you add ji.JobIdentifier to the select list, you'll see which job the department should belong to, even when there is no match.
回答2:
You will need to cross out the catalogs (with a cross join), Department and Jobs, and then make a left join to your relationship table
http://sqlfiddle.com/#!6/277ec/30
回答3:
how does this do?
select d.Department, CASE WHEN d.DepartmentId = jj.DepartmentID THEN jj.JobID ELSE NULL END
from Department d, (SELECT j.JobID, jd.DepartmentID FROM Jobs j left outer join JobDetail jd on j.JobID = jd.JobID) as jj
回答4:
I'm not sure if I understand exactly what you're trying to do so lets start at the beginning. In what way does the following statement prove inadequate?
Select
d.Department,
j.JobIdentifier
From
Departments d
Left Join JobDetails jd On d.Id = jd.DepartmentId
Left Join Jobs j On jd.JobId = j.Id
来源：https://stackoverflow.com/questions/14147248/repeat-all-base-table-rows-for-each-left-join-row-match