Solved

Recursive selection of table content

Posted on 2010-11-28
9
277 Views
Last Modified: 2012-05-10
I'm trying to move my understanding of recursion to tsql.

id     name                subid
1      Steve              3
2      Margot              3
3      Family              6
4      Other Family      6
5      Evert              4
6      Department      7
7      City                 8
8      State              9
9      Country      

I understand that sql allow 32 levels of recursives deep.
I have to approaches with this test:

1 Make a recursive stored procedure that gives a result with all related subid ordered

2. Make it like a path (ponder if there was a web path that needs to be concated)


And also that @@nestLevel is a sort of built in function for avoiding deeper levels.
0
Comment
Question by:dingir
  • 4
  • 4
9 Comments
 
LVL 13

Expert Comment

by:devlab2012
ID: 34225534
Can you please explain your question? Maybe Common Table Expression (CTE) is what you are looking for.
0
 
LVL 1

Author Comment

by:dingir
ID: 34225569
What is cte?

I would like to get i.e.
Sp(2) and got the concated path all recursived.
Steve, family, department, city and so on..

0
 
LVL 58

Accepted Solution

by:
cyberkiwi earned 500 total points
ID: 34225987
> Sp(2) and got the concated path all recursived.

Just one record? The entire path?
Have a look at this example
-- create a test table
create table testcte(id int,name varchar(20),subid int)
insert testcte select
1,'Steve',3 union all select
2,'Margot',3 union all select
3,'Family',6 union all select
4,'Other Family',6 union all select
5,'Evert',4 union all select
6,'Department',7 union all select
7,'City',8 union all select
8,'State',9 union all select
9,'Country', null
GO

-- create the test procedure
create proc usp_testcte @start int as
;with cte as (
-- the part before "union all" is called the anchor, it starts the recursion
	select subid, path=convert(varchar(max),name) from testcte where id=@start
	union all

-- the part after "union all" is called the recursive part. In here, "cte" references
-- the result from the most recent iteration of the recursion (starts with the anchor)
-- as long as this query generates results, it will continue to run, using the
-- previous iteration as the "table" content of "cte" ("cte" being the name given to the
-- common table expresion as the first word after "with")
	select t.subid, cte.path+'\'+t.name from cte
	inner join testcte t on t.id=cte.subid)

-- outside the CTE block, you can access the built-up results by referring to "cte" as a table
-- MAX is used here to just the last row, which in this case will be the longest
select max(path) from cte
GO

-- test it
exec usp_testcte 1

Open in new window

0
PRTG Network Monitor: Intuitive Network Monitoring

Network Monitoring is essential to ensure that computer systems and network devices are running. Use PRTG to monitor LANs, servers, websites, applications and devices, bandwidth, virtual environments, remote systems, IoT, and many more. PRTG is easy to set up & use.

 
LVL 1

Author Comment

by:dingir
ID: 34226046
Hm.. i feel this isnt the fully right way..?

Yes one record as result (2 given as parameter, the id of the table).
0
 
LVL 58

Expert Comment

by:cyberkiwi
ID: 34226328
> Hm.. i feel this isnt the fully right way..?

And why would you say that?
0
 
LVL 1

Author Comment

by:dingir
ID: 34226559
Why would i need so much code to that task?
0
 
LVL 58

Expert Comment

by:cyberkiwi
ID: 34226571
Have you read the code at all?
The first part sets up a table so that I can show you how it works in the 2nd part (the procedure you need).
The 3rd part is a sample of how you would use such a procedure.
If all you are after is a fast cut and paste answer, this would be all there is to create such a proc

create proc usp_name_me @start int as
;with cte as (
      select subid, path=convert(varchar(max),name) from testcte where id=@start
      union all
      select t.subid, cte.path+'\'+t.name from cte
      inner join testcte t on t.id=cte.subid)
select max(path) from cte
GO


Is that better?  Replace all bold parts with your actual table/field names
0
 
LVL 58

Expert Comment

by:cyberkiwi
ID: 34226574
All the lines starting with 2 dashes "--" are comments to help you understand the code, but it seems you may not be interested.
0
 
LVL 1

Author Comment

by:dingir
ID: 34341377
I care, absolutely.. i wasn't expect such a solution though. I will store it and see if it came of use later on. Thank's a lot and sorry for the big delay.
0

Featured Post

Microsoft Certification Exam 74-409

Veeam® is happy to provide the Microsoft community with a study guide prepared by MVP and MCT, Orin Thomas. This guide will take you through each of the exam objectives, helping you to prepare for and pass the examination.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Introduction SQL Server Integration Services can read XML files, that’s known by every BI developer.  (If you didn’t, don’t worry, I’m aiming this article at newcomers as well.) But how far can you go?  When does the XML Source component become …
For both online and offline retail, the cross-channel business is the most recent pattern in the B2C trade space.
Via a live example, show how to extract insert data into a SQL Server database table using the Import/Export option and Bulk Insert.
Via a live example, show how to set up a backup for SQL Server using a Maintenance Plan and how to schedule the job into SQL Server Agent.

860 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question