Please start any new threads on our new site at http://forums.sqlteam.com. We've got lots of great SQL Server experts to answer whatever question you can come up with.

Our new SQL Server Forums are live! Come on over! We've restricted the ability to create new threads on these forums.

SQL Server Forums
Profile | Active Topics | Members | Search | Forum FAQ
 
Register Now and get your question answered!
Username:
Password:
Save Password
Forgot your Password?

 All Forums
 SQL Server 2008 Forums
 Transact-SQL (2008)
 Complex Grouping query ?
 Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic  

jcdenmark
Starting Member

2 Posts

Posted - 11/29/2012 :  15:07:53  Show Profile  Reply with Quote
To anyone who might help...

I have a conventional junction table, e.g. for Contacts associated with a folder, a FoldersContacts table, including basically the FolderID, the ContactID and a ContactCategory.

- Querying for folders associated with 1 specific Contact (Category irrelevant) is trivial;
- Querying for folders associated with BOTH of 2 ContactIDs, regardless of category, can be done by the below, I think (but maybe prone to error below?)...
e.g. SELECT FolderID from FoldersContacts WHERE ContactID IN('A','B')
GROUP BY FolderID
HAVING COUNT(FolderID)>=2;

Querying for folders with ALL OF 3 different contacts;
The following does not work:

SELECT FolderID from FoldersContacts WHERE ContactID IN('A','B', 'C')
GROUP BY FolderID HAVING COUNT(FolderID)>=3

....because there are many cases where the COUNT(FolderID)>=3, but only 1 or 2 of 'A','B','C' are listed in the FoldersContacts table, albeit with different ContactCategories.

Question: how do I efficiently query for all folders having all of 3 (or more) Contacts associated with them ?

I have tried using a sub-query to retrieve all the contacts for a particular FolderID, and then establishing whether the (DISTINCT) Contacts Listed completely match a temp Contact table with 'A', 'B', 'C'; I did get this working:

SELECT DISTINCT FolderID
FROM FoldersContacts AS FTOP
WHERE
NOT EXISTS
(SELECT DISTINCT ContactID FROM 3Contacts LEFT JOIN
(SELECT ContactID FROM FoldersContacts WHERE
FolderID=TTMAIN.FolderID) AS SQ
ON 3Contacts.ContactID=SQ.ContactID
WHERE SQ.ContactID IS NULL
)

This works in test, but on a larger production table (e.g. FoldersContacts with 23000 records), it is UNBELIEVABLY SLOW.

Any ideas on speeding this up ? I'm sure I'm missing something basic.

And then once I have all the folders having Contacts 'A','B','C', can I then use a similar routine to further limit by Contact Category ?

Any help massively appreciated...I been really struggling with this...

Jim

jimf
Flowing Fount of Yak Knowledge

USA
2875 Posts

Posted - 11/29/2012 :  15:41:11  Show Profile  Reply with Quote
SELECT FolderID
from FoldersContacts
WHERE ContactID IN('A','B', 'C')
GROUP BY FolderID
HAVING COUNT(DISTINCT ContactID) = 3

Jim

Everyday I learn something that somebody else already knew
Go to Top of Page

jcdenmark
Starting Member

2 Posts

Posted - 11/29/2012 :  16:10:01  Show Profile  Reply with Quote
Brilliant....thanks.
Go to Top of Page

jimf
Flowing Fount of Yak Knowledge

USA
2875 Posts

Posted - 11/29/2012 :  17:32:45  Show Profile  Reply with Quote
You're welcome.

Jim

Everyday I learn something that somebody else already knew
Go to Top of Page
  Previous Topic Topic Next Topic  
 Reply to Topic
 Printer Friendly
Jump To:
SQL Server Forums © 2000-2009 SQLTeam Publishing, LLC Go To Top Of Page
This page was generated in 0.12 seconds. Powered By: Snitz Forums 2000