SELECT j.job_id

FROM Jobs As j

INNER JOIN Activities As a

  ON (j.job_id = a.job_id)

WHERE a.activity_type_id IN (1, 3)

GROUP BY j.job_id

HAVING Count(*) = 1

Use fn_split()

http://www.sqlservercentral.com/forums/shwmessage.aspx?forumid=8&messageid=302389#bm302391

We don't know enough about the tables to determine if a JOIN is correct.

>>select mem_id from dbo.conc_table00

If mem_id is not unique, using a JOIN instead of EXISTS will generate dupes...

DISTINCT works across all columns in the resultset, not just the column immediately after the keyword DISTINCT.

SELECT DISTINCT id, name gives you all the unique id AND name combinations. If...

Did you mean derived table ?

As in (using system tables for an example):

Select TableName, NumberOfColumns

From

-- Parenthesis starts derived/virtual table definition

(

  Select Object_Name(id) As TableName, Count(*) As NumberOfColumns

  From...

You don't need CASE .. WHEN, it's just a simple OR:

WHERE

  (@Status_code IS NOT NULL AND status_code = 10)  OR

  (@status_code IS NULL AND status_code IN (9,11,12) )

-- Table for sample data

Create Table #Test

(

  Name varchar(20),

  Event  varchar(20),

  Time  datetime

)

-- Populate with sample data

Insert Into #Test

Select 'Name1', 'Start',  '8/16/06 8:00 AM' Union All

Select 'Name2', 'Start', '8/16/06...

If you want to start by looking at the SQL, then some warning signs to look for:

- Use of sub-SELECTS within the main SELECT. Creates cursor-like sluggishness

- Use of user-defined...

>>I think i'll start writing the SP

That's 1 approach.

A better approach in the long run would be to design the database correctly using standard normalization techniques.

If a "story" can have...

Use fn_split() function:

http://www.sqlservercentral.com/scripts/viewscript.asp?scriptid=157

This constructs a table, which you join to instead of using IN ().

Why on earth would you index a float data type ?

A float is an approximate number, indexing it makes no sense.

The error is pretty explicit is it not ?

It's telling you it expects @cmd to be of type 'ntext/nchar/nvarchar', but you've declared it as varchar.

>>an identity value is a bad canidate for a clustered index because of all of the split pages at the end of the file.

On the contrary, clustering on an identity...

>>this is not working

Need to be a bit more specific.

What is the actual error you're getting ? Are the tables involved in referential integrity ? You can't drop tables if...

sp_help and sp_helpconstraint