Add records to a SQL table based on values in another table

A query like this one (implemented here as a stored procedure for clarity) should do what you are looking for, as I understand it.  Here are the assumptions I will be making:

     "I want to add a record to the DrgGrp table for every time a a record exists in the Drg table with the text "CAP" anywhere in the FORM field "

We will be looking for the distinct word CAP, not the letters CAP embedded in another word.  Thus, the WHERE clause will be a bit complex, eg

WHERE form LIKE 'CAP %'
    OR form LIKE '% CAP'
    OR form LIKE '% CAP %'

If other separators than a space may occur, a variation would be needed.  For example, if the word could end with a period, space, or comma, then the WHERE would look like:

WHERE form LIKE 'CAP[ ,.]%'
    OR form LIKE '% CAP'
    OR form LIKE '% CAP[ ,.]%'


     "and the records being added will all have the same value for DrgGrp.GroupID (lets say for the example of the query to make it 199)."

Passed as a parameter into the stored procedure


     "If several records exist with the same DrgGrp.DrgID the DrgGrp.Seq must be incremented to the next higher number"

Assuming that the condition should be "If several records exist with the same DrgGrp.DrgID and DrgGrp.GroupID....."


Given these assumptions:

CREATE PROCEDURE drg_AddDrgGrpRecords
    @GroupID int
AS
BEGIN

CREATE PROCEDURE drg_AddDrgGrpRecords
    @GroupID int
AS
BEGIN
    CREATE TABLE #tmp (
        id INT IDENTITY(1,1) PRIMARY KEY CLUSTERED,
        DrgID INT NOT NULL
    )
 
    INSERT INTO #tmp (DrgID)
    SELECT ID 
    FROM Drg
    WHERE form LIKE 'CAP[ ,.]%'
        OR form LIKE '% CAP'
        OR form LIKE '% CAP[ ,.]%'
 
    INSERT INTO DrgGrp (
        DrgID,
        GroupID,
        Created,        ---- Not updated in sample, can be commented out if not needed
        Seq,
        CentralMaintFieldMask
        )
    SELECT 
        DrgID,
        @GroupID,
        CURRENT_TIMESTAMP,    -- Not updated in sample, can be commented out if not needed
        (SELECT COUNT(*) FROM #tmp x WHERE x.DrgID = t.DrgID AND x.id <= t.id) Seq,
        0
    FROM #tmp t
END

Select allOpen in new window

Where do I specify DrgGrp.GroupID in the stored procedure that will be used for all the records that get added?

Sorry, I just ran the stored procedure and see that it prompts for the value.  I will try it now and let you know the results shortly.

Looks like it works well but the portion that is supposed to account for existing Seq #'s is not working.  My SQL is very basic but looking at the query it look like it might be because it is looking for existing Seq #'s in the TMP table instead of the DrgGrp table?

Right now when it is adding the records it simply uses Seq "1" but I think that's because there are never more than 1 seq #'s in the TMP table.

Worked Great.  Thanks bhess1

When I comment out where you have suggested I get the following error:

Msg 121, Level 15, State 1, Procedure drg_AddDrgGrpRecords, Line 26
The select list for the INSERT statement contains more items than the insert list. The number of SELECT values must match the number of INSERT columns.

Can you please help with this?

Ok, fixed the last error but now getting this when I run the stored procedure:

(565 row(s) affected)
Msg 2601, Level 14, State 1, Procedure drg_AddDrgGrpRecords, Line 15
Cannot insert duplicate key row in object 'dbo.DrgGrp' with unique index 'IXC_DrgGrp_DrgID_GroupID'.
The statement has been terminated.

(1 row(s) affected)