Do you wonder if your IT business is truly profitable or if you should raise your prices? Learn how to calculate your overhead burden using our free interactive tool and use it to determine the right price for your IT services. Start calculating Now!
SQL Syntax Counter Needed
I have to update a counter field in one table while pulling the next counter number from another table and update that table. For example:
Table 1) SYSCNTRL has a field called INDXCNTR
Table 2) ITEMMSTR has item numbers and needs the next available INDXCNTR from SYSCNTRL. This field is also called INDXCNTR
So as I loop through records in ITEMSTR I need to get the value of SYSCNTRL.INDXCNTR and assign it to this record. Immediately afterward I need to increment the value of SYSCNTRL.INDXCNTR by one. Then I continue the process until all records are updated.
I have done this before with cursors but I am trying to move away from using cursors.
This is for SQL Server 2005.
Table 1) SYSCNTRL has a field called INDXCNTR
Table 2) ITEMMSTR has item numbers and needs the next available INDXCNTR from SYSCNTRL. This field is also called INDXCNTR
So as I loop through records in ITEMSTR I need to get the value of SYSCNTRL.INDXCNTR and assign it to this record. Immediately afterward I need to increment the value of SYSCNTRL.INDXCNTR by one. Then I continue the process until all records are updated.
I have done this before with cursors but I am trying to move away from using cursors.
This is for SQL Server 2005.
Do more with
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
I take it an identity column won't work for your purposes? That can auto-increment. Or a GUID Identity column is a pain in the nether regions.
I would suggest a custom function for this. But without details it is hard to give a straight out answer. Do you want to break the ID out by state or region? Dates? Both?
For example the 15th order from Pennsylvania today (57 ay of the year) might look like PA14057015. That gives you a detail that order 00015 doesn't. So identifying that early makes sense as opposed trying do it later.
I would suggest a custom function for this. But without details it is hard to give a straight out answer. Do you want to break the ID out by state or region? Dates? Both?
For example the 15th order from Pennsylvania today (57 ay of the year) might look like PA14057015. That gives you a detail that order 00015 doesn't. So identifying that early makes sense as opposed trying do it later.
Table1 - master
Table2 - detail
Table1 ID int NOT NULL IDENTITY(1,1), ...
Table2 ..., Table1ID int, ... -- FK to Table1 (ID)
DECLARE @LatestID int
INSERT INTO Table1 (...) -- don't include ID
SET @LatestID = @@IDENTITY
INSERT INTO Table2 (...,Table1ID,...)
VALUES (..., @LatestID, ...)
Table2 - detail
Table1 ID int NOT NULL IDENTITY(1,1), ...
Table2 ..., Table1ID int, ... -- FK to Table1 (ID)
DECLARE @LatestID int
INSERT INTO Table1 (...) -- don't include ID
SET @LatestID = @@IDENTITY
INSERT INTO Table2 (...,Table1ID,...)
VALUES (..., @LatestID, ...)
BEGIN TRY
BEGIN TRANSACTION
DECLARE @rows_updated int
UPDATE i
SET INDXCNTR = s.INDXCNTR + i_keys.row_num
FROM (
SELECT key_col, /*key_col2, ... */
ROW_NUMBER() OVER(ORDER BY key_col) AS row_num
FROM dbo.ITEMMSTR
--WHERE <conditions_indicating_IND
) AS i_keys
INNER JOIN dbo.ITEMMSTR i ON
i.key_col = i_key.key_col
-- WITH (TABLOCKX) is required -- do NOT remove it!
CROSS JOIN dbo.SYSNCTRL s WITH (TABLOCKX)
SET @rows_updated = @@ROWCOUNT
UPDATE dbo.SYSNCTRL
SET INDXCNTR = INDXCNTR + @rows_updated
COMMIT TRANSACTION --this will release the TABLOCKX
END TRY
BEGIN CATCH
IF XACT_STATE() <> 0
ROLLBACK TRANSACTION
--...other error logic here...
END CATCH
END CATCH
BEGIN TRANSACTION
DECLARE @rows_updated int
UPDATE i
SET INDXCNTR = s.INDXCNTR + i_keys.row_num
FROM (
SELECT key_col, /*key_col2, ... */
ROW_NUMBER() OVER(ORDER BY key_col) AS row_num
FROM dbo.ITEMMSTR
--WHERE <conditions_indicating_IND
) AS i_keys
INNER JOIN dbo.ITEMMSTR i ON
i.key_col = i_key.key_col
-- WITH (TABLOCKX) is required -- do NOT remove it!
CROSS JOIN dbo.SYSNCTRL s WITH (TABLOCKX)
SET @rows_updated = @@ROWCOUNT
UPDATE dbo.SYSNCTRL
SET INDXCNTR = INDXCNTR + @rows_updated
COMMIT TRANSACTION --this will release the TABLOCKX
END TRY
BEGIN CATCH
IF XACT_STATE() <> 0
ROLLBACK TRANSACTION
--...other error logic here...
END CATCH
END CATCH
I just got back into town and will take a look a these suggestions. The detail of the situation is this. I have a table with one record in it. This record will contain a company ID, a bunch of other columns, and a column that is an index to a note field. This field is defined as an integer. Every time a new item is added to inventory my application goes out and gets what the current value for the next index, assigns it to the item record just added and then increments by one to get ready for the next item insert. So I insert items into one table while I increment this index value in another table.
The problem I face right now is that I have been given an import file and I have to dump these items into the database and bypass the application. I have already imported these items but now I need to populate the index value on each item with a unique incremented index value.
P.S. Every table in my database already has an identity column so that will not work.
The problem I face right now is that I have been given an import file and I have to dump these items into the database and bypass the application. I have already imported these items but now I need to populate the index value on each item with a unique incremented index value.
P.S. Every table in my database already has an identity column so that will not work.
Premium Content
You need an Expert Office subscription to comment.Start Free Trial