I'm getting my head around the MERGE statement in SQL server. I generally use it to insert/update a single row, which I realise isn't the only use, but it's something I seem to do quite often.
But what happens if you want to insert a value of 1, or update to increment the value and output the incremented value eg:
CREATE TABLE [Counter] (
[Key] VARCHAR(255) NOT NULL PRIMARY KEY,
[Value] INT NOT NULL
);
DECLARE #paramKey VARCHAR(255);
SET #paramKey = 'String';
MERGE [Counter] AS targt
USING (Values(#paramKey)) AS source ([Key])
ON (targt.[Key] = source.[Key])
WHEN MATCHED THEN
UPDATE SET Value = Value +1
WHEN NOT MATCHED THEN
INSERT ([Key], Value)
VALUES (source.[Key], 1);
-- but now I want the new value!
Is there a way of doing this? I notice the output clause in https://msdn.microsoft.com/en-gb/library/bb510625.aspx but it doesn't seem to work with scalars (I could output to a single row-ed table variable but that seems wrong):
-- using table variables but seems
DECLARE #paramKey VARCHAR(255), #value int;
SET #paramKey = 'String'
DECLARE #Tab table (
[Value] INT
)
MERGE Counter AS targt
USING (Values(#paramKey)) AS source ([Key])
ON (targt.[Key] = source.[Key])
WHEN MATCHED THEN
UPDATE SET Value = Value +1
WHEN NOT MATCHED THEN
INSERT ([Key], Value)
VALUES (source.[Key], 1)
OUTPUT inserted.[Value] INTO #Tab;
-- can now use #Tab as a single rowed table variable
Is there a better option?
Related
I'm writing a stored procedure to insert data from a form into two tables. One table has an autonumbered identity field. I need to insert the data into that table, find the newly created autonumber, and use that number to insert data into another table. So, to boil it down, I have a one-to-many link between the two tables and I need to make sure the identity field gets inserted.
Is this code the best way to do something like this, or am I missing something obvious?
CREATE PROCEDURE [dbo].[sp_Insert_CRT]
(
#TRACKING_ID int,
#CUST_NUM int,
#TRACKING_ITEM_ID int,
#STATEMENT_NUM nvarchar (200) = null,
#AMOUNT numeric (15, 2),
#BBL_ADJUSTED int = NULL,
#PAID_VS_BILLED int = NULL,
#ADJUSTMENT_TYPE int = NULL,
#ENTERED_BY nvarchar (10) = NULL,
#ENTERED_DATE date = NULL,
#AA_STATUS int = NULL
)
AS
BEGIN
-- Insert data into CRT_Main, where Tracking_ID is an autonumber field
INSERT into tbl_CRT_Main
(
-- TRACKING_ID
CUST_NUM
,TRACKING_ITEM_ID
,STATEMENT_NUM
,AMOUNT
)
VALUES
(
-- #TRACKING_ID
#CUST_NUM
,#TRACKING_ITEM_ID
,#STATEMENT_NUM
,#AMOUNT
)
-- Find the newly generated autonumber, and use it in another table
BEGIN TRANSACTION
DECLARE #TrackID int;
SELECT #TrackID = coalesce((select max(TRACKING_ID) from tbl_CRT_Main), 1)
COMMIT
INSERT into tbl_CRT_Admin_Adjustment
(
TRACKING_ID
,BBL_ADJUSTED
,PAID_VS_BILLED
,[ADJUSTMENT_TYPE]
,[ENTERED_BY]
,[ENTERED_DATE]
,AA_STATUS
)
VALUES
(
#TrackID
,#BBL_ADJUSTED
,#PAID_VS_BILLED
,#ADJUSTMENT_TYPE
,#ENTERED_BY
,#ENTERED_DATE
,#AA_STATUS
)
END
SELECT #TrackID = coalesce((select max(TRACKING_ID) from tbl_CRT_Main), 1)
No, don't do this. This will get you the maximum value of TRACKING_ID yes, but that doesn't mean that's the value that was created for your INSERT. If multiple INSERT statements were being run by different connections then very likely you would get the wrong value.
Instead, use SCOPE_IDENTITY to get the value:
SET #TrackID = SCOPE_IDENTITY();
Also, there is no need to wrap the above in an explicit transaction like you have with your SELECT MAX(). Instead, most likely, the entire batch in the procedure should be inside it's own explicit transaction, with a TRY...CATCH so that you can ROLLBACK the whole batch in the event of an error.
I am trying to use T-SQL Merge to check for the existence of records and update, if not then insert.
The update works fine, but the insert is not working.
Any and all help on this would be gratefully received.
DECLARE
#OperatorID INT = 2,
#CurrentCalendarView VARCHAR(50) = 'month';
WITH CTE AS
(
SELECT *
FROM dbo.OperatorOption
WHERE OperatorID = #OperatorID
)
MERGE INTO OperatorOption AS T
USING CTE S ON T.OperatorID = S.OperatorID
WHEN MATCHED THEN
UPDATE
SET T.CurrentCalendarView = #CurrentCalendarView
WHEN NOT MATCHED BY TARGET THEN
INSERT (OperatorID, PrescriptionPrintingAccountID, CurrentCalendarView)
VALUES (#OperatorID, NULL, #CurrentCalendarView);
When would a row Selected from OperatorOption not already exist in OperatorOption?
If you're saying this code does not insert - you're right it doesn't because the row has to be there to begin with (in which case it won't insert), or the row is not there to begin with, in which case there is nothing in the source dataset to insert.
Does
SELECT *
FROM dbo.OperatorOption
WHERE OperatorID = #OperatorID
return anything or not?
This does not work the way you think it does. There is nothing in the source CTE.
The answer to 'was a blank dataset missing from the target' is 'No' so nothing is inserted
To do this operation, I use this construct:
INSERT INTO dbo.OperatorOption
(OperatorID, PrescriptionPrintingAccountID, CurrentCalendarView)
SELECT #OperatorID, NULL, #CurrentCalendarView
WHERE NOT EXISTS (
SELECT * FROM dbo.OperatorOption
WHERE OperatorID = #OperatorID
)
It does not matter you are inserting values as variables. It thinks there is nothing to insert.
You need to produce data that does not match.
Like this:
DECLARE #OperatorID INT = 3, #CurrentCalendarView VARCHAR(50) = 'month';
declare #t table (operatorID int, CurrentCalendarView varchar(50));
insert into #t values (2, 'year');
MERGE #t AS TARGET
USING (SELECT #OperatorID, #CurrentCalendarView) AS source (operatorID, CurrentCalendarView)
on (TARGET.operatorID = Source.operatorID)
WHEN MATCHED THEN
UPDATE SET TARGET.CurrentCalendarView = #CurrentCalendarView
WHEN NOT MATCHED BY TARGET THEN
INSERT (OperatorID, CurrentCalendarView)
VALUES (source.OperatorID, source.CurrentCalendarView);
select * from #t
Insert probably isn't working because your source CTE does not produce any rows. Depending on how your table is organised, you might need to select from some other source, or use table valued constructor to produce source data.
I have a bigint PK column which is NOT an identity column, because I create the number in a function using different numbers. Anyway, I am trying to save this bigint number in a parameter #InvID, then use this parameter later in the procedure.
ScopeIdentity() is not working for me, it saved Null to #InvID, I think because the column is not an identity column. Is there anyway to select the record that was just inserted by the procedure without adding an extra ID column to the table?
It would save me a lot of effort and work if there is a direct way to select this record and not adding an id column.
insert into Lab_Invoice(iID, iDate, iTotal, iIsPaid, iSource, iCreator, iShiftID, iBalanceAfter, iFileNo, iType)
values (dbo.Get_RI_ID('True'), GETDATE(),
(select FilePrice from LabSettings), 'False', #source, #user, #shiftID, #b, #fid, 'Open File Invoice');
set #invID = CAST(scope_identity() AS bigint);
P.S. dbo.Get_RI_ID('True') a function returns a bigint.
Why don't you use?
set #invId=dbo.Get_RI_ID('True');
insert into Lab_Invoice(iID,iDate,iTotal,iIsPaid,iSource,iCreator,iShiftID,iBalanceAfter,iFileNo,iType)
values(#invId,GETDATE(),(select FilePrice from LabSettings),'False',#source,#user,#shiftID,#b,#fid,'Open File Invoice');
You already know that big id value. Get it before your insert statement then use it later.
one way to get inserted statement value..it is not clear which value you are trying to get,so created some example with dummy data
create table #test
(
id int
)
declare #id table
(
id int
)
insert into #test
output inserted.id into #id
select 1
select #invID=id from #id
I'm dealing with a table in which a bunch of arbitrary settings are stored as VARCHAR(255) values. The particular one I'm tasked with dealing with is a sequence number that needs to be incremented and returned to the caller. (Again, note that the sequence "number" is stored as VARCHAR, which is something I don't have any control over).
Because it's a sequence number, I don't really want to select and update in separate steps. When I've dealt with this sort of thing in the past with actual numeric fields, my method has been something like
UPDATE TABLE SET #SEQ_NUM = VALUE = VALUE + 1
which increments the value and gives me the updated value in one swell foop. I thought in this situation, I'd try the same basic thing with casts:
DECLARE #SEQ_NUM VARCHAR(255)
UPDATE SOME_TABLE
SET #SEQ_NUM = VALUE = CAST((CAST(VALUE AS INT) + 1) AS VARCHAR)
WHERE NAME = 'SOME_NAME'
The actual update works fine so long as I don't try to assign the result to the variable; as soon as I do, I receive the following error:
Msg 549, Level 16, State 1, Line 4 The collation
'SQL_Latin1_General_CP1_CI_AS' of receiving variable is not equal to
the collation 'Latin1_General_BIN' of column 'VALUE'.
I understand what that means, but I don't understand why it's happening, or by extension, how to remedy the issue.
As an aside to fixing the specific error, I'd welcome suggestions for alternative approaches to incrementing a char sequence "number".
From one of the comments, sounds like you may have already hit on this, but here's what I would recommend:
UPDATE TABLE
SET VALUE = CAST((CAST(VALUE AS INT) + 1) AS VARCHAR)
OUTPUT inserted.VALUE
WHERE NAME = 'SOME_NAME'
This will output the new value like a SELECT statement does. You can also cast inserted.VALUE to an int if you wanted to do that in the SQL.
If you wanted to put the value into #SEQ_NUM instead of outputing the value from the statement/stored procedure, you can't use a scalar variable, but you can pump it into a table variable, like so:
DECLARE #SEQ_NUM AS TABLE ( VALUE VARCHAR(255) );
UPDATE TABLE
SET VALUE = CAST((CAST(VALUE AS INT) + 1) AS VARCHAR)
OUTPUT inserted.VALUE INTO #SEQ_NUM ( VALUE )
WHERE NAME = 'SOME_NAME'
SELECT VALUE FROM #SEQ_NUM
Maintaining a sequential number manually is by no means a solution I'd like to work with, but I can understand there might be constraints around this.
If you break it down in to 2 steps, then you can work around the issue. Note I've replaced your WHERE clause for this example code to work:
CREATE TABLE #SOME_TABLE ( [VALUE] VARCHAR(255) )
INSERT INTO #SOME_TABLE
( VALUE )
VALUES ( '12345' )
DECLARE #SEQ_NUM VARCHAR(255)
UPDATE #SOME_TABLE
SET [VALUE] = CAST(( CAST([VALUE] AS INT) + 1 ) AS VARCHAR(255))
WHERE 1 = 1
SELECT *
FROM #SOME_TABLE
SELECT #SEQ_NUM = [VALUE]
FROM #SOME_TABLE
WHERE 1 = 1
SELECT #SEQ_NUM
DROP TABLE #SOME_TABLE
You can continue using the quirky update in OP but you have to split the triple assignment #Variable = Column = Expression in the UPDATE statement to two simple assignments of #Variable = Expression and Column = #Variable like this
CREATE TABLE #SOME_TABLE (
NAME VARCHAR(255)
, VALUE VARCHAR(255) COLLATE Latin1_General_BIN
)
INSERT #SOME_TABLE SELECT 'SOME_NAME', '42'
DECLARE #SEQ_NUM VARCHAR(255)
/*
-- this quirky update fails on COLLATION mismatch or data-type mismatch
UPDATE #SOME_TABLE
SET #SEQ_NUM = VALUE = CAST((CAST(VALUE AS INT) + 1) AS VARCHAR)
WHERE NAME = 'SOME_NAME'
*/
-- this quirky update works in all cases
UPDATE #SOME_TABLE
SET #SEQ_NUM = CAST((CAST(VALUE AS INT) + 1) AS VARCHAR)
, VALUE = #SEQ_NUM
WHERE NAME = 'SOME_NAME'
SELECT *, #SEQ_NUM FROM #SOME_TABLE
This simple rewrite prevents db-engine complaining on difference in data-type between #Variable and Column too (e.g. VARCHAR vs NVARCHAR) and seems like a more "portable" way of doing quirky updates (if there is such thing)
I am trying to match data from an external source to an in house source. For example one table would have a field with a value of "black blue" and another table would have a field with a value of "blue black". I am trying to figure out how to check if all individual words in the first table are contained in a record the 2nd table in any order. It's not always two words that need to be compared it could be 3 or 4 as well. I know I could use a cursor and build dynamic sql substituting the space with the AND keywod and using the contains function but I'm hoping not to have to do that.
Any help would be much appreciated.
Try doing something like this: Split the data from the first table on the space into a temporary table variable. Then use CHARINDEX to determine if each word is contained in the second table's record. Then just do this for each word in the first record and if the count is the same as the successful checks then you know every word from the first record is used in the second.
Edit: Use a Split function such as:
CREATE FUNCTION dbo.Split (#sep char(1), #s varchar(512))
RETURNS table
AS
RETURN (
WITH Pieces(pn, start, stop) AS (
SELECT 1, 1, CHARINDEX(#sep, #s)
UNION ALL
SELECT pn + 1, stop + 1, CHARINDEX(#sep, #s, stop + 1)
FROM Pieces
WHERE stop > 0
)
SELECT pn,
SUBSTRING(#s, start, CASE WHEN stop > 0 THEN stop-start ELSE 512 END) AS s
FROM Pieces
)
Here's another method you could try, you could sample some simple attributes of your strings such as, length, number of spaces, etc.; then you could use a cross-join to create all of the possible string match combinations.
Then within your where-clause you can sort by matches, the final piece of which in this example is a check using the patindex() function to see if the sampled piece of the first string is in the second string.
-- begin sample table variable set up
declare #s table(
id int identity(1,1)
,string varchar(255)
,numSpace int
,numWord int
,lenString int
,firstPatt varchar(255)
);
declare #t table(
id int identity(1,1)
,string varchar(255)
,numSpace int
,numWord int
,lenString int
);
insert into #t(string)
values ('my name');
insert into #t(string)
values ('your name');
insert into #t(string)
values ('run and jump');
insert into #t(string)
values ('hello my name is');
insert into #s(string)
values ('name my');
insert into #s(string)
values ('name your');
insert into #s(string)
values ('jump and run');
insert into #s(string)
values ('my name is hello');
update #s
set numSpace = len(string)-len(replace(string,' ',''));
update #s
set numWord = len(string)-len(replace(string,' ',''))+1;
update #s
set lenString = len(string);
update #s
set firstPatt = rtrim(substring(string,1,charindex(' ',string,0)));
update #t
set numSpace = len(string)-len(replace(string,' ',''));
update #t
set numWord = len(string)-len(replace(string,' ',''))+1;
update #t
set lenString = len(string);
-- end sample table variable set up
-- select all combinations of strings using a cross join
-- and sort the entries in your where clause
-- the pattern index checks to see if the sampled string
-- from the first table variable is in the second table variable
select *
from
#s s cross join #t t
where
s.numSpace = t.numspace
and s.numWord = t.numWord
and s.lenString = t.lenString
and patindex('%'+s.firstPatt+'%',t.string)>0;