Cross post

See here also http://www.sqlservercentral.com/Forums/Topic406155-338-1.aspx

Let me rephrase that...

How do you know the count is 1 for both records?

How do you know how to get the count?

DECLARE @Target TABLE (RowID INT, Data VARCHAR(12))

INSERT @Target

SELECT 1, 'Peso' UNION ALL

SELECT 2, 'rm' UNION ALL

SELECT 3, 'Nosepicker'

DECLARE @Source TABLE (Data VARCHAR(12))

INSERT @Source

SELECT...

Also see

http://www.sqlservercentral.com/articles/Advanced+Querying/pivottableformicrosoftsqlserver/2434/

Same class?

http://www.sqlservercentral.com/Forums/Topic406041-9-1.aspx#bm406087

It's called SCOPE.

After ALTERING a table, an insert can't take place directly.

However, there is a workaround

CREATE PROCtest_proc1

AS

BEGIN

CREATE TABLE test_table1 (col1 int not null, col2 varchar(50) null)

INSERT INTO test_table1 (col1, col2)...

You can also specify which columns to check for UPDATE on.

Then there will be no nested triggers at all!

CREATE TRIGGER trRestorder ON eStore_CatalogProducts

FOR UPDATE

AS

IF UPDATE(CanBeOrdered) OR UPDATE(StockValue)

UPDATEcp

SETcp.AvailableFrom = '20400101'

FROMeStore_CatalogProducts AS...

DECLARE @r VARBINARY(100)

SET @r = 0x5065736F

select cast(@r as varchar(100))

1) SELECT * FROM INFORMATION_SCHEMA.ROUTINES

2) SELECT OBJECT_ID('...')

3) SELECT * FROM sysobjects where...

The SQL Server does not know which columns to insert!!!

CREATE PROC test_proc1

AS

BEGIN

CREATE TABLE test_table1 (col1 int not null, col2 varchar(50) null)

INSERT INTO test_table1 VALUES (101, 'Temp Data1')

INSERT INTO test_table1 VALUES...

It's hard to...

Thank you.

I thought a 90 times faster speed improvement was worth posting.

The real performance killer in the original query is the "uid2 NOT IN" operation.

SELECT COUNT(Col1),

COUNT(Col2),

COUNT(Col3),

COUNT(*)

FROM Table1

See second version of fnParseList function here

http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=76033