there is no requirement that the insert statement have an explicit column list, although its obviously best practice. as long as the datatypes are compatible and there is not...

also from BOL (re: Inserts):

If a column is not in column_list, the SQL Server 2005 Database Engine must be able to provide a value based on the definition of the...

from BOL (about Computed Columns):

A computed column cannot be the target of an INSERT or UPDATE statement.

You could also use the Fuzzy Grouping Transformation in Integration Services for this.

is this what you're after:

DROP TABLE [#TEST]

CREATE TABLE #TEST ([UId] int, PId int, PV varchar(10) )

INSERT INTO #TEST VALUES(33138, 9, 'True')

INSERT INTO #TEST VALUES(33138, 8, 'False')

INSERT INTO #TEST VALUES(33138, 5,...

might speed up a little to parameterize and then toggle between the SELECTs for review and then UPDATE:

DECLARE@old VARCHAR(20)

, @new VARCHAR(20)

SET@old = 'Roda'

SET@new = 'Road'

UPDATEAddressSETStreetName = REPLACE(StreetName,' '+@old+' ',' '+@new+'...

Its the Name of the task and its description. By default, these are both "Send Mail Task" until you change thru edit (or expression). Not real useful to...

Could you add space on either side of the string you are replacing? That would eliminate most false catches.

UPDATE Address

SET StreetName = REPLACE(StreetName,' Roda ',' Road ')

WHERE StreetName LIKE...

take a look at this thread and see if that helps: http://www.sqlservercentral.com/Forums/Topic577774-148-1.aspx

here is a really good demo.

http://www.jumpstarttv.com/using-the-file-system-task-in-ssis_508.aspx

Yes, with an expression. I've attached an image to demonstrate.

Try the system variable @[System::SourceName]

You can do it using a ForEach Loop Container using a File enumerator. Your File System Task would be inside the ForEach Loop. No scripting necessary, just some...

I did a little test yesterday afternoon against a query with 2 varchar(5) columns, ~250k rows (109 - 123 distinct records*). I built 2 Data Flows, with one Data...

My two cents would be to use the TSQL. I've never tested the ssis Aggregate transform, but would be surprised if it outperformed sql server (of course it probably...