For the rows you have not seen before you would not stage. You would use
an OLE DB Destination adapter stright into SQL. You shouldn't really stage
to text file either. The recommended way would be to use a Raw File to do
> Allan,
>
> thank you!
>
> here is what I got from your message, correct me if i am wrong. thank
> you!
>
> 1. use Lookup and save those new records (cannot find records) to a
> flat file. (There is no PKey in my flat file, I will test if SSIS can
> lookup for all rows/cells)
> 2. insert the data in the flat file to the SQL database. (as a SET)
> do you know there is a better way than this? I just wonder if the
> records keep on increasing, then it will be slower and slower.
>
> thank you!
>
> On Jul 2, 10:20 pm, Allan Mitchell <a...@no-spam.sqldts.com> wrote:
>
>> Hello Bob,
>>
>> You need to handle this yourself and the usual way to do this would
>> be to
>> either use a Lookup transform or a Merge Join transform. You can
>> still choose
>> to handle this the ELT way and stage the rows from the flat file and
>> then
>> do a SET based INSERT afterwards but the transforms are the cuter way
>> to go.
>> --
>>
>> Allan
>> Mitchell
http://wiki.sqlis.com|
http://www.sqlis.com|
http://www.sqldts. >> com|
http://www.konesans.com >>
>>> Hi,
>>>
>>> does anyone know whether there is an option in SSIS to insert only
>>> difference? For example, I have an appended txt file to insert. I
>>> have inserted the first part of data a while ago so I only want to
>>> insert the new records this time.
>>>
>>> thank you!- Hide quoted text -
>>>
>> - Show quoted text -
>>