Just a quick blurb on the sql

select min(key1), person_id, 'contract1_id', ...yada...

INTO #t2 --(from above)

FROM contract_connect_table

where contract_id in ('contract1_id', 'contract2_id')

GO

Delete from contract_connect_table

GO

insert into contract_connect_table

select * from #t2

I would think it would be possible to create a parm table with the different values.

Select yada

FROM parm_table

WHERE @time between time1 and time2

and @ServerTimeDif between svrTime1...

SELECT * INTO #JobInfo

FROM OPENROWSET('sqloledb', 'server=(local);trusted_connection=yes'

, 'set fmtonly off exec msdb.dbo.sp_help_job @execution_status=4')

This works to get...

Very correct, the above derived tables on the original link_table performance would rapidly degrade.

The simplest approach

SELECT distinct customer

FROM customer join link-table

where link-table.product in (yada)

Sorry, neglected to catch that in the question.

Another option would be to create the sql statement in a cursor

and then exectue the statement. Storing the result in a tmp...

Below is a fairly simple method. Note joining the table on itself (link table) can be cpu intensive.

SELECT distinct ct.customer_name

FROM (select customerid from linktable where productlid = a) a

JOIN...

While testing/digging into...

dynamic properties Task:

Connections

ODBC Connection (name)

OLE DB Properties

Extended Properties

.. attach a global variable to this value gv_file_dsn...