Yuk, nasty. Easiest solution: change the field delimiter character to something else in the source data.

Otherwise (and this may not work perfectly), you could consider running a replace routine on...

Use a UNC path to get round the unmapped drive issue.

Ensure that the SQL Agent service user is a domain user which has at least read access to the folder...

I guess I'm just not understanding what you're trying to do.

What is the scope of your variable? It needs to be package level, not task level.

What is written to the...

I suggest that you change your approach slightly.

To get the variable evaluated as you want, you need to put the SQL into an expression (and the expression can use your...

So the SQL command text is in a variable. How are you executing the command?

--edit

Are you executing the command through the use of an Expression?

A single package will do the job for you. An EXECUTE SQL task can be used to run the sp's. Followed by one or more dataflow tasks to export the...

Where are you entering the text of the SQL query containing the variable?

Wow, you just got a cool new name! 😀

Well done on solving that one by yourself. My solution would have worked too though - at no point did I mention deleting dups, all I said was 'processing' them,...

A variant of the temp table method is probably still the easiest for you.

Bring all the data into a temp table, run a query to identify & process the dups...

I think you might need to MULTICAST to enable the 'reuse' you are after ...

This sounds like a pure SQL question, not SSIS? Anyhow, there is a way, using Common Table Expressions (or CTEs).

Using your example, it would be something like:

With Data as

(select productname,...

I think I detect a hint of sarcasm 🙂

Good luck - please post back if you find the solution.

But did you try changing the datatype?