I still think your first answer was better (Excel 2007). Even better than that, however,...

Instead of using a File System task, use a script task containing appropriate code. It works much better, in my experience and requires just three or four lines of code.

OK ... may I ask what 'Transfer SQL Server Objects' gives you that a dataflow task does not? Just wondering why you went down this road.

Sure - do a multicast to all of the tables you need to update ...

Why not just copy the data you require? Deleting & recreating table objects repeatedly seems a little inefficient.

Is this because you have the same data in multiple tables (ie, redundant data)?

If there is no delimiter, it is, by definition, not a CSV file.

It's a fixed-width file and you can use SSIS to import it and break it into its constituent...

ameda16 (7/27/2010)

---

an idea: use no cashing instead of full precaching, the default caching type

No caching, in which the reference dataset is accessed by each row in the rowset. To...

So your DF contains duplicates? Why not get rid of them early in the DF and save yourself the trouble? If not, what happens when the lookup is successful? How...

Looks like you are running that code for every row in your pipeline and then assigning a single result - presumably based on the values in the last row in...

Define your flat file source to be fixed width (not CSV). Then define your fields based on start pos / length etc - haven't done it for ages, so can't...

So you want the same value in every row?

That's fairly easy - here's how.

Create a package scope integer variable to hold the count.

Create an Execute SQL task to run the...

Please provide more detail - your requirement is difficult to understand. An example would help.

Have a look at the MERGE JOIN transformation - maybe that can do what you need. Difficult to know for sure without seeing examples of your data.

This thread looks very similar to this one:

http://www.sqlservercentral.com/Forums/Topic955373-148-1.aspx#bm955545

Why are you asking the same, or nearly the same, question again? What was wrong with the solutions proposed?