Hello, I have to import flat files with headers and footers. I found how not to import headers but still have issues with footers. Is there a quick way to do that?

You can use a first dataflow to read the file (no destination component need it) and store the number of rows in a variable; then in a second dataflow, add an extra column to enumerate each row, and use a conditional split to discard the rows last x rows.

Just to clarify a bit, if your footer has a different format than the rest of the rows, this might be a bit more complicated. The first data flow needs to import the entire row as a single string column. You'd then have to re-import them as a single column, use the conditional split to remove the footers) and save it to a new flat file. After that, you can reload them again with the appropriate column definitions. There are other methods of handling this: http://www.sqlservercentral.com/articles/Integration+Services/61880/ http://agilebi.com/cs/blogs/jwelch/archive/2007/05/07/handling-flat-files-with-varying-numbers-of-columns.aspx

Adding a row number and conditional split will still produce the warning, as the file is being read by the flat file source adapter. Might not help much. Single column read approach has less code then using a script task to parse out footers before using the flat file source adapter.GagandeepSingh