I have some data in an Excel spreadsheet which includes legitimate #N/A values. I need to exclude these values when loading them into a fact table. I was not able to use a row filter to eliminate values that were null. Instead, I had to set a NOT NULL constraint on the target column, turn of batch updates, and set the ignore insert errors.

By rights, the Excel Input module needs its own application-specific filter to deal with #DEF, #N/A, #VALUE!, #DIV/0, #NUM!, #NULL!, #NAME?, #REF! and null values. Whether any of these also belong in the row filter is a matter of debate. The filter probably needs to handle NAN errors at a minimum.

Post edited by: pquirk, at: 05/23/2006 11:27