The EXTRA_DATA error/warning was a thorn in my side for the last couple of days. I added a column to a dataset I was loading to a model from a Text File Reader, and I could not figure out for the life of me what I was doing different from before the column add. It turned out, where I was trying to unselect data was too early.
Problem: Suppose you are bringing in a delimited file to use in a Text File Reader that has an extra field at the beginning, which you don't want to have in your output. Finding the correct place to disable this data can difficult. If you disable it before your Text File Reader, or simply don't bring it into your Text File Reader, then you run the risk of getting an error or warning that states EXTRA_DATA for every row you try to bring in, as seen below.
I have seen the EXTRA_DATA error in 3 scenarios
- Deselecting the “Use” box in the Metadata definition on columns you don’t want.
-
Selecting all columns in the Metadata definition, but not bringing the column into your Text File Reader
-
- Or a combination of the two above cases
Solution: Keep all columns selected in the Metadata definition, and bring the column into the Text File Reader, and select the “Ignore” check box in the Text File Reader.
While this is a very specific case, I hope it brings relief to at least a few other developers!
