When n=1, the first row of the big file is written to both output files. This is because it was
a header row containing field names and needed to be the first row in both output files. The
process keeps writing rows to Smallfile1 until it reaches 3,500,000 rows. At this point, it
begins writing rows to SmallFile2.
The result was two CSV files of a more manageable size that contained all the data of the
original big file. The files were linked into the Access database and the union query worked
successfully.
The point at which the output procedure changes from one file to the other is arbitrary, so
it takes a bit of guesswork to decide where best to set this.
Chapter 30: Handling Large Text Files 311