Unfortunately, I am not at liberty to show a sample of the data we intend to load. Besides, without the associated documentation, the data raw would be exceedingly difficult to read. However, I can give you more details as to the general structure.
The original datafile contains hundreds of thousands of rows of positional data. Each row is assumed to have the same character length. Character position 5 designates the category of the row. The position of the subcategory changes depending on the category value.
For instance, if the category is "A", then the subcategory would always be located at character position 13. If the category is "P", then the subcategory would always be located at position 6.
Much like how the category value determines the subcategory placement, the category and subcategory values together also determine how the rest of the columns are positioned in the row.
In order to determine which table the row belongs to and which character positions refer to which columns, you would have to:
1. Determine the position of the subcategory value based on the category value at position 5
2. Determine the position of the rest of the columns based on both the category and subcategory values.
I hope this helps. Thanks for the tip. I'll see if tFileInputMSPositional has what I need.