The documentation (word) file that describes the format does not document several fields that are included in the header line of the CSV file. And there are 85 fields in the header (first line), but the remaining lines of CSV data have 89 elements!

Given that I am attempting to write a parser for this CSV file this is causing me fits!

Is there any way to clean up the documentation so that it matches the actual format, and maybe get the header line to match the actual body of the data since CSV is all about a 1-to-1 match between the header and row level data, and its nearly impossible to "guess" the order of the fields if the header line is in error?