Any kind of commerical utility will always involve having to map the fields, so if possible I prefer doing it in PHP or Perl for free. I don't know how many fields you are working with, but here are two alternatives. If you can get the flat file to windows and have MS Access then you can import the flat file into an Access table in which you have mapped all the fields, because it's easier to do visually if you have many fields. Then use ODBC to read the rows in the Access table, massage the data for MySQL, and write the fields back out to another flat file with comma delimiters. All code can be done easily with PHP or Perl. Then MySQL can automatically suck in the comma delimited file.
Another method would be to do everything in PHP or Perl. Read the lines in the flat file one by one and separate each field out, using substr() for example in PHP, massage each field for MySQL, then write out the fields to another flat file delimited.
I prefer PHP and always have the cgi version installed on my machines in order to perform tasks like these, although you could call the script through a browser. 900 records is quite small, but having to map all the fields is what will take time and if the data format isn't acceptable to MySQL. If you want sample PHP code for reading in flat files, processing the data, and writing out to another flat file let me know.
