I have a large SDF-format text file (around 80,000 lines) that I want to import into a VFP cursor, do some processing, then insert each record into an empty SQL table.

What is the best method to update the back-end table? I would like to use a table-buffered remote view and TABLEUPDATE(.T.), but as far as I can see with the view-designer, in order to do this I need a primary key, and with this data-set there is no single field or combination of fields that can be guaranteed to be unique for each record.

Another method is to use SPT with 'INSERT INTO MyTable (field list) VALUES (value list)', but the table contains 27 fields, which makes this option rather cumbersome, and I guess there will be a large performance hit if called once for each record.

Are there any other ways around this problem? Any help would be appreciated.


Alan