I would typically load data via bulk insert from a text file into a database. All as nvarchar. I have it all setup as stored procedures which means very fast turn around with minimal rewriting of the scripts. Occasionally however the amount column can cause as issue. For example some extracts we receive contain "-" for 0, others will show negative figure as (10.00). This means when I try to cast amount column as a float I get an error.
What is the best way to catch and identify these random characters in an automated fashion?