>>Yeah, yeah... but if he has a baseline of scanning possible tables, and basic testing, HE would know better the types of corruption he's getting and what to look for... I'm not trying to guess for him. I've been working w/VFP since 3, and Fox since 1987... I've only had 3 instances of file corruption my entire history working with people... The most recent was over 10 years ago and haven't seen any since.
>
>Let's get that somewhere halfway. If there was a corruption, it usually happens (at least did so in DOS 4.0 :) on whole blocks, not just individual fields. So there's no need to test ALL fields, just look for values which are out of whack, i.e. numbers out of range, impossible dates/datetimes, non-ascii strings. If there's a corruption, most likely there'll be a block missing somewhere in the middle of a table, or a foreign block inserted. In the first case, the next block will be misaligned, in the latter it's most likely to contain some binaries (from an image, someone's temp file, any garbage possible).
>
>So the short way to check for this would be to go through tagged fields, and check for the min() and max() of any such field. If any one of them is not in the expected range, alarm the admin to check manually.


GREAT SOLUTION ASSESSMENT Dragan... Even if you did a scan across all fields for the min/max would be a great starter for his needs.