>>>>Hi,
>>>>
>>>>Say I have a table that has column MY_INT_FLD I(4). There could be some records in this table that have the same value in this field. And some records that have 0 in this field. I want to create one or two (or maybe even three) UPDATE command(s) that will set each record of this table to a unique value and it has to be greater than 0. Is it possible? TIA.
>>>
>>>Try:
>>>

>>>update myTable set Field = Field + 1000  + recno() ;
>>>where Field in (select Field from myTable ;
>>>group by Field having count(*) > 1) 
>>>

>>>I add 1000 assuming that the total number of records is less, so we always end up with unique values after update.
>>
>>Thank you. The goal was not just to have a unique value (what Sergey suggested would have done it). But the goal was to preserve all the existing values and only change the value in records where duplicate exists.
>
>That's exactly what my script does, if you noticed. I only update fields which are duplicates. The having clause should be > 1, not > 0 (typo).

I am curious, as I try to understand your syntax and learn how UPDATE works. It seem like VFP, when executing UPDATE, processes all records at once and then actually updates the values. Otherwise, in the case of two records with the same value, only one would be set to a new value. That is, when the first instance of the duplicate record is set to a new value, subsequent sub-query would not found duplicate records. Do I understand this correctly?