Hi Jos

(loooong time no see - everything ok ?)

Importing from Excel always has some risks.
For data not allowing mistakes I always did a 2 step dance:
import into char or memo fields
then fill correct field structure from there.
Getting the field structure from import wizard is ok,
you might run even an import on the data to compare later.

Then copy Stru twice to new tables,
alter fields to char(maxlen) or memo in one and
fill that from a CSV you created earlier from the excel file.
All data is readable, no problem to add validation
or import in a loop.

Had to do this with marketing or poll data a couple of years
regularly - more then 3 dozen per year and less than 25%
had no error to fix
(typical was phone number in excel not column - defined as numeric
or column entry error inserting a blank field or overwriting one entry).

We also built a routine to read Excel via COM-
either to verify a specific column for all rows on field type
or to document Excel field types in every row -
checking a few columns more than the table was supposed to have

Do not read calculated data if possible - calculate yourself,
as the formula of 1 row might have been damaged.

HTH
thomas

>Hello All, it's been a while!
>
>I have an XLS file (5/95 version). I am trying to import this into a DBF using the APPEND FROM ... TYPE XL5 command. It works but I get an error with Numeric Overflow - Data was lost. The XLS file can contain #NUM values in some numeric cells - the data is not important and I dont mind losing that. All the other data is imported OK. Question: is it possible to import the file without the program being halted to display this overflow error message?
>
>TIA