Hi,
I am reading Excel sheets using the Excel Source in a Data Flow. The driver decides the data type of the column based on the first 8 ( ) values which it reads. How can this default be changed to, say, 1000, please
This is because I am reading columns in which the first, say, 20 cells have no values and this is causing the driver to make ALL the values in the column to be null.
Thanks.

Reading Excel Sheets - Column DataTypes - Change default
thepaul
Hope this works
-Raj
jh2005
For information about increasing the number of rows sampled by increasing the value of TypeGuessRows under the HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Jet\4.0\Engines\Excel registry key, see
-Doug
onur6n
And, it is not possible to read an Excel file using a Flat File Connection Manager.
Does anybody else have any other suggestions
Thanks.
reachrishikh
Once upon a time, it was theoretically a connection string option named MaxScanRows for the Excel ODBC driver, but oops! someone forgot to actually implement it in the code and it never worked.
Tweaking the registry value is the only option. I suppose that this could be done in a Script task...in fact, you could set it back to its previous setting at the end of the package.
The default setting always used to be 8, but I've found machines where it's set to 25 -- not sure whether Office does that I believe I've also read in the past that 0 (zero) = "all rows," but I don't have that in writing, and analyzing 16,000+ rows to guess at a column's datatype would be a bit excessive.
-Doug
williamwon
jay2068