January 11, 2012 at 4:55 pm
According to BOL http://msdn.microsoft.com/en-us/library/ms141036.aspx, the only unsigned integer type that converts to a SQL data type is DT_UI1, which converts to tinyint.
I have a package that imports a flat file, and uses a MERGE JOIN inner join to filter the rows on only the ID column using a view on the SQL server. (These are never negative numbers.) The destination is a table on the SQL server. The join columns are an ID column. When I casted both as unsigned ints (DT_UI4), the import completes, but seems to leave out some rows in the destination table, even though those have a match in the view.
My question is, is the correct way to cast these columns is to cast both as DT_I4? (we need a four-byte int as the column is 7 characters long) And is it working partially because of implicit conversion?
Thanks much,
Elaine
January 25, 2012 at 10:09 am
Seems the left out rows problem may have been an SP issue. When experiencing the missing row problem we were running SP1. Since we have installed SP3 (SMS 2008), we have not had missing rows:
January 26, 2012 at 12:15 am
One small question, just to be sure:
were the two inputs for the MERGE JOIN sorted?
Need an answer? No, you need a question
My blog at https://sqlkover.com.
MCSE Business Intelligence - Microsoft Data Platform MVP
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply