Re: getUdateCount() vs. RETURNING clause

[Thomas Kellerer <spam_eater(at)gmx(dot)net>]

Oliver Jowett, 25.11.2009 01:40:
> Thomas Kellerer wrote:
>
> > boolean hasResult = pstmt.execute();
> >
> > if (hasResult) {
> >  ResultSet rs = pstmt.getResultSet();
> >  if (rs != null && rs.next()) {
> >    int newId = rs.getInt(1);
> >    System.out.println("newid: " + newId);
> >  }
> > }
> >
> > int affected = pstmt.getUpdateCount();
> > System.out.println("affected: " + affected);
> >
> > I do see the returned ID from the ResultSet but getUpdateCount() always
> > returns -1 even though I know that a row was updated.
> > What am I missing here?
>
> Update counts and result sets (for a particular result) are mutually
> exclusive. If getResultSet() returns non-null then getUpdateCount() must
> return -1; see the javadoc for those two methods.

Hmm, my understand was a bit different. 

The Javadocs simply say

There are no more results when the following is true:
((stmt.getMoreResults() == false) && (stmt.getUpdateCount() == -1))

It doesn't state that they are mutually exclusive. And for me it seemed to imply that I can call getMoreResults() and getUpdateCount() in a loop in order to process everything that is returned.

Additionally the Javadocs for getUpdateCount() says:

"Gets the *current* result as an update count" and ".. if there are no more results it returns -1"

The word "current" here also let me to believe I can call those methods multiple times. 

Regards
Thomas