Michael J. Segel wrote:
And to Jean's point.
The reason you test the SQL syntax against another database is to help to
determine if you truly have a bug or not. Simple rule of thumb. 90% of the
time, what you think is a bug, is really something that you did wrong.
I think you may have confused my point with the points of others.
My point addressed the DERBY-539 documentation improvement.
It doesn't hurt to document actual product behavior -- and it doesn't
particularly matter how other database vendors handle it. *Derby*
behaves in a specific manner that can be documented more clearly to
avoid confusion for end users.
As for introducing behavior changes, of course we're interested in how
other databases handle it, especially if the behavior conforms to a
standard or not. But that was not the point of my post -- and it also
isn't the point of DERBY-539.
-jean
