|
Marking this as a regression as it appears to be a serious failure where the tests expects results and gets:
> SQLSTATE(null): java.sql.SQLException: Stream of column value in result cannot be retrieved twice I don't think this is a regression. Yes, previously you could call getXXXStream() in embedded twice on the column to read the stream twice, but the second read returned the wrong results. Look at the comments et c. in
The change was discussed on the list, the JDBC spec says portable code should only read a column once. So now we have getXXXStream(0 calls which can only be called once, and thus always return the correct results. Any applications depending on the old behaviour are returning the worng results ot their users. Thanks Dan,
It sounds like the best thing to do is to fix up the 10.1 test to do the right thing and document in the release notes as a change that might affect existing applications. Does that sound correct? - Changing the priority from 'Critical' to 'Medium'.
- Since the tests in 10.1 branch needs to be updated, the fix verison should also reflect that - 10.1.2.3 Removing this from :"Regression Test Failure". Dan pointed out that this is probably an intended behaviour change from
Unsetting Fix Version on Unassigned issues. Sending mail for these because these should be reviewed to see if they are still valid issues.
Comments indicate that this is an expected failure due to
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
DERBY-721r357435 | tmnk | 2005-12-17 21:15:39 -0800 (Sat, 17 Dec 2005) | 1 line
-
DERBY-721_4 - State of InputStream retrieved from resultset is not clean , ifthere exists previous InputStream . Patch to throw exception if user get stream
for a value in result set twice . by Tomohito Nakayama (tomonaka@basil.ocn.ne.jp)
The test was returning results but now throws this error when combining 10.2 client with 10.1 server, so at first glance looks like a regression.