|
||||||||
|
This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira |
||||||||
|
||||||||
|
This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira |
||||||||
That's difficult because I'd have to find out what happens in spring's classes concerning transaction management, re-implement the stuff there, setup everything without spring. At the moment that's not possible for me.
But please have a look to NonBatchingBatch and my fixed version. The changes I made do not change the behaviour of this class other than ensuring that all the cleanup that happens when the query succeeds also happen when the query fails. Also code from the abstract superclass could be used instead of the (incompletely) copied code that is pressent there at the moment. Independent of the question if this is really a bug or not it will improve the code quality.