Bug 116309 - [Undo] NPE in OperationHistoryActionHandler #2
Summary: [Undo] NPE in OperationHistoryActionHandler #2
Status: VERIFIED FIXED
Alias: None
Product: Platform
Classification: Eclipse Project
Component: UI (show other bugs)
Version: 3.1   Edit
Hardware: PC Windows XP
: P3 normal (vote)
Target Milestone: 3.1.2   Edit
Assignee: Susan McCourt CLA Friend
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-11-14 14:26 EST by Michael Van Meekeren CLA Friend
Modified: 2006-01-10 20:08 EST (History)
0 users

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Michael Van Meekeren CLA Friend 2005-11-14 14:26:32 EST 
see bug 115761 , this bug is to track and test the fix for the 3.1.2 stream and
should be verified as soon as a build is available.
Comment 1 Susan McCourt CLA Friend 2005-11-14 19:17:12 EST 
fix was released in the R3_1_maintenance branch >20051114.
to be verified on next 3.1.2 build.
Comment 2 Susan McCourt CLA Friend 2005-11-16 18:50:52 EST 
Per explanation in bug #116673, I'm adding extra-paranoid safeguards to 
OperationHistoryActionHandler in both 3.1.2 and 3.2 stream.   Leaving this bug 
open until next maintenance build after M20051116.
Comment 3 Susan McCourt CLA Friend 2005-11-16 19:00:02 EST 
errr...I mean leaving this bug marked "fixed" until the new safeguards from 
bug 116673 can be verified on the 3.1.2 stream
Comment 4 Michael Van Meekeren CLA Friend 2006-01-10 10:27:02 EST 
while testing 3.1.2 I realized it would be nice to have a test or other sample to reproduced this bug for verification and regression testing in future.  Do we have a regression test for this?  (likely needs to be a session test as it is a shutdown issue)
Comment 5 Susan McCourt CLA Friend 2006-01-10 11:20:01 EST 
I'll put a regression test in today as part of verifying this bug.
Comment 6 Susan McCourt CLA Friend 2006-01-10 20:08:40 EST 
verified on M20060109-1200 through inspection of source.
I agree a regression test would be nice.  I tried several approaches to force the error in a "natural way" and cannot do so after several quite detailed attempts, including a forced dispatch of async messages after setting up the condition.  It appears the original test case that found the problem was actually running the action handlers, so simply closing a workbench window while performing asyncs from the action handler does not cause the problem.  I will pursue putting together a regression test using code from the original bug reporter.

For the purposes of this bug, marking it verified.  Any regression test added will be added to the 3.2 stream.  See bug #115761 for info on the test case.