Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 356415 - [variables] Refreshing view followed by collapse-all action creates blank entry in view.
Summary: [variables] Refreshing view followed by collapse-all action creates blank ent...
Status: CLOSED DUPLICATE of bug 356414
Alias: None
Product: Platform
Classification: Eclipse Project
Component: Debug (show other bugs)
Version: 3.7   Edit
Hardware: PC Linux-GTK
: P3 major (vote)
Target Milestone: ---   Edit
Assignee: Platform-Debug-Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on: 356414
Blocks:
  Show dependency tree
 
Reported: 2011-08-31 17:23 EDT by Pawel Piech CLA
Modified: 2012-01-25 12:18 EST (History)
2 users (show)

See Also:

Attachments
Screenshot of bug. (132.40 KB, image/png)
2011-08-31 17:24 EDT, Pawel Piech CLA 		no flags Details
Workaround patch in case fixing the underlying bug 356414 is a problem. (704 bytes, patch)
2011-09-06 18:20 EDT, Pawel Piech CLA 		no flags Details | Diff
View All 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 Pawel Piech CLA 2011-08-31 17:23:01 EDT 
This bug applies only if a debugger refreshes a view (using IModelDelta.REFRESH), or user refreshes it through a debugger-contributed action.  Then user selects collapse-all action.

What follows is a view that looks very broken (see attachment).

This bug seems to be caused by a low-level SWT issue (bug 356414), but it could be worked around by having the collapse-all action call TreeViewer.refresh() after calling collapseAll().
Comment 1 Pawel Piech CLA 2011-08-31 17:24:29 EDT 
Created attachment 202569 [details]
Screenshot of bug.
Comment 2 Pawel Piech CLA 2011-09-06 18:20:27 EDT 
Created attachment 202848 [details]
Workaround patch in case fixing the underlying bug 356414 is a problem.
Comment 3 Martin Oberhuber CLA 2012-01-24 04:37:39 EST 
Pawel, how do you want to proceed on this ?
Comment 4 Michael Rennie CLA 2012-01-24 14:18:29 EST 
(In reply to comment #2)
> Created attachment 202848 [details]
> Workaround patch in case fixing the underlying bug 356414 is a problem.

My vote would be to not apply a workaround to debug.

Reasons for my vote:
1. the encounter rate for this issue is very low
2. the workaround introduces a negative performance impact
3. we will inevitably forget about the workaround, and when SWT fixes the other bug we will have a performance hit with no side benefits.

I would mark this issue as 'readme' and push on SWT to fix the underlying problem.
Comment 5 Pawel Piech CLA 2012-01-25 00:15:33 EST 
(In reply to comment #4)
> My vote would be to not apply a workaround to debug.
I don't think there's any real performance impact in the workaround, but I'd also rather fix the real bug, especially since it's a low level one.
Comment 6 Pawel Piech CLA 2012-01-25 12:18:29 EST 
Fixed by SWT

*** This bug has been marked as a duplicate of bug 356414 ***