Bug 493819 - Real instead of pseudo conflicts on containment ref changes when resource attachment change involved
Summary: Real instead of pseudo conflicts on containment ref changes when resource att...
Status: ASSIGNED
Alias: None
Product: EMFCompare
Classification: Modeling
Component: Core (show other bugs)
Version: 3.1.0   Edit
Hardware: PC Windows NT
: P3 normal (vote)
Target Milestone: ---   Edit
Assignee: Laurent Delaigue CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-05-17 10:34 EDT by Laurent Delaigue CLA
Modified: 2016-05-18 09:09 EDT (History)
3 users (show)

See Also:


Attachments
Test case to reproduce the issue (58.37 KB, application/zip)
2016-05-17 10:34 EDT, Laurent Delaigue CLA
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Laurent Delaigue CLA 2016-05-17 10:34:32 EDT
Created attachment 261800 [details]
Test case to reproduce the issue

Conflicts are detected as REAL instead of PSEUDO between containment ref changes when they are contained in a resource that is controlled/uncontrolled

To reproduce, compare branches "expected3" with "expected" with MatchBasedConflictDetector active.
Many real conflicts are displayed, they should all be pseudo-conflicts.
Note that this does not happen when comparing expected2 with expected because then there are no ResourceAttachmentChanges involved.
Comment 1 Eclipse Genie CLA 2016-05-17 10:38:01 EDT
New Gerrit change created: https://git.eclipse.org/r/72932
Comment 2 Philip Langer CLA 2016-05-17 10:42:22 EDT
This bug might be related to bug 493527 and its fix, right? At least it sounds like the symptoms are pretty similar.
Comment 3 Laurent Delaigue CLA 2016-05-18 09:09:54 EDT
(In reply to Philip Langer from comment #2)
> This bug might be related to bug 493527 and its fix, right? At least it
> sounds like the symptoms are pretty similar.

No quite, although the two bugs lie in the same area, clearly.
In bug 493527, the problem only exists with the Default Conflict Detector. This old conflict detector is known to have issues and for the time being we focus the fixes on the scalable conflict detector.
We'll look at the default conflict detector later on once the wrong conflicts are dealt with.