Ticket #627 (accepted defect)
Opened 2011-06-21T11:01:32-05:00
Last modified 2013-01-22T15:01:00-06:00
Rotating image can cause window size to be wrong
| Reported by: | bdezonia | Owned by: | curtis |
|---|---|---|---|
| Priority: | major | Milestone: | imagej2-unscheduled |
| Component: | UI Framework | Version: | |
| Severity: | serious | Keywords: | |
| Cc: | Blocked By: | ||
| Blocking: | #1629 |
Description (last modified by bdezonia)
Using 7feeec6da90f475fea86cb0e43bfca37dcdfb828: If you load a new Image and rotate it by 90 the window size can be wrong. Usually its wrong just the first time you rotate (the window does not resize). But I have had it that after that failed window transformation further rotate 90s are also wrong over and over as the untransformed original window then also rotates by 90 and stays off by 90.
Change History
comment:1 Changed 2011-06-21T13:11:04-05:00 by bdezonia
comment:2 Changed 2011-06-21T13:14:30-05:00 by bdezonia
- Summary changed from Rotateing image can cause window size to be wrong to Rotating image can cause window size to be wrong
comment:4 Changed 2011-07-29T13:32:18-05:00 by bdezonia
- Milestone changed from imagej-2.0-beta1 to imagej-2.0-alpha5
comment:6 Changed 2011-08-18T13:13:38-05:00 by bdezonia
- Status changed from new to closed
- Resolution set to fixed
Fixed in 64c877fbd4981345c0994d2c3e6016acf8fbfb7a. Had to rework some code that may have been added to avoid some other bug. Commented and dated decision in source in case we rediscover bug.
comment:7 Changed 2011-09-26T14:20:36-05:00 by bdezonia
- Status changed from closed to reopened
- Resolution fixed deleted
comment:8 Changed 2011-09-26T16:09:48-05:00 by bdezonia
Can still duplicate this bug in alpha 5 and also latest version. There must be a race condition that was never realy fixed. Or switching from ThreadSafeEventService to SwingEventService has somehow caused problems. Investigate.
comment:9 Changed 2011-09-28T15:21:48-05:00 by bdezonia
- Owner changed from bdezonia to curtis
- Status changed from reopened to assigned
At times I have been able to "fix" this bug with changes to when updates(), redoLayouts(), and invokeLaters() happen. But no longer.
When these "fixes" have worked bug #605 would also go away. I believe they are the same bug. I will be closing #605 while noting that any fix to this bug should test that #605 is also fixed.
In the meantime I have made some code changes to make calls to update() and redoLayout() consistent throughout IJ2. Those commits were: 77667ced3ad5d107f906a72036cd6b3a4600a5ca, 2ea227f8de12310eda084de55f20b16772de94c8, & a9a7f0adcc653f2e7cadc1726671a49ecdc8e6c0.
This ticket will go to a back burner until Curtis' work on eliminating redoLayout() shakes out.
comment:10 Changed 2011-10-24T12:04:24-05:00 by curtis
- Milestone changed from imagej-2.0-alpha5 to imagej-2.0-beta1
Waiting on ticket #828.
comment:12 Changed 2012-02-27T10:59:37-06:00 by curtis
- Milestone changed from imagej-2.0-beta1 to imagej-2.0-beta2
comment:14 Changed 2012-07-12T10:26:18-05:00 by curtis
- Milestone changed from imagej-2.0.0-beta3 to imagej-2.0.0-beta4
comment:16 Changed 2012-09-06T14:06:14-05:00 by curtis
- Blocking 1358 added; 1325 removed
- Milestone changed from imagej-2.0.0-beta4 to imagej-2.0.0-beta5
comment:17 Changed 2012-12-12T09:31:01-06:00 by bdezonia
Still a problem on 12-12-12
comment:19 Changed 2013-01-22T15:01:00-06:00 by bdezonia
- Milestone changed from imagej2-b7-ndim-data to imagej2-unscheduled
Seems to happen regularly with Boats