Project

General

Profile

Bug #2918

Bug #2677: fix drawing and functional differences between P2J GUI and 4GL GUI

frame title active state difference for frames without focusable widgets

Added by Eugenie Lyzenko over 8 years ago. Updated over 7 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Hynek Cihlar
Target version:
FWD - Cleanup and Stabilization for GUI
Start date:
Due date:
% Done:

100%

billable:
No
vendor_id:
GCD
case_num:

frame_activation_rect2_test_4gl_step0.jpg - 4gl initial state (40.2 KB) Eugenie Lyzenko, 12/04/2015 08:39 AM

frame_activation_rect2_test_4gl_step1.jpg - 4gl after step 2 (44.8 KB) Eugenie Lyzenko, 12/04/2015 08:39 AM

frame_activation_rect2_test_4gl_step2.jpg - 4gl after step 3 (48.3 KB) Eugenie Lyzenko, 12/04/2015 08:39 AM

frame_activation_rect2_test_p2j_step2_diff_20151204.jpg - p2j after step 3 (45.1 KB) Eugenie Lyzenko, 12/04/2015 08:39 AM

History

#1 Updated by Eugenie Lyzenko over 8 years ago

This issue is happening when the frame has no focusable widgets inside and for certain conditions. The example - rectangle/rect_test2.p
1. Start the test, the frame title is the same as in 4gl.
2. Click on the frame title. The title become active in both 4gl and p2j
3. Press spacebar to continue testing. In 4gl the frame title is not active while in p2j it is active.

And this is the difference. See the pictures attached.

#2 Updated by Hynek Cihlar about 8 years ago

  • Assignee set to Hynek Cihlar

#3 Updated by Hynek Cihlar about 8 years ago

The issue is fixed in task branch 2918a revision 11005. The changes depend on 2823a, rebase is pending.

#4 Updated by Greg Shah about 8 years ago

Code Review Task Branch 2918a Revision 11005

I'm fine with the changes. Some minor comments:

1. In UiUtils, the locateFrame() and locateRootFrame() could use a common worker method by passing a predicate (lambda).

2. Please add history entries to TopLevelWindow, UiUtils and AbstractWidget.

#5 Updated by Hynek Cihlar about 8 years ago

2918a rebased against trunk revision 11007, the branch is now at revision 11010.

#6 Updated by Hynek Cihlar about 8 years ago

2918a revision 11008 resolves the issues from review. It passed GUI regression tests and ChUI regression testing in progress.

Please review.

#7 Updated by Greg Shah about 8 years ago

Code Review Task Branch 2918a Revision 11011

Please update the javadoc for UiUtils.getParentFrames() to explain that if the given widget is itself a frame, then it will be included in the resulting list.

Otherwise the changes are fine.

#8 Updated by Hynek Cihlar about 8 years ago

Greg Shah wrote:

Code Review Task Branch 2918a Revision 11011

Please update the javadoc for UiUtils.getParentFrames() to explain that if the given widget is itself a frame, then it will be included in the resulting list.

This is good point. I had unintentionally changed the semantics of locateFrame(), the method in its original implementation does not return the passed in frame widget. And I think it also makes sense for getParentFrames() not to include the passed in frame.

Also I fixed the original logic of locateFrame() that would not properly resolve the parent frame when DropDown is passed in.

Please review revision 11012. I am also restarting ChUI regression tests.

#9 Updated by Greg Shah about 8 years ago

Code Review Task Branch 2918a Revision 11012

I'm good with the changes.

#10 Updated by Greg Shah about 8 years ago

How is testing coming along?

I expect to be ready for 1811u to go into testing soon, but we are waiting on 2918a.

#11 Updated by Hynek Cihlar about 8 years ago

Greg Shah wrote:

How is testing coming along?

I expect to be ready for 1811u to go into testing soon, but we are waiting on 2918a.

Sorry, I forgot to report back. The test failed with some unexpected errors. I had to restart it.

#12 Updated by Hynek Cihlar about 8 years ago

ChUI regression testing finished again with many false negatives, restarted.

#13 Updated by Hynek Cihlar about 8 years ago

This time 2918a ChUI testing hang in Ctrl-C part. Testing restarted.

#14 Updated by Hynek Cihlar about 8 years ago

2918a rebased against trunk rev 11008. Also I fixed two regressions in the branch - an NPE in GUI alert-box and wrong focus state in ChUI.

The branch has passed Ctrl-C tests, main part is in progress.

#15 Updated by Hynek Cihlar about 8 years ago

2918a revision 11017 has passed regression testing. I am about to rebase the branch against latest trunk.

#16 Updated by Hynek Cihlar about 8 years ago

2918a rebased against trunk 11010 as revision 11019. There were no conflicting changes. GUI regression tests went fine, ChUI regression testing is in progress.

#17 Updated by Greg Shah about 8 years ago

Considering the low risk of regressions on the ChUI side, please go ahead and merge to trunk now.

#18 Updated by Hynek Cihlar about 8 years ago

2918a merged to trunk as revision 11011. The branch has been archived.

#19 Updated by Hynek Cihlar about 8 years ago

  • % Done changed from 0 to 100

#20 Updated by Greg Shah about 8 years ago

  • Start date deleted (12/04/2015)
  • Status changed from New to Closed
  • Target version set to Milestone 16

#21 Updated by Greg Shah over 7 years ago

  • Target version changed from Milestone 16 to Cleanup and Stabilization for GUI

Also available in: Atom PDF