Bug 1423 - Newt SWT GLWindow Fullscreen Wrong Coordinates in Embedded Layout
Summary: Newt SWT GLWindow Fullscreen Wrong Coordinates in Embedded Layout
Status: RESOLVED FIXED
Alias: None
Product: Jogl
Classification: JogAmp
Component: swt (show other bugs)
Version: 2.4.0
Hardware: All all
: P4 normal
Assignee: Sven Gothel
URL:
Depends on: 1422
Blocks: 674
  Show dependency tree / graph
 
Reported: 2020-01-15 13:58 CET by Marcel Au
Modified: 2023-01-24 09:02 CET (History)
1 user (show)

See Also:
Type: DEFECT
SCM Refs:
e03204605f95dc3586d8e3582250211954542be5
Workaround: ---

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 Marcel Au 2020-01-15 13:58:23 CET 
If I switch a GLWindow to fullscreen (GLWindow.setFullscreen()) the top left display starts at the 0,0 coordinates of the embedded view and not at the monitor 0,0 coordinates. 

I tried to correct this with the newTChild.setPosition method but this had no effect.
Comment 1 Sven Gothel 2020-01-16 02:01:31 CET 
Excellent, thank you.
Comment 2 Marcel Au 2020-01-16 12:08:43 CET 
Hello Sven,

with the latest built (1502) fullscreen works for Linux (as well as the correct layout on high dpi).

On Windows and Mac, however still the same top-left of the view.
Comment 3 Sven Gothel 2020-01-22 09:19:59 CET 
(In reply to Marcel Au from comment #2)
commit e03204605f95dc3586d8e3582250211954542be5

    Bug 1423: NEWT Child Window on OSX: Don't use parent derived location in fullscreen mode
    
    WindowDriver OSX: useParent(..) -> useParentLocation(..) and add !isFullscreen() as a criteria to use the parent derived location.
Comment 4 Sven Gothel 2020-01-22 09:22:34 CET 
(In reply to Marcel Au from comment #2)
On Windows I cannot validate the bug, i.e. fullscreen works for me.

I used the VM as described in Bug 1422 comment 15
Comment 5 Sven Gothel 2020-01-22 09:27:47 CET 
(In reply to Sven Gothel from comment #3)
This commit works for me on OSX as it resolved the issue.

I had no time to produce a new RC yet though.
Comment 6 Marcel Au 2020-01-22 12:53:46 CET 
(In reply to Sven Gothel from comment #5)

Hello Sven,

WindowDriver OSX: useParent(..) -> useParentLocation(..) and add !isFullscreen() as a criteria to use the parent derived location.

I' don't understand. How should I call the above? Or do you have already committed the solution?
Comment 7 Marcel Au 2020-01-22 14:22:19 CET 
Hello Sven,

fullscreen works for me on windows (highdpi) now if I set/unset/set fullscreen.

At the first fullscreen event the top-left view coordinates are used.

If I then set fullscreen to false and then again fullscreen the view the correct coordinates are used.
Comment 8 Marcel Au 2020-01-28 08:45:15 CET 
(In reply to Sven Gothel from comment #5)

Hello Sven could you produce a new RC. 

I would like to test the fullscreen feature on Mac. 

My latest test was with built 1502.
Comment 9 Marcel Au 2020-02-04 08:18:21 CET 
Works with built 1503 on MacOSX.
Comment 10 Sven Gothel 2020-02-04 14:28:38 CET 
(In reply to Marcel Au from comment #9)
so all platforms (MacOS, Linux and also Windows) working?

On Windows there should have never been an issue though .. hmm.
Comment 11 Marcel Au 2020-02-04 16:12:54 CET 
(In reply to Sven Gothel from comment #10)
As far as I've tested it, yes.

I don't know why I have to call the fullscreen action on Windows two times but it works anyway as a workaround without any time lag.
Comment 12 Sven Gothel 2023-01-24 09:02:34 CET 
see comment 11