Test CTabItem rendering

[basilevs]

Pixel color verification for CTabItem's tabs.

A regression test for #3260 where tabs fail to render if
NSView.setClipsToBounds() is omited before scheduling redraw.

This test fails to reproduce the problem, but is useful as rendering is
not tested in this suite yet.

[merks]

It seems odd to use an all caps name for something that’s not a constant.

[basilevs]

Colors in SWT depend on Display, which I access via Shell, which is initialized after test starts. I could try to use Display.getCurrent() but this would be riskier (Display can be recreated). Therefore these are constants effectively but not in practice. Should I make them lowercase?

[merks]

Actually there are now Color constructors without a device and colors need not be disposed so they don’t really depend on a display/device.

When I see an all caps name, I expect it to static final, so yes, I’d prefer the name to match general expectations…

[basilevs]

I've made constants final using new constructors.

[github-actions]

Test Results

  177 files  +   26    177 suites  +26   29m 8s ⏱️ + 9m 13s
4 676 tests  -    53  4 652 ✅  -  54   20 💤  -  3   4 ❌ + 4 
6 743 runs  +1 020  6 571 ✅ +970  156 💤 +34  16 ❌ +16 

For more details on these failures, see this check.

Results for commit 6789053. ± Comparison against base commit 9ab3898.

This pull request removes 57 and adds 4 tests. Note that renamed tests count towards both.

org.eclipse.swt.graphics.ImageWin32Tests ‑ testDisposeDrawnImageBeforeRequestingTargetForOtherZoom
org.eclipse.swt.graphics.ImageWin32Tests ‑ testDrawImageAtDifferentZooms(boolean)[1] true
org.eclipse.swt.graphics.ImageWin32Tests ‑ testDrawImageAtDifferentZooms(boolean)[2] false
org.eclipse.swt.graphics.ImageWin32Tests ‑ testImageDataForDifferentFractionalZoomsShouldBeDifferent
org.eclipse.swt.graphics.ImageWin32Tests ‑ testImageShouldHaveDimesionAsPerZoomLevel
org.eclipse.swt.graphics.ImageWin32Tests ‑ testRetrieveImageDataAtDifferentZooms(boolean)[1] true
org.eclipse.swt.graphics.ImageWin32Tests ‑ testRetrieveImageDataAtDifferentZooms(boolean)[2] false
org.eclipse.swt.graphics.ImageWin32Tests ‑ test_getImageData_fromCopiedImage
org.eclipse.swt.graphics.ImageWin32Tests ‑ test_getImageData_fromImageForImageDataFromImage
org.eclipse.swt.tests.win32.Test_org_eclipse_swt_dnd_DND ‑ testByteArrayTransfer
…

org.eclipse.swt.tests.junit.Test_org_eclipse_swt_custom_CTabFolder ‑ test_tabsAreRendered(boolean, int)[1]
org.eclipse.swt.tests.junit.Test_org_eclipse_swt_custom_CTabFolder ‑ test_tabsAreRendered(boolean, int)[2]
org.eclipse.swt.tests.junit.Test_org_eclipse_swt_custom_CTabFolder ‑ test_tabsAreRendered(boolean, int)[3]
org.eclipse.swt.tests.junit.Test_org_eclipse_swt_custom_CTabFolder ‑ test_tabsAreRendered(boolean, int)[4]

♻️ This comment has been updated with latest results.

[basilevs]

Windows CI is passing, Linux and MacOS are not. The test is passing on my MacOS. Configuration issue? The test seems trivial - it should work, as long as graphical context is available.

How do I get screenshots? They are saved to /tmp and are not archived.

[akurtakov]

Would you please rebase on master ?

[basilevs]

Tests started to fail on Windows CI. They passed there on a previous rebase. Screenshots prove tabs are not rendered in 100ms. What am I doing wrong?

Now:
Gtk3 - pass
Gtk4 - fail
Macos aarch64 - pass
Macos x86_64 - fail
Windows - fail

[basilevs]

@HeikoKlare Windows regression might be related to recent pushes of GC redundancy fixes. Could you please run these tests with and without the changes?