Refactor Image(Display, int, int) in Tests (No Ops)

[ShahzaibIbrahim]

Replacing Image(Display, int, int) with Image(Display, ImageGcDrawer, int, int) in Tests. Basic Case

• requires -> Refactor Image(Display, int, int) in Tests (Automated Tests) #2229

[github-actions]

Test Results

   545 files  ±0     545 suites  ±0   32m 32s ⏱️ + 2m 44s
 4 405 tests ±0   4 387 ✅ ±0   18 💤 ±0  0 ❌ ±0 
16 741 runs  ±0  16 599 ✅ ±0  142 💤 ±0  0 ❌ ±0 

Results for commit 06fbe94. ± Comparison against base commit 3d93396.

♻️ This comment has been updated with latest results.

[HeikoKlare]

This PR overlaps with #2229. Please consider the comments I made there and please revise the PRs so that they do not change the same lines of code in the semantically same but syntactically different ways.

[fedejeanne]

@ShahzaibIbrahim is there anything else to do here from your side or have all comments been addressed? I see that some conversations are still open.

Also, the test org.eclipse.swt.tests.junit.Test_org_eclipse_swt_SWT.test_isLocal failed failed in 2 consecutive runs. Can you look into that? If it's a flaky test, please report it.

[ShahzaibIbrahim]

@ShahzaibIbrahim is there anything else to do here from your side or have all comments been addressed? I see that some conversations are still open.

Also, the test org.eclipse.swt.tests.junit.Test_org_eclipse_swt_SWT.test_isLocal failed failed in 2 consecutive runs. Can you look into that? If it's a flaky test, please report it.

All the comments were addressed but it needs to be re-reviewed.

The comments in #2229 have been addressed

[ShahzaibIbrahim]

@akurtakov the test test_isLocal fails for this PR only in the CI. Do you have any idea what is it about?

[fedejeanne]

@akurtakov the test test_isLocal fails for this PR only in the CI. Do you have any idea what is it about?

My educated guess is that this check fails because SWT was not built for Linux in this PR and the check basically says "you're not testing your changes, you're testing master right now".

But this is irrelevant since this PR doesn't change any production code that needs to be tested, it changes only the tests. So technically, it is OK to test master (more precisely: the latest official drop).

Am I seeing this right, @akurtakov ?

[akurtakov]

I am traveling so I can't look into things this week. Please remind me next week if there is still a problem to look into it.

[fedejeanne]

Thank you for fixing it, @akurtakov !

May I ask what was the trick? If you merely rebased on master then you must have magic hands because I tried that too and it didn't work... 😅

[akurtakov]

I have just rebased it :) . My gut feeling is that this error appears in a period when a commit makes the build produce different artifact compared to last I-build or smth similar and it starts working after the next build is in I-build repo. I haven't looked whether that's the case so I could be totally wrong too.

[fedejeanne]

Ok, but at least I have something to try the next time I see this check fail. Thank you for the hint!