Wait for refresh job at FileBufferFunctions.tearDown() by trancexpress · Pull Request #3825 · eclipse-platform/eclipse.platform.ui

trancexpress · 2026-03-27T13:13:50Z

FileBuffersForFilesInLinkedFolders.test16_3() fails infrequently during CI jobs, the fail indicates a missing resource delta notification.

RefreshJob is triggered during file buffer operations in test2 and test6, and can notify test listeners of deltas. To avoid this potential problem, this change adds a wait for the refresh family at: tearDown()

See: #2528

trancexpress · 2026-03-27T13:22:45Z

Normally the test listener is notified with this stack trace:

"main" #1 [169732] prio=5 os_prio=0 cpu=787.66ms elapsed=559.41s tid=0x00007f0af401bff0 nid=169732 at breakpoint [0x00007f0af91fa000]
   java.lang.Thread.State: RUNNABLE
        at org.eclipse.core.filebuffers.tests.FileBufferFunctions$14Listener.stateChanging(Unknown Source)
        at org.eclipse.core.internal.filebuffers.TextFileBufferManager$9.run(TextFileBufferManager.java:771)
        at org.eclipse.core.runtime.SafeRunner.run(SafeRunner.java:47)
        at org.eclipse.core.internal.filebuffers.TextFileBufferManager.fireStateChanging(TextFileBufferManager.java:768)
        at org.eclipse.core.internal.filebuffers.ResourceFileBuffer$SafeFileChange.preRun(ResourceFileBuffer.java:74)
        at org.eclipse.core.internal.filebuffers.ResourceFileBuffer$FileSynchronizer.resourceChanged(ResourceFileBuffer.java:183)
        at org.eclipse.core.internal.events.NotificationManager$1.run(NotificationManager.java:342)
        at org.eclipse.core.runtime.SafeRunner.run(SafeRunner.java:47)
        at org.eclipse.core.internal.events.NotificationManager.notify(NotificationManager.java:331)
        at org.eclipse.core.internal.events.NotificationManager.broadcastChanges(NotificationManager.java:185)
        at org.eclipse.core.internal.resources.Workspace.broadcastPostChange(Workspace.java:473)
        at org.eclipse.core.internal.resources.Workspace.endOperation(Workspace.java:1619)
        at org.eclipse.core.internal.resources.Resource.move(Resource.java:1644)
        at org.eclipse.core.internal.resources.Resource.move(Resource.java:1588)
        at org.eclipse.core.filebuffers.tests.FileBuffersForFilesInLinkedFolders.moveUnderlyingFile(Unknown Source)
        at org.eclipse.core.filebuffers.tests.FileBufferFunctions.test16_3(Unknown Source)

At ResourceFileBuffer.FileSynchronizer.resourceChanged(IResourceChangeEvent) there are some conditions which might prevent the notification, most notably at the start of the method:

	@Override
	public void resourceChanged(IResourceChangeEvent e) {
		IResourceDelta delta= e.getDelta();
		if (delta != null) {
			delta= delta.findMember(fFile.getFullPath());
		}

		if (delta != null && fIsInstalled) {

The listener should be notified on CHANGED and REMOVED deltas, depending on delta flags. Considering FileBuffersForFilesInLinkedFolders.moveUnderlyingFile() calls IFile.move(), I'm not sure what can be missing in the delta... but I don't have ideas other than logging and hopefully seeing something in the log.

fIsInstalled should be true as long as the buffer is connected... When running FileBuffersTestSuite there were no connect() or disconnect() calls in parallel, only from the main thread. So I'm not sure this is the cause; I've added a check for it just in case, but it seems unlikely.

...lipse.core.filebuffers.tests/src/org/eclipse/core/filebuffers/tests/FileBufferFunctions.java

github-actions · 2026-03-27T14:06:16Z

Test Results

852 files ±0 852 suites ±0 55m 6s ⏱️ -58s
7 884 tests ±0 7 641 ✅ ±0 243 💤 ±0 0 ❌ ±0
20 154 runs ±0 19 498 ✅ ±0 656 💤 ±0 0 ❌ ±0

Results for commit e6dfa25. ± Comparison against base commit 04a5221.

♻️ This comment has been updated with latest results.

trancexpress · 2026-03-27T14:16:18Z

Failed test is: #2622

trancexpress · 2026-03-27T15:29:44Z

Repeating the test case many times via @ClassTemplate does seem to reproduce a fail, in another test method:

org.opentest4j.AssertionFailedError: Wrong listener notifcation in org.eclipse.core.filebuffers.tests.FileBuffersForFilesInLinkedFolders:
listener.count: 0 (expected: 1)
listener.newLocation: null
newLocation: /project/folderA/MovedLinkedFile
buffers identical: false
	at org.eclipse.core.filebuffers.tests.FileBufferFunctions.test15(FileBufferFunctions.java:905)

The test code mentions: https://bugs.eclipse.org/bugs/show_bug.cgi?id=409810

Output from the logging:

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:18:38.914
!MESSAGE PASS in test14()

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:18:38.916
!MESSAGE [main] thread logging resource change event
Delta:
	kind: 4, flags: 0, resource: R/
	kind: 4, flags: 0, resource: P/project
	kind: 1, flags: 0, resource: F/project/folderA

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:18:38.916
!MESSAGE [Worker-0: Refreshing workspace] thread logging resource change event
Delta:
	kind: 4, flags: 0, resource: R/
	kind: 4, flags: 0, resource: P/project
	kind: 4, flags: 0, resource: F/project/LinkedFolder
	kind: 2, flags: 8192, resource: L/project/LinkedFolder/FileInLinkedFolder
	kind: 4, flags: 0, resource: F/project/folderA
	kind: 1, flags: 4096, resource: L/project/folderA/MovedLinkedFile

!ENTRY org.eclipse.core.filebuffers.tests 4 0 2026-03-27 16:18:38.921
!MESSAGE FAIL in test15()
!STACK 0
org.opentest4j.AssertionFailedError: Wrong listener notifcation in org.eclipse.core.filebuffers.tests.FileBuffersForFilesInLinkedFolders:
listener.count: 0 (expected: 1)
listener.newLocation: null
newLocation: /project/folderA/MovedLinkedFile
buffers identical: false
	at org.junit.jupiter.api.AssertionUtils.fail(AssertionUtils.java:38)
	at org.junit.jupiter.api.Assertions.fail(Assertions.java:138)
	at org.eclipse.core.filebuffers.tests.FileBufferFunctions.test15(FileBufferFunctions.java:905)

A run from Eclipse that passes has this logging:

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:26:29.556
!MESSAGE [main] thread logging resource change event
Delta:
	kind: 4, flags: 0, resource: R/
	kind: 4, flags: 0, resource: P/project
	kind: 1, flags: 0, resource: F/project/folderA

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:26:29.563
!MESSAGE [main] thread logging resource change event
Delta:
	kind: 4, flags: 0, resource: R/
	kind: 4, flags: 0, resource: P/project
	kind: 4, flags: 0, resource: F/project/LinkedFolder
	kind: 2, flags: 8192, resource: L/project/LinkedFolder/FileInLinkedFolder
	kind: 4, flags: 0, resource: F/project/folderA
	kind: 1, flags: 4096, resource: L/project/folderA/MovedLinkedFile

!ENTRY org.eclipse.core.filebuffers.tests 1 0 2026-03-27 16:26:29.579
!MESSAGE PASS in test15()

The delta is the same, but its sent by this thread in the fail case:

!MESSAGE [Worker-0: Refreshing workspace] thread logging resource change event

In the pass case its from the main thread:

!MESSAGE [main] thread logging resource change event

During 100 executions, I see the refresh job is triggered during two tests:

	at org.eclipse.core.internal.refresh.RefreshManager.refresh(RefreshManager.java:104)
	at org.eclipse.core.internal.localstore.FileSystemResourceManager.asyncRefresh(FileSystemResourceManager.java:205)
	at org.eclipse.core.internal.localstore.FileSystemResourceManager.isSynchronized(FileSystemResourceManager.java:839)
	at org.eclipse.core.internal.resources.Resource.isSynchronized(Resource.java:1533)
	at org.eclipse.core.internal.filebuffers.ResourceFileBuffer.isSynchronized(ResourceFileBuffer.java:465)
	at org.eclipse.core.filebuffers.tests.FileBufferFunctions.test2(FileBufferFunctions.java:156)

	at org.eclipse.core.internal.refresh.RefreshManager.refresh(RefreshManager.java:104)
	at org.eclipse.core.internal.localstore.FileSystemResourceManager.asyncRefresh(FileSystemResourceManager.java:205)
	at org.eclipse.core.internal.localstore.FileSystemResourceManager.isSynchronized(FileSystemResourceManager.java:839)
	at org.eclipse.core.internal.resources.Resource.isSynchronized(Resource.java:1533)
	at org.eclipse.core.internal.filebuffers.ResourceFileBuffer.revert(ResourceFileBuffer.java:337)
	at org.eclipse.core.filebuffers.tests.FileBufferFunctions.test6(FileBufferFunctions.java:249)

It seems to not be always triggered. I'm not sure if the added logging here actually helps with reproduction.

trancexpress · 2026-03-27T16:36:52Z

The Windows delete problem occurred:

Contains: Problems encountered while deleting files.
Contains: Could not delete: D:\a\eclipse.platform.ui\eclipse.platform.ui\tests\org.eclipse.ui.tests\target\work\data\TestProject\TestFolder\Foo.txt.
java.nio.file.FileSystemException: D:\a\eclipse.platform.ui\eclipse.platform.ui\tests\org.eclipse.ui.tests\target\work\data\TestProject\TestFolder\Foo.txt: The process cannot access the file because it is being used by another process

Then maybe a follow-up error:

Resource '/TestProject' already exists.
org.eclipse.core.internal.resources.ResourceException(/TestProject)[374]: java.lang.Exception: Resource '/TestProject' already exists.
	at org.eclipse.core.internal.resources.ResourceException.provideStackTrace(ResourceException.java:42)
	at org.eclipse.core.internal.resources.ResourceException.<init>(ResourceException.java:38)
	at org.eclipse.core.internal.resources.Resource.checkDoesNotExist(Resource.java:353)
	at org.eclipse.core.internal.resources.Resource.checkDoesNotExist(Resource.java:340)
	at org.eclipse.core.internal.resources.Project.assertCreateRequirements(Project.java:94)
	at org.eclipse.core.internal.resources.Project.create(Project.java:316)
	at org.eclipse.core.internal.resources.Project.create(Project.java:302)
	at org.eclipse.core.internal.resources.Project.create(Project.java:297)
	at org.eclipse.ui.tests.navigator.AbstractNavigatorTest.createTestProject(AbstractNavigatorTest.java:55)
	at org.eclipse.ui.tests.navigator.AbstractNavigatorTest.createTestFolder(AbstractNavigatorTest.java:62)
	at org.eclipse.ui.tests.navigator.AbstractNavigatorTest.createTestFile(AbstractNavigatorTest.java:70)
	at org.eclipse.ui.tests.navigator.NavigatorTest.testInitialPopulation(NavigatorTest.java:36)

I don't know this fail though:

Test that all matches have been added again ==> expected: <892> but was: <879>
org.opentest4j.AssertionFailedError: Test that all matches have been added again ==> expected: <892> but was: <879>
	at org.junit.jupiter.api.AssertionFailureBuilder.build(AssertionFailureBuilder.java:151)
	at org.junit.jupiter.api.AssertionFailureBuilder.buildAndThrow(AssertionFailureBuilder.java:132)
	at org.junit.jupiter.api.AssertEquals.failNotEqual(AssertEquals.java:197)
	at org.junit.jupiter.api.AssertEquals.assertEquals(AssertEquals.java:150)
	at org.junit.jupiter.api.Assertions.assertEquals(Assertions.java:563)
	at org.eclipse.search.tests.filesearch.SortingTest.testSorted(SortingTest.java:70)

trancexpress · 2026-03-27T17:27:26Z

Another fail due to resource deletion:

Contains: Could not delete 'D:\a\eclipse.platform.ui\eclipse.platform.ui\tests\org.eclipse.ui.tests\target\work\data\org.eclipse.ui.tests.dialogs.ResourceItemLabelTest_1774631750502\test.txt'.
org.eclipse.core.runtime.CoreException: Could not delete: D:\a\eclipse.platform.ui\eclipse.platform.ui\tests\org.eclipse.ui.tests\target\work\data\org.eclipse.ui.tests.dialogs.ResourceItemLabelTest_1774631750502\test.txt.

And this fail: #3581

CleanupAddon should ensure that partStack is not rendered anymore, as all childs have been removed ==> expected: <true> but was: <false>
org.opentest4j.AssertionFailedError: CleanupAddon should ensure that partStack is not rendered anymore, as all childs have been removed ==> expected: <true> but was: <false>
	at org.junit.jupiter.api.AssertionFailureBuilder.build(AssertionFailureBuilder.java:151)
	at org.junit.jupiter.api.AssertionFailureBuilder.buildAndThrow(AssertionFailureBuilder.java:132)
	at org.junit.jupiter.api.AssertTrue.failNotTrue(AssertTrue.java:63)
	at org.junit.jupiter.api.AssertTrue.assertTrue(AssertTrue.java:36)
	at org.junit.jupiter.api.Assertions.assertTrue(Assertions.java:214)
	at org.eclipse.e4.ui.tests.workbench.PartRenderingEngineTests.ensureCleanUpAddonCleansUp(PartRenderingEngineTests.java:2453)

FileBuffersForFilesInLinkedFolders.test16_3() fails infrequently during CI jobs, the fail indicates a missing resource delta notification. RefreshJob is triggered during file buffer operations in test2 and test6, and can notify test listeners of deltas. To avoid this potential problem, this change adds a wait for the refresh family at: tearDown() See: eclipse-platform#2528

trancexpress · 2026-03-27T19:13:53Z

Lets see if this helps, if not I guess we add resource delta logging as initially intended...

trancexpress force-pushed the gh2528 branch from 2614a37 to c91b641 Compare March 27, 2026 13:23

trancexpress commented Mar 27, 2026

View reviewed changes

...lipse.core.filebuffers.tests/src/org/eclipse/core/filebuffers/tests/FileBufferFunctions.java Outdated Show resolved Hide resolved

trancexpress marked this pull request as draft March 27, 2026 14:20

trancexpress force-pushed the gh2528 branch from c91b641 to 0f8e3b4 Compare March 27, 2026 15:37

trancexpress changed the title ~~Log resource deltas during FileBuffersForFilesInLinkedFolders~~ Wait for refresh job at FileBufferFunctions.tearDown() Mar 27, 2026

trancexpress force-pushed the gh2528 branch 2 times, most recently from d12a109 to cc02776 Compare March 27, 2026 15:45

trancexpress force-pushed the gh2528 branch from cc02776 to 7d79209 Compare March 27, 2026 16:40

trancexpress marked this pull request as ready for review March 27, 2026 17:58

trancexpress force-pushed the gh2528 branch from 7d79209 to e6dfa25 Compare March 27, 2026 18:25

trancexpress merged commit 157e7ed into eclipse-platform:master Mar 27, 2026
18 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Wait for refresh job at FileBufferFunctions.tearDown()#3825

Wait for refresh job at FileBufferFunctions.tearDown()#3825
trancexpress merged 1 commit intoeclipse-platform:masterfrom
trancexpress:gh2528

trancexpress commented Mar 27, 2026 •

edited

Loading

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Uh oh!

github-actions bot commented Mar 27, 2026 •

edited

Loading

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

trancexpress commented Mar 27, 2026 •

edited

Loading

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

trancexpress commented Mar 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Uh oh!

github-actions bot commented Mar 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Test Results

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

trancexpress commented Mar 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Uh oh!

trancexpress commented Mar 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

trancexpress commented Mar 27, 2026 •

edited

Loading

github-actions bot commented Mar 27, 2026 •

edited

Loading

trancexpress commented Mar 27, 2026 •

edited

Loading