Deflake PreCommit Java watchForNewFiles#38320
Conversation
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request aims to improve the stability of the test suite by refactoring a flaky test case in FileIOTest. Additionally, it includes minor build configuration improvements for the DirectRunner and removes deprecated compatibility logic in the Python SDK to streamline the codebase. Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. Footnotes
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates Gradle task registrations to use the configuration avoidance API and refactors Java tests in FileIOTest.java to separate boundedness checks. Feedback highlights a critical regression in the Python SDK where version-specific logic was accidentally deleted and identifies that the new Java test will fail without disabling abandoned node enforcement since it lacks a run() call.
Abacn
changed the title
|
Checks are failing. Will not request review until checks are succeeding. If you'd like to override that behavior, comment |
|
Need a windows machine to debug the failing test (moved from needsRunnerTest to core unit test making it gets exercised on windows test workflow) |
There was a problem hiding this comment.
Code Review
This pull request migrates Gradle task declarations to the tasks.register API for better configuration efficiency and refactors FileIOTest.java to separate boundedness checks into a new test method. However, the new testMatchWatchForNewFiles_UnboundedPCollection method introduces a compilation error by attempting to concatenate a Path object with a String and will trigger a test failure unless abandoned node enforcement is disabled for the unexecuted TestPipeline.
| public void testMatchWatchForNewFiles_UnboundedPCollection() { | ||
| // Additional scenarios for testMatchWatchForNewFiles. Only construct the pipeline and check | ||
| // output pcoll | ||
| final Path watchPath = tmpFolder.getRoot().toPath().resolve("watch"); | ||
|
|
||
| PCollection<MatchResult.Metadata> matchAllMetadata = | ||
| p.apply("create for matchAll new files", Create.of(watchPath + "/*")) | ||
| .apply( | ||
| "match filename through matchAll", | ||
| FileIO.matchAll() | ||
| .continuously( | ||
| Duration.millis(100), | ||
| Watch.Growth.afterTimeSinceNewOutput(Duration.standardSeconds(1)))); | ||
|
|
||
| PCollection<MatchResult.Metadata> matchUpdatedMetadata = | ||
| p.apply( | ||
| "match updated", | ||
| FileIO.match() | ||
| .filepattern(watchPath.resolve("first").toString()) | ||
| .continuously( | ||
| Duration.millis(100), | ||
| Watch.Growth.afterTimeSinceNewOutput(Duration.standardSeconds(1)), | ||
| true)); | ||
|
|
||
| assertEquals(PCollection.IsBounded.UNBOUNDED, matchAllMetadata.isBounded()); | ||
| assertEquals(PCollection.IsBounded.UNBOUNDED, matchUpdatedMetadata.isBounded()); | ||
| } |
There was a problem hiding this comment.
This new test method has two significant issues:
- Compilation Error: At line 359,
watchPath + "/*"will not compile becausewatchPathis ajava.nio.file.Pathobject and Java does not support the+operator forPathandString. It should usewatchPath.resolve("*").toString()to be correct and consistent with the rest of the file. - Test Failure (Abandoned Nodes): The test uses the
TestPipelinerulepbut does not callp.run(). By default,TestPipelinethrows an exception if nodes are added but the pipeline is not executed. Since this test only checks the boundedness of the PCollections, you should disable the abandoned node enforcement usingp.enableAbandonedNodeEnforcement(false).
Here is the corrected method:
public void testMatchWatchForNewFiles_UnboundedPCollection() {
// Additional scenarios for testMatchWatchForNewFiles. Only construct the pipeline and check
// output pcoll
p.enableAbandonedNodeEnforcement(false);
final Path watchPath = tmpFolder.getRoot().toPath().resolve("watch");
PCollection<MatchResult.Metadata> matchAllMetadata =
p.apply("create for matchAll new files", Create.of(watchPath.resolve("*").toString()))
.apply(
"match filename through matchAll",
FileIO.matchAll()
.continuously(
Duration.millis(100),
Watch.Growth.afterTimeSinceNewOutput(Duration.standardSeconds(1))));
PCollection<MatchResult.Metadata> matchUpdatedMetadata =
p.apply(
"match updated",
FileIO.match()
.filepattern(watchPath.resolve("first").toString())
.continuously(
Duration.millis(100),
Watch.Growth.afterTimeSinceNewOutput(Duration.standardSeconds(1)),
true));
assertEquals(PCollection.IsBounded.UNBOUNDED, matchAllMetadata.isBounded());
assertEquals(PCollection.IsBounded.UNBOUNDED, matchUpdatedMetadata.isBounded());
}
Abacn
changed the title
|
still flaky |
1 participant
FileIOTest.testMatchWatchForNewFiles is very flaky, probably due to a heavy pipeline test 4 scenarios put into a single test. Consider split it into two test, for matchMetadata and matchAllUpdatedMetadata, construct and run pipeline; for matchAllMetadata and matchUpdatedMetadata, only construct pipeline and check the output PCollection is unbounded.
Previously it was testing full combination of two parameters. Picking two out of four covering all parameters the test coverage remains unchanged.
Please add a meaningful description for your change here
Thank you for your contribution! Follow this checklist to help us incorporate your contribution quickly and easily:
addresses #123), if applicable. This will automatically add a link to the pull request in the issue. If you would like the issue to automatically close on merging the pull request, commentfixes #<ISSUE NUMBER>instead.CHANGES.mdwith noteworthy changes.See the Contributor Guide for more tips on how to make review process smoother.
To check the build health, please visit https://github.com/apache/beam/blob/master/.test-infra/BUILD_STATUS.md
GitHub Actions Tests Status (on master branch)
See CI.md for more information about GitHub Actions CI or the workflows README to see a list of phrases to trigger workflows.