nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Where do I view pipeline test reports?

Jason Sheedy

According to the current docs, maven test results are picked up automatically and test reports generated. I can see in the tear down section of my build pipeline that it is picking up the results (output below), but can't see the reports anywhere. Can someone please advise where to view test reports?

Finished scanning for test reports. 

Found 22 test report files.

All test reports aggregated into 44 test cases

Uploading test results

Finished uploading test results.

Accepted answers

All comments

Andrew Sheehan_nuhgh4

@Jason Sheedy I have found that you can only view the results of failed tests. When you have a pipeline with failing tests, you will then be able to select a dropdown that appears where the Logs header is and select Test Results. Alternatively, you can click on the x number of failures message on the unit testing step and that will take you to the test results as well.

Hope this helps!

Jason Sheedy

Thanks Andrew. It seems a like a bit of an oversight to not display passing tests. I can't think of any good reason why they wouldn't include it. Maybe that's only available in the Enterprise version.

Sheraz Sharif

I would also like to see reports where all tests passed.

Philip Hodder

We now display the successful tests too. But don't generate a full report. Is there anything specific you'd want to see in a larger report for a 100% tests passing?

Jason Sheedy

Thanks @Philip HodderLooks like my test reports are being captured in the build teardown, but I don't see anything in the UI to indicate failed or passing tests.

<span>All test reports aggregated into 44 test cases<br></span><span>Uploading test results<br></span><span>Finished uploading test results.</span><span>Found 22 test report files.<br></span>

When you say "should be showing on the step", is that displayed in the log output of the individual build or the main pipelines screen that shows all builds?

Jason Sheedy

Correction. I do see failing tests, but it's not displayed in the same way as the screenshot you posted.

Philip Hodder

@Jason Sheedy

Sorry, I just showed a subset of the page. Here's the whole result:

Screen Shot 2018-01-16 at 10.36.54 am.png The same view as you're looking at.

Thanks,

Phil

Jason Sheedy

Thanks Phil. I'm not seeing the same thing. Could there be some difference because it's an academic team account? (See below)

Screenshot from 2018-01-16 12-45-08.png

Fabrice Veniard

Thks for the quick answer and sorry for misplacing my original reply. I misunderstood the interface ^^

Anyway, failing tests failing the pipeline works like it's supposed to.

In case of full success, I would see a green bar in the log output and/or interface on the left (equivalent to the big red one on failure)

Philip Hodder

No problem. I'll pass on your feedback to the rest of the team.

Fabrice Veniard

Thks !

Also, I don't have the test details on the left, close to the link to see the build configuration as per your screenshots.

Maybe I could have these in there too when they succeed ?

Anyway, thks for the follow up

Ralf Hauptmann

Hey guys,

after watching your webinar last week I started using Pipelines on our PHP API.

I was reading this thread carefully but still cant see a way to access the error reports created see:

I know there is a way upload this to S3 but I cant find the documentation anymore.
Wouldn't it be the easiest way to attach this report to the email you send anyway on failure:

Hope you can help me and others here.
I already checked: https://confluence.atlassian.com/bitbucket/test-reporting-in-pipelines-939708543

Cheers from Fishburners Sydney

unknown

I don't know If I 'm missing sth here but phpunit can fail on deprecation notices(if not configured otherwise) but the tests will pass. As a result you got a failed pipeline with successful tests. Well, I don't know if it is an uncaught use case by your side, maybe it is worth mentioning it.

jackiecustodio

Hi guys.. from https://confluence.atlassian.com/bitbucket/test-reporting-in-pipelines-939708543.html, says pipelines only support XML xUnit-compatible reports... what if i have an ExtentReport that spits out a TestAutomationReport.html at the end of every test run (and/or attaches a screenshot when a bdd step fails)... how do I integrate this in Bitbucket pipelines and actually have a Test Report tab or whatever that business users can jump into to see my colorful dashboard ??

unknown

Hey Jackie,

I believe the reports should be strictly in XML format.

Eduardo Rosales Fernandez_aktkyd

Hi guys, I have this step in gitlab:

gradle-test:
    stage: test
script:
        - ./gradlew test -PnodeInstall --no-daemon
artifacts:
        reports:
            junit: build/test-results/test/TEST-*.xml
paths:
            - build/test-results/
            - build/jacoco/
expire_in: 1 day

how I transform it to bitbucket pipeline???

Philip Hodder

Hi Eduardo,

Bitbucket Pipelines automatically scans for test reports after each step of a pipeline.

So if you have test reports output in your build, then it should be parsed and shown in the UI of the build result.

We got some details on how test reports works in Pipelines here: https://confluence.atlassian.com/bitbucket/test-reporting-in-pipelines-939708543.html

Let me know how you go with that.

Thanks,

Phil

Jeannie wang

How bitbucket can show the red banner saying "failed" if the result reports errors?

My case shows that the build can still be deployed even though some tests are failed

Philip Hodder

@Jeannie wang the state of your step (in this case successful), is based on the exit code of your script. In this case, the command you're using to run your tests is still returning a successful exit code, even though you have a failing test.

Bitbucket Pipelines only looks for the presence of failed tests in the xUnit output of your tests.

You'll need to investigate why you test command is returning itself as successful. Perhaps you have a flag you need to enable/disable to avoid ignoring certain test results?