search

pandas-dev / pandas

Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more
https://pandas.pydata.org
BSD 3-Clause "New" or "Revised" License
42.57k stars 17.56k forks source link

CI/TST: Better information about the tests in azure-pipelines #29214

Open datapythonista opened 4 years ago

datapythonista commented 4 years ago

Copying from https://github.com/astropy/astropy/pull/8445#issuecomment-546140223

If you install the newest version of pytest-azurepipelines (1.0.0rc1) https://pypi.org/project/pytest-azurepipelines/1.0.0rc1/ I've switched the output from Junit to Nunit XML as its better supported in Azure Pipelines. You can then run this Azure plugin https://marketplace.visualstudio.com/items?itemName=AnthonyShaw.vss-pytest and you get an extra tab in the UI with all the information from the Pytest run, including skip reasons, fail reasons, fixtures, versions, parameters.

I guess we want this.

alimcmaster1 commented 4 years ago

+1 for this - do we just have to wait for a pytest-azurepipelines release

mlline00 commented 4 years ago

I hooked up Azure Pipelines to my pandas repo and verified everything was working correctly. Then, I modified azure-pipelines.yml to just run the windows tests and committed the change. The windows CI tests ran. Next, I committed an update which set pytest-azurepipelines=v1.0.0rc1 in azure-windows-36.yaml and azure-windows-37.yaml. The windows CI tests ran a second time.

Information displayed on the pytest tab is identical between the two CI runs.

Based on the comment above, I expected to see some additional/richer information on the pytest tab. Maybe I'm missing something?

datapythonista commented 4 years ago

No idea, can you share a link to this tab please?

avinashpancham commented 4 years ago

Do we still need some work on this? I worked with pytest-azurepipelines on Azure DevOps before, but I do not know how to access the Pandas CI/CD pipeline. Can someone maybe point me in the right direction?