ci: Fix functional tests step for pull requests in Windows GHA job

[hebasto]

This functionality has been broken since the Windows runner image version 20240128.1.0.

Fixes #29534.

[DrahtBot]

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

Code Coverage

For detailed information about the code coverage, see the test coverage report.

Reviews

See the guideline for information on the review process.

Type	Reviewers
ACK	fanquake, m3dwards

If your review is incorrectly listed, please react with 👎 to this comment and the bot will ignore it on the next update.

[hebasto]

cc @maflcko

[fanquake]

I can ACK 57e6e22 this only based on the fact that in this PR, the native Windows functional tests run: https://github.com/bitcoin/bitcoin/actions/runs/8119259315/job/22194887783#step:27:72, and that the native Windows functional tests are not currently running on master: https://github.com/bitcoin/bitcoin/actions/runs/8131828989/job/22239779585#step:27:63.

However it's not at all clear from the PR description or linked issue what broke, why it did, is this an intermittent thing, is there an upstream issue etc

Feels like these images should be pinned if they are going to randomly break, and then our tests will silently stop running?

[hebasto]

I doubt whether image version pinning is available. Regarding the root of the issue, I suspect a PowerShell version update. However, I cannot confirm it at this moment.

…

-- Hennadii Stepanov

On Tue, 5 Mar 2024 at 17:03, fanquake ***@***.***> wrote: I can ACK 57e6e22 <57e6e22> this only based on the fact that in this PR, the native Windows functional tests run: https://github.com/bitcoin/bitcoin/actions/runs/8119259315/job/22194887783#step:27:72, and that the native Windows functional tests are not currently running on master: https://github.com/bitcoin/bitcoin/actions/runs/8131828989/job/22239779585#step:27:63 . However it's not at all clear from the PR description or linked issue what broke, why it did, is this an intermittent thing, is there an upstream issue etc Feels like these images should be pinned if they are going to randomly break, and then our tests will silently stop running? — Reply to this email directly, view it on GitHub <#29535 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH3PXPQMMFYFNOESMDHC5XLYWX3HZAVCNFSM6AAAAABECUBYRWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNZZGIZTMMBSHE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[m3dwards]

ACK 57e6e22 as a way to get the tests running again quickly.

As a follow up I think that it would be good to know it doesn't work on powershell and also why the test runner complaining that '' isn't a valid test isn't a failure.

[fanquake]

Looks like we could/should also change the test_runner exit condition for the no valid scripts case, from 0 to 1, so that it is an error. @m3dwards you might want to follow up with that?