Add TEST_SCOPE_WSGI_SERVER_FIXTURE and use it in PLONE_ROBOT_TESTING …

[datakurre]

…layer (used by Volto Cypress tests

[mister-roboto]

@datakurre thanks for creating this Pull Request and helping to improve Plone!

TL;DR: Finish pushing changes, pass all other checks, then paste a comment:

@jenkins-plone-org please run jobs

To ensure that these changes do not break other parts of Plone, the Plone test suite matrix needs to pass, but it takes 30-60 min. Other CI checks are usually much faster and the Plone Jenkins resources are limited, so when done pushing changes and all other checks pass either start all Jenkins PR jobs yourself, or simply add the comment above in this PR to start all the jobs automatically.

Happy hacking!

[sneridagh]

I just tested it and locally seems to work well. Of course, I have no means to "slow down" the processing to match the GHA runners pace. I just restarted the failing test. I will try to test this in a PR.

[datakurre]

Good to know test passed on second run. I could not find clear reason for the possible regression. WSGI shutdown should not affect the db.

[sneridagh]

Four runs, and all passed. Certainly this made impact on the test run. I'd say let's merge the other PR and this one, release p.a.robotframework.

For the record: https://github.com/plone/volto/runs/6335283483?check_suite_focus=true

[mauritsvanrees]

@sneridagh I have merged the two branches. Can you check if it works now? I hope I did not misinterpret some things.

In your Volto PR, the DockerfileRF should use p.a.robotframework master. And you should use the new layer VOLTO_ROBOT_TESTING.

[sneridagh]

OK! I think Asko wanted to fiddle a bit more with this one, but don't worry, we can open another PR for continue fiddling with it. The changes seemed already giving good vibrations.

Could we have another alpha release so we can test it in Volto?

Thanks!!

[sneridagh]

Regarding the DockerfileRF, I will take care of it and cleaning up after we end with the tests.

[mauritsvanrees]

I have released plone.app.robotframework 2.0.0a4.

[sneridagh]

Thanks!

[datakurre]

@mauritsvanrees I am sorry, but this was not yet ready :(

I though "in progress" label would have messaged that.

The merged version introduced unnecessary logging, but maybe that's not so harmful.

[sneridagh]

@datakurre Not a big deal!

What about this:

• Remove the logging, release another alpha.
• We create another PR with the current state, update Volto's PR to match it, so @datakurre can continue doing tests
• I'm already another PR running with a4, let's see the outcome

[datakurre]

@sneridagh @mauritsvanrees Now that I saw what was merged, no reason to panic.

This pull was not complete, because it did not really fix the issue in Volto yet. But thanks to the other merge, this only introduced the new layer only into a new VOLTO_ROBOT_TESTING.

So, @sneridagh could now work to change Volto core to use plone.app.robotframework.testing.VOLTO_ROBOT_TESTING. It does not fix the issue, but should not any new issues either.

Becase PLONE_ROBOT_TESTING fixture was left untouched, merging this pull should have had no regression.

[datakurre]

Need to add one more comment just to praise all the work @mauritsvanrees does for Plone releases :rocket:

[sneridagh]

@datakurre I will! let me know if you create another PR in here, so I will update the PR we were working on as well.

[mauritsvanrees]

Sorry, I merged too soon apparently. I don't think there is any harm in the currently released version though. Happy to do another release when needed.