PyPy fixes

[isuruf]

See commit messages for details.

[msarahan]

Needs a test for the py value change (additional split) and for the noarch change. Please remove the pyXY selector changes.

[isuruf]

For the noarch change, I have no idea how to add a test. It would require a python package with a run_exports section. Any ideas?

[msarahan]

You'll probably need to make a simple recipe. You can build it as part of the test. There's a few examples of that, especially around run_exports. They make the tests slower, but it seems like the most stable, reliable way for tests to stay relevant over time (as opposed to building a package now and putting it somewhere).

[msarahan]

LGTM. Thanks for the revisions.

[isuruf]

One test fails that might be related,
https://dev.azure.com/anaconda-conda/ccafa57b-d4b4-4525-b7e5-575388d7c030/_apis/build/builds/235/logs/115
with,

2020-02-06T20:28:07.5154326Z Traceback (most recent call last):
2020-02-06T20:28:07.5154675Z   File "/tmp/cb/popen-gw0/test_noarch_python_with_tests0/noarch_python_test_package_1581019389295/test_tmp/run_test.py", line 17, in <module>
2020-02-06T20:28:07.5154737Z     assert os.path.isdir(site_packages)
2020-02-06T20:28:07.5154780Z AssertionError

@msarahan, have you seen this before? or is this a new failure?

[mingwandroid]

Do we not really want to use a different run_export type here? Dropping all but the package name?

[isuruf]

@mingwandroid, I don't understand. Are you suggesting a different run_exports scheme ('noarch' in addition to 'weak', 'strong') that would be applied if the package is a dependency of a noarch package?

[mingwandroid]

I am.

[isuruf]

@mingwandroid, done.

[msarahan]

@mingwandroid I'm not sure where you were going with this, but please go ahead with your thoughts. I'm fine with this PR now with or without the new run_exports type.

[mingwandroid]

Great. This is exactly what I meant.

[isuruf]

Tests pass now (except for unrelated test failures)

[mingwandroid]

Does anyone have any clues about https://dev.azure.com/anaconda-conda/conda-build/_build/results?buildId=239&view=logs&j=134db9d0-24f8-5cb4-e659-1d272ab2ea38&t=1bd5db4d-dc45-5f16-3e1b-4d645fd2ab7c&l=1548

I don't think it's related but I don't see any important differences there myself.

[marcelotrevisani]

Does anyone have any clues about https://dev.azure.com/anaconda-conda/conda-build/_build/results?buildId=239&view=logs&j=134db9d0-24f8-5cb4-e659-1d272ab2ea38&t=1bd5db4d-dc45-5f16-3e1b-4d645fd2ab7c&l=1548

I don't think it's related but I don't see any important differences there myself.

It seems that one of the packages which the CRAN tests are using was archived. And because of that, you need more options to download it
We might mock those access to the CRAN repo with the current data which they return. After all, those tests are not testing the connection with the cran but the parser itself.

[isuruf]

Yes, those 2 tests are failing on master too and seems to be an issue with r-rsed being archived.

[isuruf]

ping

[isuruf]

Thanks @marcelotrevisani. Is there a plan for a release?

[marcelotrevisani]

Thanks @marcelotrevisani. Is there a plan for a release?

Not that I am aware of... :/
Mike usually generated the releases, I can see with him how to generate it then

[github-actions]

Hi there, thank you for your contribution!

This pull request has been automatically locked because it has not had recent activity after being closed.

Please open a new issue or pull request if needed.

Thanks!