search

eclipse-m2e / m2e-core

Eclipse Public License 2.0
113 stars 115 forks source link

Merge https://github.com/tesla/m2e-core-tests into this repo #531

Open kwin opened 2 years ago

kwin commented 2 years ago

Providing PRs which need adjusted tests is a hassle like this and probably the original reason for separating the tests from m2e core (licensing issues) are resolved meanwhile, or you even state that several modules in this repo have another license.

HannesWell commented 2 years ago

This would definitely be more convenient for everybody who contributes to m2e and this issue was already raised for a long time, see https://github.com/tesla/m2e-core-tests/issues/64 and Bug 541525 mentioned there. The problem seems not to be the licensing itself but the IP reviews.

Nevertheless I would appreciate if this would happen, but from what I have read in those issues I'm not sure if those that have to do the work are willing to do it.

Maybe @mickaelistria can say more about this?

mickaelistria commented 2 years ago

Providing PRs which need adjusted tests is a hassle

FWIW, it's not a mandatory step. Build (Jenkinsfile) is configured to always fetch m2e-core-tests master independently of what's the commit linked.

the original reason for separating the tests from m2e core (licensing issues) are resolved meanwhile

It would require some investigation as there is really a lot of content in this repo, and I'm not sure we can track all contributors to ensure they agree to submit it under Eclipse Contributor Agreement. That's far from being trivial and guaranteed to complete well.

I personally don't plan to invest in it more than I already did in the past because I'm certain it's a lot of work, but really unsure that success is at the end; too much uncertainity.

HannesWell commented 2 years ago

It indeed looks difficult to find out who has contributed to this repo. When you run git shortlog -sne --group=committer --group=author on the m2e-core-tests submodule you get entries like the following.

533  eu <eu@42f58313-09e3-4aa2-bb18-3b3e6a6bc9e8>
  1  unknown <matthew@.(none)>

One probably needs some inside knowledge of the company (I assume Sonatype?) at the time the contributions happen. There are other entries that are obvious but there are also many unclear ones.