Frequently we face situations when we change tox.ini in contrib (e.g., adding a new env or instrumentation or fixing a env name) we are required to manually update workflows in core as well (we need this to test core changes don't break contrib). Here are some PRs:
Maybe there's a better way to do that without the need to wait someone deal with a broken PR in core because tox -e generate-workflows is failing in CI.
Right now, the solution is: we need to manually fix workflows (i.e., run tox -e generate-workflows locally) and commit the changes (this is the process for keep workflows synced between -core and -contrib).
We need to have this explicit in documentation as well.
Frequently we face situations when we change tox.ini in contrib (e.g., adding a new env or instrumentation or fixing a env name) we are required to manually update workflows in core as well (we need this to test core changes don't break contrib). Here are some PRs:
https://github.com/open-telemetry/opentelemetry-python/pull/4258 https://github.com/open-telemetry/opentelemetry-python/pull/4256 https://github.com/open-telemetry/opentelemetry-python/pull/4239 https://github.com/open-telemetry/opentelemetry-python/pull/4192 https://github.com/open-telemetry/opentelemetry-python/pull/4158
Maybe there's a better way to do that without the need to wait someone deal with a broken PR in core because
tox -e generate-workflows
is failing in CI.Right now, the solution is: we need to manually fix workflows (i.e., run tox -e generate-workflows locally) and commit the changes (this is the process for keep workflows synced between -core and -contrib).