You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When upgrading to 0.3.0 we noticed that the filenames of the failing tests were no longer correctly reported. This did not happen in 0.2.0 for us. However, also in 0.2.0 we saw incorrect line numbers for the failing tests.
In 0.3.0, the filename and line reported instead points to code within a hook (@pytest.hookimpl(tryfirst=True)) where we wrap pyfuncitem.obj in a lambda.
The text was updated successfully, but these errors were encountered:
@ke-kialo can you share a MRE? It's definitely tricky to track the code when there's plugins and hooks, and we don't have a test case which uses something like that.
When upgrading to 0.3.0 we noticed that the filenames of the failing tests were no longer correctly reported. This did not happen in 0.2.0 for us. However, also in 0.2.0 we saw incorrect line numbers for the failing tests.
In 0.3.0, the filename and line reported instead points to code within a hook (
@pytest.hookimpl(tryfirst=True)
) where we wrappyfuncitem.obj
in a lambda.The text was updated successfully, but these errors were encountered: