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
I would like for a scala-steward run to fail if any of the updates fails (i.e., if any of the updates prints "Unable to bump version for update ...").
I have noticed that there are two cases where "Unable to bump version for update" occurs:
True positive: the version should have been updated, the log occurs, and in reality the version was not updated.
False positive: the version should have been updated, the log occurs, but actually the version was updated.
In the worst case, scala-steward always succeeds (exit code = 0), but never makes an update. So, I would rather be extra cautious and just fail the whole run if either of these occur.
For now, I have worked around this as follows:
tee the scala-steward logs to a file
grep the file for "Unable to bump version for update"
fail the Github Actions job if the grep returns non-empty
This workaround is good enough for me, for now. But I'm wondering if it would make sense to add this kind of strict, all-or-nothing mode as a command line argument?
The text was updated successfully, but these errors were encountered:
Are there any other cases where scala-steward can fail and still complete w/ a successful response? I feel like I remember seeing some others at times.
Maybe another way to solve this is to emit some sort of JSON digest, summarizing everything that scala-steward did during its run. Then the caller can parse this digest and apply their own definition of success or failure. Obviously not a small task, but just mentioning it as an idea.
I would like for a scala-steward run to fail if any of the updates fails (i.e., if any of the updates prints "Unable to bump version for update ...").
I have noticed that there are two cases where "Unable to bump version for update" occurs:
In the worst case, scala-steward always succeeds (exit code = 0), but never makes an update. So, I would rather be extra cautious and just fail the whole run if either of these occur.
For now, I have worked around this as follows:
tee
the scala-steward logs to a fileThis workaround is good enough for me, for now. But I'm wondering if it would make sense to add this kind of strict, all-or-nothing mode as a command line argument?
The text was updated successfully, but these errors were encountered: