Brett,
Trust me, I don't enjoy this discussion no more that you, but I have to
respond.
Brett Porter wrote:
I'm sorry you lost some time investigating it, but I made every
attempt to do this properly.
At the time I made the change, I cleaned out the checkout and did a
build without getting any test errors. This was confirmed by the grid:
https://grid.sonatype.org/ci/view/Mercury/job/mercury-ant/6/
It appears your change to test using compilation was only checked in
afterwards. I'm afraid I'm still working on my ability to predict the
future :)
This statement suggests that I am a dumb coder, who submits tons of jars
to SVN for the pleasure of just having them there. I admit that I did
not commit the tests using those jars right away.
But give me some credit: everything has a reason. And if this reason is
not clear - ask, don't assume you know everything and can improve
without knowing. I did acknowledge your suggestion about the size of
test repo, and started fixing it. If you would have just suggested the
solution, provided a script in jira - that would only raise a lot of
gratitude.
But hindering a pre-alpha quality project by assuming things and
changing still unstable data, this is simply not fair. Losing a day over
such a trivial matter - I simply did not expect anyone to do such a thing.
I apologize if this sounds harsh, but believe me - the sole purpose is
to improve our process, make sure that this does not happen in the future.
So the proposal is: change the rules to say the following: "if you don't
work on an actively developed project - don't start modifying it without
consulting the team, working on it. If you do find a bug or improvement
- communicate with developers via issue tracking system and other means"
This is not predicting the future - just common sense.
Thanks,
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
