Julian>actions over the last few weeks have left me angry, depressed, and
burned out with the project

That is sad. Please accept my apologies for hurting your feelings.

Julian>Of course, a lot of the things he is removing are things that I
personally have created, so I naturally take this more personally than most
people

I did not even think about targeting someone specific.
The tools like Maven and Checkstyle were great 10 years ago, however, the
tools are lagging behind up to the point they severely impact
developer experience.

I can elaborate, but take "the order of the imports" issue for example.
Checkstyle produces errors, however, they are really hard to understand,
especially for new contributors.
On top of that, it turned out, Checkstyle accepted slight variations of the
import order, thus our codebase was using different imports at the same
time.

I replaced Checkstyle with Spotless, so the import order violations are now
printed in a human-readable format, and imports can be adjusted
automatically.
On top of that, I added .editorconfig, so the order of the imports is
automatically configured in IDEA.

Does that mean I removed Checkstyle configuration for import order? It does.
Did I remove Julian's code? Probably I did. I did not investigate who
created the import order configuration before.

Julian>We should go back to being solely Java.

Does that imply we remove Quidem, Cassandra, Druid, ElasitcSearch, Geode,
Kafka, MongoDB, Pig, Piglet, Spark and Splunk languages as well?
All of the above are non-Java languages, so if you suggest being "solely
Java", that should imply we treat the above as languages as well.

Julian>The general reaction was against the idea[1].

The general reaction was for "tests with $" and for "converting all the
things".
$ is important in that discussion.

Julian>After receiving this feedback, he went ahead anyway[2].

Julian, I'm afraid you are overreacting. [1] and [2] are not related.

Did I convert the test that was discussed in [1]?
Of course, I did not.

Did I convert all the uses of $ to something else?
Of course, I did not.

Did I convert ANY test with $?
Of course, I did not.

Did I convert a test in [2]?
Yes, I did. The converted test has absolutely nothing complicated, and its
readability much improved.

Note: I had received positive feedback on similar changes:
https://lists.apache.org/thread.html/5510cb6a825452d2b79600d4293c0ce487932ed51b1e22a3003452f7%40%3Ccommits.calcite.apache.org%3E
so

The old Java source in [2] missed \n characters in certain lines:
https://github.com/apache/calcite/commit/b93ec5a9edb7459696385e6adad67b92b6d406d7#diff-97678cea1fb54ed4236764d6a492e386L151-L152
It was absolutely not visible in the Java sources, and now the formatting
is much easier to follow.

Vladimir

Reply via email to