zero323 commented on pull request #32122:
URL: https://github.com/apache/spark/pull/32122#issuecomment-817718455
Personally, I don't see a problemem with having this (with addressedw
current and possibly future comments) added separately.
Annotations are pretty large and not fixed thing - current state is a result
of a few years of maintenance and multiple passes and there still problems that
haven't been noticed so far. Not to mention there are parts that are morenmy
personal an opinion about style (i.e. list vs. tuple usage or balance between
false positives and false negatives), which are probably disputable if you come
from different background. There are also some decisions that we made during
initial port (some tests where omitted, some stubs where removed, we decided to
prefer ignores instead of adding new annotations in quite a few places).
We should probably avoid fixing individual, low priority (not bug, not
reported as an issue) annotations in separate PRs, to keep review work.
managable, but I don't think its a problem here.
Also, there are cases where annotations require a lot of caution, like ML or
UDFs.
I am bit preoccupied lately, but if you plan to work more on this part and
have any questions, please let me know.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
