KAFKA-7658: Follow up to original PR#8027
Conversation
There was a problem hiding this comment.
Renaming this variable in this class and all its sub-classed to make it more clear what it is for
There was a problem hiding this comment.
I agree with this
"subTopologySourceNodes" is more clearly named
There was a problem hiding this comment.
Creating this MaterializedInternal was actually a bug as if the passed in "store name" was null, no store name would be generated, and we would fail with an NPE. It's fixed implicitly with the refactoring (this bug was exposed after I rewrote a test -- leave a comment there, too)
There was a problem hiding this comment.
oh! this is good
thank you,
It was my mistake
There was a problem hiding this comment.
Just some cleanup to get rid or warnings.
There was a problem hiding this comment.
The is nothing to be deprecated -- also the comment is confusing as maintainMs / segmentInterval are not use in this method
There was a problem hiding this comment.
Adding couple of new test to verify we don't accept nulls.
There was a problem hiding this comment.
This line was added in the original PR for no reason -- just removing it again.
There was a problem hiding this comment.
Making the test a little bit fancier.
There was a problem hiding this comment.
Changing this test, to also change the key type to make sure we can pass in serdes correctly.
There was a problem hiding this comment.
Overwrite key-serde to match type (this was exposing the "store name == null" bug).
There was a problem hiding this comment.
Due to the refactoring of the KStreamImple#toTable() overloads, we get different names -- but as nothing is released yet, there is backward compatibility concern.
There was a problem hiding this comment.
This check is redundant, because we know there is only one sub-topology (ie, copartitionGroup) from the topologyDescription check above.
There was a problem hiding this comment.
Just a code simplification -- no need to not use props similarly to the other tests.
There was a problem hiding this comment.
This is an actual fix in the test code.
There was a problem hiding this comment.
This looks better than what I did, go for it!
My original hotfix PR is just to unblock the JDK11 jenkins job.
|
LGTM Thank you for fixing the shortcomings. :) |
|
Java 11 timed out. Java8: We can retrigger a test after a review. |
There was a problem hiding this comment.
oh! this is good
thank you,
It was my mistake
There was a problem hiding this comment.
I agree with this
"subTopologySourceNodes" is more clearly named
|
Test results are gone. Retest this, please. |
|
Java 11: Java 8: Retest this please. |
|
Spotbug failure -- unclear why -- could not reproduce locally -- rebased and pushed a new version just in case. |
048a738 to
d953f70
Compare
|
Not sure why, but it seems the build was never scheduled correctly. Retest this please |
- add more unit tests - fix bug for auto-repartitioning - some code cleanup
d953f70 to
295398d
Compare
|
Rebased this PR to resolve merge conflicts. |
|
Java 11 passed. Java 8: Retest this please. |
|
Java 11 timed out. Java 8 passed. Hence, we got one green build each. Merging this now. |
Follow up to original PR #7985 for KIP-523 (adding `KStream#toTable()` operator) - improve JavaDocs - add more unit tests - fix bug for auto-repartitioning - some code cleanup Reviewers: High Lee <yello1109@daum.net>, John Roesler <john@confluent.io>
|
Merged to |
Follow up to original PR apache#7985 for KIP-523 (adding `KStream#toTable()` operator) - improve JavaDocs - add more unit tests - fix bug for auto-repartitioning - some code cleanup Reviewers: High Lee <yello1109@daum.net>, John Roesler <john@confluent.io>
Follow up to PR #7985. Call for review @vvcephei @highluck
For PR must be cherry-picked to
2.5branch.