Caideyipi commented on PR #17871:
URL: https://github.com/apache/iotdb/pull/17871#issuecomment-6905706306

   Re-review against fixed base `cca1dbbde6d0dffa38495c92c92a48b2626d2f99` 
reviewed PR head `49816e970fe1f5df80f2b762e56b213703272005`. I found and fixed 
four behavioral issues: moving averages lost the finite mean when an 
intermediate window sum overflowed; spline coordinates could collapse dense 
timestamp clusters at both `long` endpoints; DTW matches could remain stale or 
depend on partial-state merge order, and `parseTime` rejected the valid 
`Long.MAX_VALUE` boundary. Regression tests cover overflow recovery, 
endpoint-dense spline input, non-monotonic/interleaved pattern input, and exact 
long bounds.
   
   Validation: clean offline source-reactor build of `library-udf` and its 
IoTDB dependency closure passed all 146 tests (0 failures, errors, or skips); 
Checkstyle, Spotless, and `git diff --check` passed. DTW retains effective 
input in O(N) memory and deferred rebuilding costs O(N * W^2); no performance 
benchmark or full-project build was run. CI was not polled.
   
   Reviewed SHA: `49816e970fe1f5df80f2b762e56b213703272005`. Final local and 
confirmed source-branch/PR-head SHA: 
`91e48ebf715a2890a2f3116e8066bfebe251f763`. The PR remains Draft.


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to