[Improvement][LocalOrder] Add tests about keeping consistent with FixedSize when no skew optimization #336
Conversation
…edSize when no skew optimization
|
PTAL @jerqi |
| // If ShuffleServer is flushing the file at this time, the length in the index file record may be greater | ||
| // than the length in the actual data file, and it needs to be returned at this time to avoid EOFException | ||
| if (dataFileLen != -1 && totalLen >= dataFileLen) { | ||
| if (dataFileLen != -1 && totalLen > dataFileLen) { |
There was a problem hiding this comment.
Yes. we need. This has been discussed in #320
Codecov Report
@@ Coverage Diff @@
## master #336 +/- ##
============================================
+ Coverage 61.31% 61.37% +0.06%
- Complexity 1526 1529 +3
============================================
Files 186 186
Lines 9438 9445 +7
Branches 924 924
============================================
+ Hits 5787 5797 +10
+ Misses 3341 3340 -1
+ Partials 310 308 -2
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
common/src/main/java/org/apache/uniffle/common/segment/LocalOrderSegmentSplitter.java
Outdated
Show resolved
Hide resolved
|
There are conflicts between LOCAL_ORDER and getInMemoryShuffleData. |
You are right. I will fix it by creating a reference copy to avoid breaking down original sequence. |
Fixed #336 |
3 participants
What changes were proposed in this pull request?
Why are the changes needed?
More stability
Does this PR introduce any user-facing change?
No
How was this patch tested?
UTs