discard the extended checksum part in row values
What problem does this PR solve?
Issue Number: ref https://github.com/pingcap/tidb/issues/42747
Problem Description:
Row values now may have extended checksums since pingcap/tidb#42859, so client-java need to discard these checksums on read.
What is changed and how does it work?
Do not copy the checksum part to RowV2.data. Actually the current implementation is ok since RowV2.data can be only accessed by RowV2.getData. This PR just makes it more safer.
The latest updates on TiDB Cloud Branch. Learn more about TiDB Cloud ↗︎
| BranchName | ClusterID | Status | Updated |
|---|---|---|---|
| discard-row-checksum_739_5b77567 | 3059033 | running | 2023-04-24T06:34:36Z |
Codecov Report
Patch coverage has no change and project coverage change: -0.18 :warning:
Comparison is base (
cb26d58) 37.93% compared to head (eecfdcf) 37.76%.
Additional details and impacted files
@@ Coverage Diff @@
## master #739 +/- ##
============================================
- Coverage 37.93% 37.76% -0.18%
+ Complexity 1613 1603 -10
============================================
Files 278 278
Lines 17517 17522 +5
Branches 1992 1994 +2
============================================
- Hits 6645 6617 -28
- Misses 10208 10237 +29
- Partials 664 668 +4
| Impacted Files | Coverage Δ | |
|---|---|---|
| src/main/java/org/tikv/common/codec/RowV2.java | 0.00% <0.00%> (ø) |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
roughly approve. But will client-java return row with checksums col before this optimize? checksum seems a feature of tidb and it does not exist in TiKV, which means it may not influence client-java. have you tested this pr locally?