morningman commented on PR #68713:
URL: https://github.com/apache/doris/pull/68713#issuecomment-5984818783
<!-- doris-repo-review:v1:begin -->
### Local pipeline review — ✅ PASS
```yaml
schema: doris-repo-review/v1
status: PASS
pr: apache/doris#68713
commit: 99e33c47bd19fe75e8477231f74e9dabfaf1c853
base: 0dcd2a31a7bb00ba9a817c2600f64e770efea54d
reviewed_at: 2026-10-05T05:56+08:00
reviewer: morningman
model: claude-fable-5-1
effort: max
findings: {blocker: 0, major: 0, minor: 3, nit: 3}
rounds: 2
converged: true
```
**Notes for maintainers**
- Coverage the review could not reach: CI at this head ran only the license,
clang-format, checkstyle, secrets, large-file and title checks - no BE UT, FE
UT or regression results exist (no `run buildall`). The new V2 reader case
`be/test/format_v2/jni/jni_table_reader_test.cpp:316-349`, the gate tests and
the fluss regression suite were verified by reading only; the text-only gates
(build hygiene, fe-connector imports and metadata funnel, a clang-format 16 dry
run, an offline Groovy parse) pass locally. A `run buildall` is the evidence
still missing before merge.
- Residual risk below Major, relevant to a backport: the V1 (legacy
`FileScanner`) reader leaves the gate only on a cancel, not on a satisfied
limit as `be/src/util/jni_scan_heap_gate.h:52-55` promises
(`be/src/format/jni/jni_reader.cpp:154-158`), and that V1 block has no test. On
a branch where `enable_file_scanner_v2` is not the default, this is the default
path for paimon JNI reads with the option on.
- Rolling upgrade needs nothing beyond the PR:
`TFileRangeDesc.jni_heap_bytes` is an optional field without a default and
`enable_jni_heap_admission` a forwarded session variable; every mixed-version
combination degrades to "no declaration", i.e. the pre-PR behaviour.
<sub>Reviewed locally with the `doris-repo-review` pipeline. Repository
policy may accept this receipt for the matching commit; it is not a human
Apache approval.</sub>
<!-- doris-repo-review:v1:end -->
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
