github-actions[bot] commented on code in PR #65437:
URL: https://github.com/apache/doris/pull/65437#discussion_r3557222492
##########
be/src/exec/scan/file_scanner_v2.cpp:
##########
@@ -263,6 +268,16 @@ bool FileScannerV2::is_supported(const
TFileScanRangeParams& params, const TFile
}
}
+Status FileScannerV2::_validate_scan_range(const TFileScanRangeParams& params,
+ const TFileRangeDesc& range) {
Review Comment:
Remote batch sources still need a pre-routing escape hatch for table formats
that V2 cannot read. A Hive ACID scan can enter batch mode from
`HiveScanNode.isBatchMode()` when the partition threshold is reached, and each
fetched split is later marked `table_format_type=transactional_hive` in
`HiveScanNode.setScanParams()`. This selector now chooses FileScannerV2 before
those remote splits are fetched, but V2 neither lists `transactional_hive` as
supported nor has a `TransactionalHiveReader` factory branch; this validator
will therefore return `NotSupported` as soon as the first ACID split arrives.
The existing `FileScanner` path does handle `transactional_hive`, so large Hive
ACID queries that previously worked through remote batch can now fail by
default. Please keep transactional Hive on `FileScanner` unless/until V2
implements that reader, for example by carrying a scan-level support flag from
FE or disabling V2 for Hive ACID batch sources.
--
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]
