github-actions[bot] commented on code in PR #65446:
URL: https://github.com/apache/doris/pull/65446#discussion_r3558519440
##########
be/src/format_v2/table_reader.cpp:
##########
@@ -658,10 +658,14 @@ Status
TableReader::create_file_reader(std::unique_ptr<FileReader>* reader) {
const bool enable_mapping_timestamp_tz = _scan_params != nullptr &&
_scan_params->__isset.enable_mapping_timestamp_tz &&
_scan_params->enable_mapping_timestamp_tz;
+ const std::string hive_parquet_time_zone =
Review Comment:
The new scan parameter is only consumed by scanner v2. `FileScanOperatorX`
still selects the legacy `FileScanner` when `enable_file_scanner_v2=false` (a
user-visible session variable), in load mode, or if any range is not supported
by v2. That legacy Hive Parquet path constructs `HiveParquetReader` with only
`_state->timezone_obj()`, and the legacy INT96 converter still uses
`_convert_params->ctz`, so an HMS table with
`hive.parquet.time-zone=Asia/Shanghai` returns catalog-adjusted INT96 values
with scanner v2 but session-timezone values when the legacy scanner is
selected. Please either propagate this property through the legacy Hive Parquet
reader/converter too, or force/reject scanner-v2 fallback when this parameter
is present, and add coverage with `enable_file_scanner_v2=false`.
--
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]
