xiangfu0 opened a new pull request, #18325:
URL: https://github.com/apache/pinot/pull/18325
## Summary
- Normalize Parquet LIST `element` wrapper maps inside the Parquet native
and Parquet-Avro extraction paths.
- Keep the behavior out of `BaseRecordExtractor` so other file formats are
not affected.
- Add a regression test that writes a nested Parquet LIST schema and
verifies both Parquet readers return scalar arrays while preserving real
`element` struct fields.
## Why
Parquet list encodings can surface scalar arrays as repeated structs with a
single `element` field, causing Pinot rows to contain `[{"element":"abc"},
{"element":"xyz"}]` instead of `["abc", "xyz"]` for fields such as
`metadata.tags`.
## Testing
- `./mvnw spotless:apply -pl pinot-plugins/pinot-input-format/pinot-parquet`
- `./mvnw -pl pinot-plugins/pinot-input-format/pinot-parquet
-Dtest=ParquetListElementRecordReaderTest,ParquetRecordReaderTest test`
- `./mvnw checkstyle:check -pl
pinot-plugins/pinot-input-format/pinot-parquet`
- `./mvnw license:format -pl pinot-plugins/pinot-input-format/pinot-parquet`
- `./mvnw license:check -pl pinot-plugins/pinot-input-format/pinot-parquet`
Fixes #17420
Refs #17708
--
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]
