prashantwason commented on PR #20483: URL: https://github.com/apache/hudi/pull/20483#issuecomment-6639754578
@yihua I checked a Hive/MR read of these columns. It fails with the default file group reader, so I opened #20528. The cause is a little different from `LIST_COLUMN_TYPES`. `HiveHoodieReaderContext.setSchemas` builds the column types from the file schema, where the column is already the `member<N>` group, so Hive's parquet reader gets `struct<member0:...>` and reads the file fine. The failure comes afterwards: each row is rewritten to the required schema, where the column is still a `UNION`, and `HoodieArrayWritableSchemaUtils.rewriteRecordWithNewSchemaInternal` throws `IllegalArgumentException: should not be here?` on the first non-null union value. With `hoodie.file.group.reader.enabled=false` the same reads return correct `struct<member0:...>` rows, which match the type this PR syncs. -- 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]
