SEZ9 commented on issue #12649: URL: https://github.com/apache/seatunnel/issues/12649#issuecomment-6030367888
Thanks for the detailed report. The analysis makes sense: `seatunnelRowBatch` is cumulative, while `fieldVector` only holds the currently loaded batch, so indexing the vector with `readRowCount + i` and bounding the loop by `seatunnelRowBatch.size()` breaks for every non-empty batch after the first. The point about the loop-bound-only change turning the exception into silent nulls is a good catch. Since you're willing to submit a PR, please go ahead with the fix (loop over `rowCountInOneBatch`, index the vector from 0, keep `readRowCount + i` for the list). Please include a multi-batch test in `ArrowToSeatunnelRowReaderTest` that asserts the actual values, not just the row count, ideally covering the `ARRAY` and `TIMESTAMP` columns you mentioned. Since `DorisValueReader` and `StarRocksBeReadClient` both use this reader, please make sure the existing Doris and StarRocks tests still pass. <!-- streview-comment:1577 --> -- 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]
