thswlsqls opened a new issue, #9861: URL: https://github.com/apache/paimon/issues/9861
**Search before asking** - [x] I searched in the [issues](https://github.com/apache/paimon/issues) and found nothing similar. **Paimon version** master @ bbae07d37 **Compute Engine** Any engine; the Java API reproduces it. **Minimal reproduce step** 1. Create a primary-key table with a column `r ROW<id INT, v VECTOR(3, FLOAT)>`. 2. Write one row with a non-null `r.v` and call `prepareCommit`. The flush fails with: ``` java.lang.IllegalArgumentException: Unsupported type: VectorType at org.apache.paimon.data.NestedRow.getVector(NestedRow.java:311) at org.apache.paimon.format.parquet.writer.ParquetRowDataWriter$ArrayWriter.write(ParquetRowDataWriter.java:523) at org.apache.paimon.format.parquet.writer.ParquetRowDataWriter$RowWriter.write(ParquetRowDataWriter.java:609) ``` The Avro writer fails the same way. The write buffer holds the row as `BinaryRow`, so the format writer reads the nested row through `BinaryRow.getRow()` → `NestedRow`, whose `getVector(int)` is still the stub from #7204. `BinaryWriter.writeVector` already writes the nested slot; only the read accessor is missing. **What doesn't meet your expectations?** A nested VECTOR should be written, committed and read back like a top-level VECTOR (`BinaryRow.getVector`) and `ARRAY<VECTOR>` (`BinaryArray.getVector`). #7204 added the stub to both `BinaryArray` and `NestedRow`; #9773 fixed `BinaryArray`; `NestedRow` is the last one. **Anything else?** N/A **Are you willing to submit a PR?** - [x] I'm willing to submit a PR! -- 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]
