pan3793 edited a comment on issue #1885: URL: https://github.com/apache/iceberg/issues/1885#issuecomment-741473849
@openinx Thanks for following up on this issue, I haven't test nested field, but found another issue about Logical Types. Our ETL build on CDH-6.3.1 with Avro 1.8.2, which not support generate Java 8 time because of AVRO-2079, and without `converter`, Flink can't handle joda time properly. > I think we may need a `converter` which just like the flink's `DataFormatConverters.RowConverter` to convert the avro `GenericRecord` to `RowData` ? We really need that `converter`. All of our ETL job input and output data structures are present by Avro `GenericRecord` because it has `avrc` to define schema in Json, and provide `avro-maven-plugin` to generate Entity automatically. PS: We only use avro to manage schemas, but store in ORC. And we are trying to migrate storage format to Iceberg, if iceberg provide schema management tools like avro, may be we can also manage schema by Iceberg. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
