laskoviymishka opened a new issue, #1000: URL: https://github.com/apache/iceberg-go/issues/1000
Parent: #589 Depends on #998. When users call `Transaction.AddDataFiles` with parquet files written outside iceberg-go on a v3 table, those files have no `first_row_id` and the writer cannot retroactively fabricate one without rescanning every file. Pyiceberg's posture is to error out unless the caller explicitly sets `first_row_id` per file at registration time. Mirror that. Touches: `table/transaction.go` `AddDataFiles` (or wherever the `filesToDataFiles` validator lives) — branch on `formatVersion >= 3` and require a non-zero `first_row_id` per file. Error message should point users at supplying the row-id range explicitly so they understand the workflow. Tests: `AddDataFiles` without `first_row_id` on v3 returns a clear error; with `first_row_id` succeeds and metadata `next-row-id` advances correctly. v2 path unchanged. Spec: [Iceberg row lineage](https://iceberg.apache.org/spec/#row-lineage). -- 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]
