deepakpanda93 commented on issue #14601: URL: https://github.com/apache/hudi/issues/14601#issuecomment-5115808617
This issue was reviewed as part of the JIRA-migrated backlog triage. **Findings:** this is addressed. Hudi no longer requires `hoodie.datasource.write.recordkey.field` - if you leave it unset, Hudi auto-generates a unique record key per row, so an `insert` of rows without a natural unique key behaves predictably (append semantics) instead of colliding. Auto record-key generation landed in commit [a34067826c2f](https://github.com/apache/hudi/commit/a34067826c2f) (HUDI-5514 / HUDI-5574 / HUDI-5604 / HUDI-5535, #8107) with follow-up [a60522a6fab0](https://github.com/apache/hudi/commit/a60522a6fab0) (#8697) improving the out-of-the-box defaults for append-only use cases. Both were first released in **0.14.0**. See `AutoRecordKeyGenerationUtils.scala` on `release-1.2.0`. On the related question in the thread - `insert` vs `upsert`: `upsert` looks up the record key in the index and merges into the existing record, while `insert` skips that index lookup and appends. With auto-generated keys there is no unintended merging. Closing as fixed. -- 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]
