SEZ9 commented on issue #13223:
URL: https://github.com/apache/seatunnel/issues/13223#issuecomment-6935773998
@SEPURI-SAI-KRISHNA Thanks for the update. Here is where each of the three
rows stands from my side.
- **Iceberg (#13268)**: Holding it until #13243 and #13242 are in works for
me. The two measurements you describe cover this row:
- the append case: 1000 keys, 3500 insert, update and delete rows into a
table without a key, with the same 3500 rows at replica 1 and 4;
- the repeated-changes case at replica 4: a whole-table comparison giving
exactly the 50 remaining ids, each at its last value.
What is left: once both prerequisites are in, please make sure the PR
carries only its two `[Iceberg]` commits, then mark it ready and ping here.
- **JDBC XA**: Agreed, it stays `UNSUPPORTED`. Nothing further is needed on
this row.
- **JDBC UNIQUE gap (#13250)**: Your split answers the duplication concern:
#13250 changes only which key index `JdbcSink` hands the writer, and #13242
changes only how that index is hashed and refreshed. I have not checked the
conflict-free merge in either order myself, so I am taking it as you stated it.
What is left: the JDBC E2E with replicas 1, 2 and 4 as its own PR once #13242
is in. Please link it in this thread when it is up.
Nothing else is needed from you on these three rows for now.
<!-- streview-comment:2385 -->
--
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]