SEPURI-SAI-KRISHNA commented on PR #13126:
URL: https://github.com/apache/seatunnel/pull/13126#issuecomment-6983287164

   @DanielLeens Thanks, and for the correction: the order and the loop are 
fixed in `9809b2fabf`. `eaf81ac9fd` merges dev and resolves the conflict in 
`incompatible-changes.md` (EN and ZH), keeping both entries.
   
   **Issue 1.** `testCassandra` now checks the row with NULL cells before the 
comparison. `compareResult` takes the next sink row for every source row and 
fails with `The sink table has no row for source row id N` when the sink has 
fewer rows, and with `The sink table has more rows than the source table` when 
it has more, so a lost row fails the test instead of looping. The no-op 
`assertTrue(true)` is gone.
   
   **Suggestion.** Done. Before the job runs, the test writes sink row -1 with 
`c_text`, `c_int` and `c_list_int` set. After the job every non-key cell of 
that row must be NULL, so the test shows that a NULL is bound and overwrites 
the stored value instead of being left unset.
   
   `CassandraIT#testCassandra` passes on Zeta. With the `TypeConvertUtil` of 
dev it fails promptly: the source throws a `NullPointerException` in 
`buildSeaTunnelRow` on the NULL `blob` cell, the job ends `FAILED`, and the 
test fails on the exit code.
   


-- 
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]

Reply via email to