LuciferYang opened a new pull request, #10940:
URL: https://github.com/apache/paimon/pull/10940

   ### Purpose
   
   close #10939
   
   A stream scan returns `SnapshotNotExistPlan.INSTANCE` when there is no new 
snapshot, and the Flink enumerators, Spark's `StreamHelper` and 
`CDCSourceEnumerator` check for that instance. With query auth rules, 
`TableQueryAuthResult.convertPlan` wrapped it into a new `DataFilePlan`, so a 
Flink aligned streaming read on such a table failed with a 
`NullPointerException` in the enumerator.
   
   `convertPlan` now returns `SnapshotNotExistPlan` unchanged. It has no 
splits, so there is nothing to wrap, and the auth rules are still fetched and 
applied before planning.
   
   ### Tests
   
   - `RESTCatalogTest#testStreamScanWithAuthRulesKeepsSnapshotNotExistPlan`: a 
stream scan with row filter and masking rules returns 
`SnapshotNotExistPlan.INSTANCE` before the first commit and after it has read 
the commit.
   - `RESTCatalogITCase#testAlignedStreamingReadWithRowFilterOnEmptyTable` and 
`#testAlignedStreamingReadWithRowFilter`: an aligned streaming read with a row 
filter, started on an empty table and on a table that has data, returns the row 
the filter keeps. On master both jobs fail with the `NullPointerException`.
   
   ### API and Format
   
   no
   
   ### Documentation
   
   no
   


-- 
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