zhangyue19921010 opened a new issue, #11119:
URL: https://github.com/apache/paimon/issues/11119

   ### Search before asking
   
   - [x] I searched the issues and pull requests and found no report of this 
Java schema-evolution failure.
   
   ### Paimon version
   
   Master at `31f3db76793dd709f5d0d17a4b44cafb34174140` (2.2-SNAPSHOT).
   
   ### Compute Engine
   
   Java API, JDK 11.
   
   ### Minimal reproduce step
   
   1. Create an ordinary append table with one nullable `INT` column, `f0`, and 
no primary key.
   2. Write `1`, `2`, `3`, and `NULL`, then commit.
   3. Change `f0` to `BIGINT` and reload the table.
   4. Plan a scan with `new PredicateBuilder(table.rowType()).in(0, 
Arrays.asList(1L, null, 3L))`.
   
   The scan throws a `NullPointerException` from 
`CastExecutors.castLiteralsWithEvolution`, called by 
`SimpleStatsEvolutions.filterUnsafeFilter` during manifest filtering.
   
   ### What doesn't meet your expectations?
   
   The scan should succeed and, with `executeFilter()`, return `1` and `3`. The 
new `SchemaEvolutionTest.testInWithNullAfterIntegerWidening` reproduces the 
exception on unchanged production code.
   
   ### Anything else?
   
   The integer literal conversion path dereferences a NULL while checking for 
overflow. Returning an empty conversion result for that predicate lets the 
existing residual filtering handle it safely.
   
   AI-generated with OpenAI Codex.
   
   ### Are you willing to submit a PR?
   
   - [x] I'm willing to submit a PR!
   


-- 
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