zhangyue19921010 opened a new pull request, #11312:
URL: https://github.com/apache/paimon/pull/11312

   ### Purpose
   
   Fixes #11311
   
   Creating a table with defaults such as `ARRAY(1, NULL)`, `MAP(1, NULL)` or 
`STRUCT(1, NULL)` fails because the complex cast rules treat uppercase NULL as 
a value to parse. Whole-value NULL defaults also fail for ARRAY and ROW.
   
   Recognize unquoted NULL case-insensitively and add whole-value NULL handling 
to ARRAY and ROW. Quoted and escaped null strings remain literal text. 
Projecting a NULL row default also preserves NULL.
   
   ### Tests
   
   Added Spark SQL create/write/read regressions and cast tests for lowercase, 
uppercase and mixed-case NULL, including quoted and escaped string controls. On 
unchanged production code, all four new Spark regressions fail and the four 
existing controls pass; the focused cast baseline has eight parser errors in 
ten cases.
   
   All 77 tests in `CastExecutorTest`, `DefaultValueRowTest`, `CastedRowTest` 
and Spark `DefaultValueTest` pass with normal Maven checks enabled (JDK 11, 
Spark 3.5.9). Coverage includes projecting a NULL row default.
   


-- 
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